Systems and methods are directed to managing a digital wardrobe using machine learning technology. A wardrobe assistant system receives, via a mobile device of a user, a live video of a plurality of fashion items. Each fashion item of the plurality of fashion items in the live video is identified. The wardrobe assistant system determines details for each fashion item in substantially real-time, whereby the details include a description of each identified fashion item. One of the details can be overlaid over at least one fashion item in the live video. The wardrobe assistant system stores the plurality of fashion items and their corresponding details in a digital wardrobe associated with the user. Subsequently, a style model generates a matched style outfit using two or more of the plurality of fashion items from the digital wardrobe and the matched style outfit is displayed in a user interface.
G06Q 30/0282 - Rating or review of business operators or products
G06Q 50/00 - Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
Automated image generation and optimization is described. In one or more implementations, a sample of listing titles is extracted from listings maintained by the online marketplace within a particular category. The extracted listing titles are simplified using at least one large language model (LLM) to remove extraneous information for image generation. An image prompt is generated using the LLM based on the simplified listing titles and the particular category. One or more images are generated using a large vision model based on the image prompt. The generated images are scored using a vision language model based on predefined criteria. The image prompt is iteratively refined based on the scoring, and images are regenerated until at least one image meets a threshold score for the predefined criteria.
Music generation for item listings for an online marketplace is described. A user (e.g., seller) may list an item for sale on an online marketplace. The item listing may include information about the item, including a title, a description, and one or more images, among other information. The information may be input into a large language model (LLM), which may be trained on a set of music data and a set of inventory categories of the online marketplace. The LLM may execute to generate a music sample that corresponds to the item. The music sample may be applied to the item listing standalone, or as background music for a video corresponding to the item and included in the item listing.
Methods, systems, and computer programs are presented for replacing dialogue in a video segment. One method includes operations for analyzing content of a video to extract dialogue and meaning in the video, and for determining fragments of the video to be modified based on the extracted dialogue and meaning. The modification is based on factors comprising regional differences, cultural sensitivity, and inappropriate content. For each fragment to be modified, the following operations are performed: generate replacement speech based on the regional differences, cultural sensitivity, and inappropriate content; and generate audio for the replacement speech. The generation of the audio comprises synchronizing the audio with the video to align audio with lip movements while maintaining an emotional tone and a voice profile of each speaker in the video. Further, the method includes an operation for causing presentation on a computer display of the video with the modified one or more fragments.
G11B 27/031 - Electronic editing of digitised analogue information signals, e.g. audio or video signals
G06F 40/58 - Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
G06V 20/40 - ScenesScene-specific elements in video content
G06V 40/16 - Human faces, e.g. facial parts, sketches or expressions
G10L 13/027 - Concept to speech synthesisersGeneration of natural phrases from machine-based concepts
G10L 13/033 - Voice editing, e.g. manipulating the voice of the synthesiser
G10L 15/18 - Speech classification or search using natural language modelling
G10L 21/055 - Time compression or expansion for synchronising with other signals, e.g. video signals
G10L 25/57 - Speech or voice analysis techniques not restricted to a single one of groups specially adapted for particular use for comparison or discrimination for processing of video signals
G10L 25/63 - Speech or voice analysis techniques not restricted to a single one of groups specially adapted for particular use for comparison or discrimination for estimating an emotional state
5.
Embedding Large Language Model (LLM) Based, Personalized Trivia Quizzes in a Chat Interface
A large language model (LLM) based personalized chat quiz is described. In one or more implementations, information associated with an online marketplace is clustered into a plurality of topics. A plurality of prompts is generated, each prompt causing the LLM to generate a trivia quiz about a respective topic of the plurality of topics. Each prompt is generated by embedding a natural language description of the respective topic into a predefined prompt template. The prompts are input to the LLM, and generated trivia quizzes are received from the LLM. An indication of a customer service wait time is received. A generated trivia quiz about a personalized topic for the user based on tracked information about the user is selected from the generated trivia quizzes. The generated trivia quiz is embedded into the chat interface for presentation during the customer service wait time.
A first cluster of nodes is deployed in at least one data center, the first cluster of nodes comprising: a leader node in communication with a client device and at least one follower node in communication with the leader node. One or more additional nodes are established and configured as a set of learner nodes associated with the first cluster of nodes. A predetermined number of learner nodes from the set of learner nodes are configured as a predetermined number of follower nodes in a second cluster of nodes. The leader node and the at least one follower node are transitioned from the first cluster of nodes to the second cluster of nodes. At least one learner node from the set of learner nodes is configured as a follower node in the second cluster of nodes.
Systems and methods are directed to minimizing hallucinations in a generated summary. A summary generation system embodied within a server triggers a large language model (LLM) to generate an initial summary for a subject. Based on the initial summary, the server prompts the LLM to generate a list of factual questions about the initial summary. The server then triggers the LLM to answer the list of factual questions without knowledge of the initial summary and using internal knowledge of the LLM. Questions from the list of factual questions that received a positive answer are identified. Based on the questions, the server prompts the LLM to generate a refined summary from the initial summary. The server then generates a user interface that presents the refined summary. Approval of the refined summary triggers generation of a publication using the refined summary.
A system for testing a graphical user interface (GUI) is provided. The system accesses, by a compiler, a page controller associated with testing operations of a GUI of an application, the page controller comprising a list of functions representing respective interactions with the GUI, and the page controller defining one or more conditions associated with executing functions in the list of functions. The system calls a subset of the list of functions of the page controller according to an order defined by a test sequence, a first function in the list of functions being dependent on a result of calling a second function in the list of functions, the compiler preventing the first function from being placed in the order defined by the test sequence before the second function. The system verifies operation of the GUI of the application in response to calling the subset of the list of functions.
Some aspects relate to technologies for providing keyboard emulators on product webpages of mechanical keyboards to emulate the tactile and/our auditory features of the mechanical keyboards. To configure a keyboard emulator for a given mechanical keyboard, a user interface is provided for presentation on a first user device. An input is received via the user interface that adjusts an element of a haptic pattern and/or a sound pattern for the keyboard emulator. Based on the input, a configured haptic pattern and/or configured sound pattern is provided for the virtual keys of the keyboard emulator. Once configured, the keyboard emulator is provided on a product webpage for the mechanical keyboard for presentation on a second user device. User interaction with virtual keys of the keyboard emulator causes the second user device to provide a haptic feedback and/or sound using the configured haptic pattern and/or configured sound pattern.
G06F 3/04886 - Interaction techniques based on graphical user interfaces [GUI] using specific features provided by the input device, e.g. functions controlled by the rotation of a mouse with dual sensing arrangements, or of the nature of the input device, e.g. tap gestures based on pressure sensed by a digitiser using a touch-screen or digitiser, e.g. input of commands through traced gestures by partitioning the display area of the touch-screen or the surface of the digitising tablet into independently controllable areas, e.g. virtual keyboards or menus
G06F 3/01 - Input arrangements or combined input and output arrangements for interaction between user and computer
G06F 3/04842 - Selection of displayed objects or displayed text elements
An input, including information about an item to be published on the publication platform, is received from an interactive interface of a publication platform. The information about the item includes item elements related to the item. A semantic relationship between two or more of the item elements related to the item is identified based on receiving the input. The two or more item elements are grouped into a semantic group based on the semantic relationship. A listing for the item is published on the publication platform based on the two or more item elements in the semantic group.
Controlling access to at least one domain associated with an access point using a wildcard-free certificate and an allow list is described. Relationship data describing how data is to be routed between an access point and domains is received. Based on the relationship data, a certificate is generated that individually lists each domain associated with the access point and includes information describing data routing for the domain via the access point. Data describing at least one exception for granting a request, when the request does not specify a domain included in the wildcard-free certificate, is received and used to generate an allow list. The certificate and the allow list are used to control data communication traffic via the access point.
An object request is received and an object of the object request is matched with sub-objects that are displayed in a first view. A selection of a first sub-object of the sub-objects is received and a three-dimensional rendering of the selected first sub-object is generated in response to receiving the selection. The three-dimensional rendering has second sub-objects associated with the first sub-object where each of the second sub-objects are selectable. The three-dimensional rendering is a second view that is an expansion of the first view where the second sub-objects are shown in the second view.
Methods, systems, and computer storage media for providing a cross-domain secure authentication engine in an item listing system are described. The cross-domain secure authentication engine is an advanced, modular system designed to handle user authentication seamlessly across different service domains, while ensuring security features and an adaptive user experience. In an item listing system, the cross-domain secure authentication engine operates based on a security-focused framework that streamlines user authentication across multiple domains without sacrificing user experience. It integrates a modal containing an iframe-based sign-in interface associated with a separate authentication domain and employs a postMessage API for cross-origin communication. Key features include dynamic modal resizing—adjusting the sign-in window to different authentication steps (like email, password, or OTP)—and parameterized iframes that remove unnecessary UI elements for a cleaner user experience. The cross-domain secure authentication engine also supports security measures and supports incremental, multi-step authentication flows to reduce user friction.
G06Q 20/40 - Authorisation, e.g. identification of payer or payee, verification of customer or shop credentialsReview and approval of payers, e.g. check of credit lines or negative lists
G06F 3/0481 - Interaction techniques based on graphical user interfaces [GUI] based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
Some aspects relate to technologies for performing item retrieval on a listing platform using clusters of interchangeable parts formed using fitment match. In accordance with some aspects, item data is accessed for part item listings on a listing platform, where the item data for each part item listing includes fitment data for each of a number of different fitments. For each part item listing, a fitment hash is generated using fitment data for each fitment of the part item listing. The part item listings are clustered based on overlap of fitment hashes for the part item listings. Cluster data is stored for the part item listing clusters. The cluster data for each part item listing cluster associates a cluster identifier and an item listing identifier for each part item listing in the part item listing cluster. The cluster data can be leveraged to perform item retrieval for the listing platform.
Some aspects relate to technologies for providing personalized parts experiences to users on listing platforms. In accordance with some aspects, signal data is accessed for a user of a listing platform based on user interactions of the user on the listing platform that are associated with parts and/or applications. Signal classifications are determined for the signal data to provide classified signal data for the user. One or more user classifications for the user are then determined based on the classified signal data. At least one of those user classifications is used to select a personalized parts experience for the user. One or more application servers of the listing platform then provide the personalized parts experience to a user device of the user via one or more user interfaces presented on the user device.
Some aspects relate to technologies for generating persona simulators and employing the persona simulators for interaction simulation. The persona simulators can comprise user persona simulators that act as end users during simulated interactions and agent persona simulators that act as agent users during simulated interactions. To generate a persona simulator, interaction data for an end user or an agent user is accessed. The interaction data can be based on, for instance, transcript(s) from interaction(s) involving the end user or agent user. Parameter scores for different interaction parameters are generated from the interaction data, and in some cases, the parameter scores are used to determine an index score. The scores define the persona of the end user or agent user and are used to provide an input to a generative model to provide a persona simulator. The persona simulator is then used to generate text for messages during a simulated interaction.
G06Q 30/015 - Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk
G06N 3/006 - Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
Partner-based item recommendations are described. A computing system receives historical data associated with user engagement and transactions for a set of items. The computing system selects a set of candidate items from the set of items based on a set of categories associated with the set of items, a set of popularity metrics associated with the set of items, and a set of similarity metrics associated with the set of candidate items and the set of items. The computing system generates, for each candidate item, a respective user preference score based on the historical data and using a neural network including an attention mechanism and a weighted loss function. The computing system orders the set of candidate items according to the user preference scores and a category distribution for the set of items. The computing system outputs at least one candidate item based on the ordered set of candidate items.
Dynamic listing modification is described. A listing strategy system receives information describing a digital marketplace listing and generates a recommended strategy for the listing based on characteristics of similar listings. Based on the monitored interactions with the listing, the listing strategy system detects when the recommended strategy deviates by a threshold amount from an optimal strategy. In response to detecting a threshold deviation from the optimal strategy, the listing strategy system automatically modifies at least one characteristic of the listing based on the optimal strategy without changing a net renumeration for the listing. A listing modification summary system uses information describing the listing, the net renumeration, original listing characteristics, the recommended strategy, the optimal strategy, and modified characteristics to create a prompt. The prompt causes a machine learning system to generate a modification summary that details how observed interactions with the listing correlate to listing characteristics over time.
Methods, systems, and computer storage media for providing an advertising campaign management engine in an artificial intelligence system are described. Advertising campaign management refers to the process of planning, executing, monitoring, and optimizing advertising campaigns. Operationally, first, high-conversion design modeling and mining analyzes performance data and user preferences to identify design elements—such as image style and text layout—that drive engagement. Second, design requirement generation converts ad optimization profiles into structured prompts that guide the refinement of ad designs. Lastly, design fine-tuning leverages advanced generative models to adjust ad elements, such as images and layouts, based on the generated requirements. The advertising campaign management engine supports an iterative process that enables automated updates and optimizations, allowing ads to be tailored to specific demographics. For example, it could produce quirky designs for young adults or simplified layouts for seniors. Refined ads are then deployed and continuously monitored for performance improvements.
An immersive trading card experience is described. In one or more implementations, an image of a front of a trading card is obtained and segmented into two or more layers displayable at different depths. Listing information corresponding to the trading card is retrieved for use as the trading card rear and is adapted to include real-time trending information for additional listings of the trading card. In response to the trading card being returned in a search result, an enhanced item card is communicated to a computing device. The enhanced item card includes the trading card front with the layers displayable at different depths. A rotatable display of the enhanced item card includes the layers displayable at different depths as an enhanced image of the trading card front and includes the listing information as the trading card rear.
G06F 3/04845 - Interaction techniques based on graphical user interfaces [GUI] for the control of specific functions or operations, e.g. selecting or manipulating an object, an image or a displayed text element, setting a parameter value or selecting a range for image manipulation, e.g. dragging, rotation, expansion or change of colour
Key phrase recommendation based on token correspondence is described. In one or more implementations, a plurality of user queries entered via a search feature of a listing platform is received. Key phrases are extracted from the plurality of user queries. The key phrases are associated with an item category of the listing platform, and each have a search count that exceeds a threshold. Further, an item title of an item listed via the listing platform in the item category is obtained, and the item title includes title tokens. The key phrases are ranked based, in part, on quantities of the title tokens included in the key phrases, and the ranked key phrases are output.
Some aspects relate to technologies for a keyphrase relevance model trained to filter keyphrases when providing relevant keyphrase recommendations for promoted placement of item listings. In accordance with some aspects, a training dataset is accessed. The training dataset comprises training samples that each includes: a keyphrase, item listing data for an item listing on a listing platform, and a label indicating relevance of the keyphrase to the item listing. The relevance labels are obtained using a search relevance model of a search system for the listing platform. Positive training samples have a relevant label, while negative training samples have a non-relevant label. A keyphrase relevance model is trained using the positive training samples and the negative training samples to provide a trained keyphrase relevance model. The trained keyphrase relevance model is used to filter candidate keyphrases in order to provide keyphrase recommendations for item listings on the listing platform.
A machine learning model that augments a listing is provided. The model receives an item associated with a first listing and extracts a first vector representation comprising a first characteristic from the image content. The model searches a corpus for a second vector representation having a second characteristic using the first vector representation. The corpus has second vector representations associated with the second items. The model compares the first characteristic with the second characteristic and determines a match between the first vector representation and the second vector representation. The second characteristic is associated with a second item that is associated with a second listing. The model augments the first listing based on the second listing and displays the augmented first listing.
A system for automatically generating auto-scaling configurations for graphics processing unit (GPU) models is described. A provides users a platform to perform a load and performance (LnP) test on a model of a GPU of a computing device. The LnP test may result in a set of metrics associated with the GPU model. For example, the metrics may include throughput metrics (e.g., transactions per second (TPS)) and GPU utilization metrics. In some examples, the metrics may be based on a service level agreement (SLA), which may include latency, error rate, and GPU utilization requirements. Based on a scaling threshold determined from the metrics and a utilization requirement, the system may output an auto-scaling configuration for the GPU model. The GPU may operate using the auto-scaling configuration, where the auto-scaling configuration may enable the GPU model to scale up or scale down, for example, based on changes in traffic.
An artificial intelligence (AI)-based system for component compound identification is described. The system may receive images associated with a product listing for an online marketplace. The product listing may be for a medicinal substance (e.g., a prescription drug). The system may generate a prompt for a large language model (LLM) based on the images and prompt the LLM to identify component compounds (e.g., ingredients) of the medicinal substance. For example, the LLM may recognize text from the images that indicates the component compounds and their respective amounts (e.g., weights or percentages). The system may generate an output based on prompting the LLM, the output indicating the identified component compounds and a determination of whether the product listing is approved or denied.
In an example embodiment, a transaction log index is used to index log records. The transaction log index comprises an in-memory transaction log index and an on-disk transaction log index. The log records are temporarily indexed in the in-memory transaction log index until the storage server can ensure that the corresponding log records are stored without any missing log records in a sequence, as well as possibly based on other correctness and performance considerations.
Methods, systems, and computer storage media for providing a network management engine in a cloud computing system are described. Network management engine is an end-to-end system that oversees, monitors, and optimizes the full lifecycle of network traffic, from the initial client request to the final data response. By capturing, analyzing, and managing network packets, it ensures seamless data flow and tracks key performance metrics to identify and resolve latency or congestion issues. Network management engine supports monitoring, and fault detection for efficient data flow, capturing metrics, and analyzing performance. Network management engine includes network packet management extension engine and network performance management engine. Network packet management extension engine is a specialized engine designed for calculating and analyzing packet latency at various stages of a network path. Network performance management engine is a specialized engine that identifies and manages deviations from expected network behavior, particularly in areas where latency may occur.
A search query for an item having at least one object is received. An item identification (id) for the item is then determined using the at least one object. A first database having item identification (ids) associated with object types is accessed. Using the item id, a determination is made whether the first database has an object type that corresponds to the item id. A second database that has categories corresponding to object types is accessed to determine a category that corresponds to the object type. A third database having project names that correspond to the object type and the category is accessed to determine a project name associated with the item. A list of recommended items based on the project name is generated and presented to a user.
Systems and methods are directed to authenticating items using deformation analysis. An authentication system receives a request to authenticate an item. In response, the authentication system accesses two images of the item, whereby a first image shows the item hanging without a weight load and a second image shows the item hanging with the weight load. An image analyzer analyzes the two images to determine deformation differences of the item between the two images. The authentication system then determines an authentication score for the item by applying the deformation differences to a machine learning model trained with training data derived from authentic and counterfeit items. Based on the authentication score transgressing an authenticity threshold, an indication of authenticity of the item is provided.
A transaction system selects an individual item from an inventory associated with an online seller in an online e-commerce platform. The transaction system obtains, from a database of the online e-commerce platform, information associated with the individual item. The transaction system processes the information associated with the individual item by one or more large language models (LLMs) and one or more machine learning models to identify a target item that is associated with the individual item and generates a recommendation for presentation to the online seller to expand the inventory associated with the online seller to include the target item.
A message from a first device of a first user for sending to a second device of a second user is received. Context for the message is generated based on a chat history between the first user and the second user. The chat history comprises one or more messages exchanged between the first user and the second user prior to receiving the message. The message is classified as attempted extortion by a machine learning model trained to classify the intention of messages based on content of the message and the context for the message. Based on classifying the message as attempted extortion, the message is prevented from being sent to the second device of the second user.
Methods, systems, and computer storage media for providing a network management engine in a cloud computing system are described. Network management engine is an end-to-end system that oversees, monitors, and optimizes the full lifecycle of network traffic, from the initial client request to the final data response. By capturing, analyzing, and managing network packets, it ensures seamless data flow and tracks key performance metrics to identify and resolve latency or congestion issues. Network management engine supports monitoring, and fault detection for efficient data flow, capturing metrics, and analyzing performance. Network management engine includes network packet management extension engine and network performance management engine. Network packet management extension engine is a specialized engine designed for calculating and analyzing packet latency at various stages of a network path. Network performance management engine is a specialized engine that identifies and manages deviations from expected network behavior, particularly in areas where latency may occur.
Alternative options are presented to users at the point of personalization of a search engine to generate personalized query suggestions. Seed items within a module carousel comprising item cards on a homepage are identified. Embeddings are retrieved for at least a portion of the seed items. Based on the embeddings, related item embeddings corresponding to related items that were purchased by other users are identified. A list of queries for each of the related items that were purchased by the other users is retrieved. A ranking for each query of the list of queries is generated. Based on the ranking, queries of the list of queries are presented in place of at least one of the item cards within the module carousel. In various aspects, the queries of the list of queries broaden the intent, narrow the intent, or diversify the intent.
In the implementation of techniques for generating guidance digital content with artificial intelligence, a system receives digital content and metadata corresponding to item listings for an item category. The system filters the digital content based on predefined quality metrics and metadata to identify a subset of reference digital content, where each reference digital content meets or exceeds a predefined quality threshold. Via one or more artificial intelligence models, the system classifies the subset of reference digital content into distinct perspectives, each distinct perspective representing a different view of items included in the item category. Based on this classification, via the one or more artificial intelligence models, the system generates guidance digital content for each distinct perspective. The system broadcasts the guidance digital content for display via a user interface in association with the item category.
A user query is received. Based on the user query, one or more data files are identified. Each data file is associated with metadata relevant to the user query. One or more data chunks are determined from the one or more data files using a first machine learning model. A response to the user query is generated using a second machine learning model. The response is a synthesized response that is derived from the content of the one or more data chunks.
In implementations of systems and procedures for item recommendation and visualization, a computing device receives an input digital image depicting an environment and identifies attributes of objects within the environment. The attributes of the objects are used to identify items from item catalog data that have similar attributes and are suitable for inclusion within the environment. The identified items are displayed by way of a recommendation. The attributes of the objects are further used for generation of synthesized images. The synthesized images depict objects from the input digital image virtually replaced by items from the item catalog data. The synthesized images thus support visualization of the items within the environment depicted by the input digital image.
Detecting imbalances in data traffic for distributed data centers is described. A distributed data center system can obtain a set of data samples corresponding to respective data traffic at data centers, the respective data traffic based on data associated with an application. The distributed data center system can generate subsets of weight-transformed data samples associated with respective intervals of time. The distributed data center system can generate respective indices associated with the subsets of weight-transformed data samples. The indices can be representative of a data traffic balance at the data centers. The distributed data center system can transmit a message indicating an index failing to satisfy a threshold value, the index being associated with at least one subset of weight-transformed data samples.
A plurality of query result lists generated within a predetermined time period is identified based on a user-generated search query. One or more bottom-listed query results from each query result list are determined. A plurality of scores is generated using a first machine learning model. One or more lowest-scored query results are determined based on the plurality of scores. A second machine learning model is trained using a contrastive learning framework based on one or more hard-negative query results and the user-generated search query.
A user-generated search query is received. A first machine learning model is used to retrieve a plurality of embeddings representing a plurality of query rewrite candidates. A plurality of query pairs is generated between the user-generated search query and each query rewrite candidate. A second machine learning model is used to generate a plurality of similarity scores based on the plurality of query pairs. A plurality of similarity scores is ranked to identify a query rewrite candidate used for generating query results. The query results are caused to be displayed.
Methods, systems, and computer storage media for providing an adaptive subtitle color management engine in an item listing system. The adaptive subtitle color management engine supports automatically updating a subtitle color of a video to a new subtitle color (e.g., contrast color) that is legible against the changing colors and brightness of levels of video-background. Subtitle data associated with video data can include a subtitle segment time stamp including a subtitle start time and a subtitle end time. Based on the subtitle file data and the subtitle segment time stamp, frames of video are identified for computing a contrast color for subtitles for the frames. The evaluation to change the subtitle color is repeated for frames associated with subtitles as defined in the subtitle file data. The contrast color is determined based on the average background color of the frame, where the average background color indicates an average pixel intensity.
A user selects an optical system on a publication platform. The optical system includes an optical device and an image formation device. A first parameter of the optical device and a second parameter of the image formation device are obtained and a field of view (FOV) of the optical system is determined based on the first parameter and the second parameter. A user selects an observation target, the observation target corresponding to a position in a predetermined image. A preview image of the selected observation target is generated and displayed on the publication platform based on the FOV of the optical system and the position of the selected observation target in the predetermined image, the preview image of the selected observation target being an estimated image of the observation target from the optical system.
A method for extracting column-level data lineage from Structured Query Language (SQL) scripts using a novel Bi-Tree model for efficient and scalable parsing of complex SQL statements is described. In one example embodiment, a computer-implemented method includes parsing a Structured Query Language (SQL) statement to generate an Abstract Syntax Tree (AST), constructing a bi-tree model based on the AST, the bi-tree model includes a context tree and a select-pattern (SP) tree, and extracting column-level lineages based on the bi-tree model.
Listwise autocomplete ranking with position debias is described. A computing system can receive a user input indicative of a portion of a search query. The computing system can receive a set of candidate queries for an autocomplete list based on the portion of the search query. The computing system can generate a respective ranking score for each candidate query of the set of candidate queries by calculating, for each candidate query of the set of candidate queries, a respective click probability prediction using a position debiased listwise click-through rate (CTR) prediction model, and calculating, for each candidate query of the set of candidate queries, a respective reward per search. The computing system can order the set of candidate queries according to the respective ranking scores. The computing system can output at least one of the candidate queries based on the ordered set of candidate queries.
A method for a publication platform involves identifying users and their corresponding metrics, classifying users based on their likelihood to respond to incentives, identifying a targeted user group, identifying items published by a user, selecting a set of items based on management rules, generating a pluggable interface component providing a consistent recommendation interface across different device operating systems, receiving attribute values of a custom incentive signal from the user, and sending a message based on the custom incentive signal to the targeted user group.
A counterfeit item detection system detects counterfeit items during an item listing processes provided by an online marketplace. The system enhances the ability of the online marketplace to identify and reject potential counterfeit items. The system comprises a trained counterfeit item detection model that is configured to receive an image and identify whether the image includes a counterfeit item. The model is trained using a data set of training images. An image of the data set is taken from a video related to the time based on identifying that the context of text associated with the video relates to counterfeit items. The text can be determined from the video's audio, and the image is obtained at a time in the video where the text corresponds to a counterfeit item context.
Systems, methods, and media are provided for separately searching a plurality of image segments in an image and displaying search results for each image segment on the same search results page. An input image may be uploaded to a search engine. An image segmentation technique and object recognition model are used to identify objects and extract image segments having the objects in the segmented image. A database search is performed for each image segment to identify search results for each image segment. The search results are generated for display on a search results page in groups. In some aspects, a first search results page may display the top ranked search results for each image segment in a group, and a second search results page may include the second ranked search results for each segment in a group.
G06F 16/583 - Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
One or more images of an item on a publication platform or to be published on a publication platform are accessed. The text content from the one or more images of the item are extracted. A description of the item is determined based on the extracted text content. The description of the item is displayed on an interactive interface on a user device. A response to the displayed description (e.g., a modification of the displayed description by a user of the user device) is received. A modified description based on the response to the displayed description is published on the publication platform.
Methods, systems, and computer storage media for providing an interest-driven recommendation engine in an item listing system are described. The interest-driven recommendation engine operates through three key stages: user interest modeling, item exploration, and recommendation delivery. In user interest modeling, user activity logs are used to build a hierarchical interest graph, capturing both short-term and long-term interests. A teacher-student Large Language Model “LLM” paradigm generates and scales these graphs efficiently, refining them to prioritize passion-driven interests over utilitarian needs. During item exploration, ranked interests are transformed into query expansions through a process called interest-to-recall. These queries are converted into embeddings, enabling the interest-driven recommendation engine to retrieve relevant items from the inventory. In recommendation delivery, a learning-to-rank neural network scores retrieved items for relevance, followed by a diversification step to balance closely aligned recommendations with inspirational suggestions. This ensures recommendations are both practical and engaging, fostering discovery and user satisfaction.
09 - Scientific and electric apparatus and instruments
35 - Advertising and business services
41 - Education, entertainment, sporting and cultural services
42 - Scientific, technological and industrial services, research and design
Goods & Services
Downloadable computer software for livestream shopping, livestream auctions, interactive retail transactions, product demonstrations, and consumer engagement; Downloadable software for transmitting live video, audio, chat, and multimedia content for e-commerce purposes; Downloadable software for streaming live human hosts for use in online shopping, auctions, and product marketing; Downloadable and recorded computer database software featuring information in the field of hobbies, collectibles and a wide variety of consumer products of others; Downloadable and recorded computer software for processing electronic credit card, debit card, cash card, gift card, wire transfer, and wallet payments and for transferring funds to and from others Providing an online marketplace for buyers and sellers of goods and services via livestream video; Online trading services in which sellers post products and services to be auctioned or sold and bidding and purchasing is done via livestream shopping and auction events; Product demonstration services, namely, live product demonstrations of trading cards, coins, comics, toys, games, clothing and accessories, handbags, watches, jewelry, sneakers, health and beauty products, electronics, and sporting goods; Providing business information to buyers about sellers and their goods and services for the purpose of facilitating transactions between buyers and sellers; Providing a searchable online advertising guide featuring the goods and services of online vendors; Providing a searchable online evaluation database featuring consumer product and services information in the nature of buyer feedback, seller ratings and reviews, and product evaluations and reviews for buyers and sellers; Advertising services; Providing consumer product information relating to confirming authenticity of consumer products, product manufacturers, and sellers of goods for the purposes of helping consumers make informed purchasing decisions Entertainment services, namely, providing videocasts in the field of interactive shopping and auctions featuring live human hosts; Providing online non-downloadable videos via livestream in the field of shopping, auctions, consumer products, and product demonstrations Software as a service (SAAS) services featuring software for livestream shopping, livestream auctions, interactive retail transactions, product demonstrations, and consumer engagement; Providing online, non-downloadable software for transmitting live video, audio, chat, and multimedia content for e-commerce purposes; Providing online, non-downloadable software for streaming live human hosts for use in online shopping, auctions, and product marketing; Maintenance and updating of computer software for others; Providing online, non-downloadable computer e-commerce software to allow users to conduct electronic business transactions in online marketplaces via a global computer network; Providing online, non-downloadable computer database software featuring information in the field of hobbies, collectibles and a wide variety of consumer products; Providing online, non-downloadable computer software and non-downloadable software development tools for use in developing further software and software applications in the field of e-commerce; Providing online, non-downloadable computer software for processing electronic credit card, debit card, cash card, gift card, wire transfer, and wallet payments and for transferring funds to and from others
Some aspects of the present technology relate to technologies for performing client testing using network footprint comparison. In accordance with some configurations, a network proxy initiates a client-side network testing process associated with a network and an application where the network proxy intercepts outbound network requests from the application via the network. A network signature of the application is then generated using the outbound network requests and this network signature is compared to a baseline network signature. In accordance with some configurations, comparing the network signature comprises identifying variances in application-requested-inputs associated with the one or more outbound network requests. Based on this comparison, a client-side network test result is generated that can be used to identify variances in the application relative to the baseline network signature.
H04L 41/0604 - Management of faults, events, alarms or notifications using filtering, e.g. reduction of information by using priority, element types, position or time
Systems and methods are directed decomposing an image using artificial intelligence (AI) and large language model (LLM) technology. The system accesses an image containing one or more objects and processes the image through an image captioning model to generate an image caption for the image. The system then creates an enhanced prompt by integrating the image caption with user inputs that describe or customize the object(s) in the image into a general prompt for a category associated with the image. The enhanced prompt triggers a text-based LLM to decompose the image into individual components and corresponding details. The system then causes presentation of a user interface that includes results from the text-based LLM, whereby the user interface include fields for each individual component.
Some aspects relate to technologies for optimizing tracking metadata for server-rendered Hypertext Markup Language (HTML) to reduce the size of cached HTML. In accordance with some aspects, a server (e.g., an origin server) generates pre-rendered HTML with markers in place of tracking metadata. The server also generates a mapping between the markers and the tracking metadata. The pre-rendered HTML and mapping are cached, for instance, on a local cache of that server and/or one or more other servers (e.g., edge servers). When a server receives a request for content, the server retrieves the pre-rendered HTML and mapping from its local cache and provides the pre-rendered HTML and mapping to the requesting client. When the client renders the content, the interactive elements in the pre-rendered HTML are populated by replacing the markers with tracking metadata using the mapping.
A generative model evaluator is built to assess the performance of generative models. An example method involves providing a set of guidelines for evaluating generative model outputs, a training data set with inputs and initially scored outputs, and an evaluation prompt. Using an evaluation model, a model-determined evaluation score is generated for the outputs. The optimization engine identifies differences between the initial evaluation scores and model-determined evaluation scores, and determines whether a difference is from an error in the initial evaluation score, the model-determined evaluation score, or the guidelines. Based on the determined error, a modification is made to the initial evaluation score, the set of guidelines, or the evaluation prompt. The process is iteratively continued using the modifications to optimize the evaluator, which can include the optimized evaluation prompt and the optimized set of guidelines.
A system is provided that adjusts a setting using user motion. The system displays a setting on a screen having a value adjustable based on user input and receives, from a head-worn device, first data that indicates activation of the setting to enable the setting to be adjusted. The system receives, from the head-worn device, motion information associated with a body part of a user of the head-worn device and adjusts the value of the setting based on the motion information. The system receives, from the head-worn device, second data indicative of a verbal cue from the user to confirm the adjusted value and detects a pattern of movement associated with the body part of the user concurrently with receiving the second data. The system causes the adjusted value to be stored in association with the setting based on receiving the second data and detecting the pattern of movement.
Methods, systems, and computer storage media for providing an LLM-based labeler development engine in a customer service management system are described. The LLM-based labeler development engine leverages ground-truth data (e.g., human-labeled data), example selection machine learning model, a few-shot learning LLM, and iterative labeling refinement operations to develop an LLM-based labeler. The labeler development engine uses the ground-truth data, the few-shot learning LLM and iterative refinement operations—including error analysis and automatic prompt correction to train the LLM-based labeler. The LLM-based labeler can be used to label a training dataset for training a machine learning model for the customer service email management system, the machine learning model supports categorizing incoming customer service emails. In this way, the LLM-based labeler operates as an intermediary labeler that is developed with a subset of data to facilitate labeling a training dataset for training the machine learning model of a customer service email management system.
Systems and methods are directed to optimizing large language model (LLM) query responses using graph Retrieval-Augmented Generation (RAG). A Graph RAG system generates a knowledge graph customized for a team based on data from one or more data sources maintained by the team. The knowledge graph is then stored for later use by a query system. The query system receives, from a client device, a query that requires context from the knowledge graph. The context is obtained from the knowledge graph on substantially real-time. The query system generates a prompt that includes the context and the query. The prompt triggers the LLM to provide a response to the query. The query system then causes display of the response on the client device.
Dynamic automatic generation of item listings is described. A computing device (e.g., or a user engagement system) receives input that indicates at least one item. The computing device generates a probability of user engagement with a listing of the at least one item. The computing device generates the probability based on a context associated with the at least one item. In some cases, the computing device displays a control selectable to automatically generate the listing of the at least one item based on the probability of the user engagement satisfying a threshold value. In some other cases, the computing device displays a control selectable to generate a configurable template for the listing of the at least one item based on the probability of the user engagement failing to satisfy a threshold value.
A system is provided that automatically modifies code for an application. The system accesses a database comprising code for an application and generates a prompt comprising instructions for identifying target conditional statements in the code for the application and modifying the code for the application to remove the target conditional statements. The system stores the code for the application in temporary storage. The system processes, by a generative machine learning model, the prompt and the code for the application in the temporary storage to automatically generate a set of new code segments for portions of the code for the application associated with the target conditional statements.
09 - Scientific and electric apparatus and instruments
16 - Paper, cardboard and goods made from these materials
35 - Advertising and business services
36 - Financial, insurance and real estate services
38 - Telecommunications services
39 - Transport, packaging, storage and travel services
41 - Education, entertainment, sporting and cultural services
42 - Scientific, technological and industrial services, research and design
45 - Legal and security services; personal services for individuals.
Goods & Services
Recorded content; databases (electronic); digital or
electronic directories; media content; electronic
publications recorded on computer media; instruction manuals
and training manuals in electronic format; downloadable
image files; electronic publications (downloadable); audio
visual recordings; computer software; programs for
computers; software downloadable from the internet;
interfaces for computers; single sign-on application
software; computer chatbot software for simulating
conversations, in particular based on artificial
intelligence; interactive software based on artificial
intelligence; virtual assistant software; computer programs
for use in telecommunications; software for document
management, for data processing and for database management;
software to facilitate electronic commerce; point-of-sale
software; software for accessing information directories
that may be downloaded from the global computer network;
software for interpretation of market information; software
to enable the searching of data; search engine software;
software relating to financial transactions, monetary
matters and finance management; software for creating
searchable databases of information and data; software for
enterprise resource planning, inventory management and
inventory monitoring; software for managing customer data;
content management system software; software for managing
dates and appointments; document management system software;
software for facilitating secure credit card transactions;
downloadable cryptographic keys for receiving and spending
crypto assets; software for operating online shops; software
for retrieving web feeds; electronic blackboards;
applications (software) for portable communications devices;
downloadable web-based applications for portable
communications devices or computers; computer software and
applications for portable communications devices and
computers for creating, editing, publishing, viewing,
searching and saving classified advertisements, in
particular for an online vehicle market for the purchase and
sale of new and used vehicles, vehicle trailers,
motorcycles, bicycles, e-bikes, building and agricultural
machines, containers, boats and other water vehicles, and
parts and fittings for vehicles, motorcycles, bicycles,
e-bikes, building and agricultural machines, boats and water
vehicles; computer software and applications for portable
communications devices and computers for creating, editing,
publishing, viewing, searching and arranging of automotive
vehicle services and of repair services for vehicles,
vehicle trailers, motorcycles, bicycles, e-bikes, building
and agricultural machines, containers, boats and other water
vehicles; software for comparing of insurances; software for
comparing of funds, in particular loans; software for
arranging of insurance and finance; virtual and augmented
reality software; artificial intelligence software; machine
learning software; e-commerce and e-payment software;
software for pay-per click-optimization; software for
automated advertising placement in the form of adverts for
customer products; software for automated retrieving of
third party addresses; downloadable software for language
translation; downloadable computer programs and downloadable
computer software for the artificial generation of human
speech and texts; downloadable software that uses artificial
intelligence processing to provide a virtual assistant with
generative artificial intelligence, generative pre-trained
transformers, and large language models for collecting,
analyzing, processing, providing, and summarizing data and
business, commercial, or financial information;
communications equipment; point-to-point communications
equipment; podcasts; pre-recorded compact discs, DVDs, HD
DVDs, UHD discs and other digital data storage media; USB
flash drives; electronic agendas; mouse pads; file and
electronic mail servers; car radios; electric cables and
wires, in particular starter cables for motors and jump
leads; warning triangles; navigation apparatus for vehicles;
vehicle drive training simulators; parts and fittings of all
the aforesaid goods, included in the class. Stationery and educational supplies; office requisites,
except furniture; printed matter; transfers [decalcomanias];
tear-off calendars; advent calendars; stickers [stationery],
in particular vehicle bumper stickers; newspapers;
periodicals; photographs [printed]; books; pamphlets; guide
books; forms, printed; paper and cardboard; parts and
fittings of all the aforesaid goods, included in the class. Advertising, marketing and promotional services; updating of
advertising information in a computer data base; banner
advertising; arranging and placing of advertisements (for
others); advertising services provided via the internet;
classified advertising, also through portals and platforms
on the internet, in particular for the purchase and sale of
new and used vehicles, vehicle trailers, motorcycles,
bicycles, e-bikes, building and agricultural machines,
containers, boats and other water vehicles and parts and
fittings for vehicles, motorcycles, bicycles, e-bikes,
building and agricultural machines, boats and water
vehicles; classified advertising, also through portals and
platforms on the internet, in particular for automotive
vehicle services and for repair services for vehicles,
vehicle trailers, motorcycles, bicycles, e-bikes, building
and agricultural machines, containers, boats and other water
vehicles; search engine marketing services; web site traffic
optimisation; search engine optimisation; pay per click
advertising; advertisements (placing of-); collecting and
compilation of business directories of goods, services,
addresses and industries and the information contained
therein; advertising of the sale and purchase of vehicles,
vehicle trailers, motorcycles, bicycles, e-bikes, building
and agricultural machines, containers, boats and other water
vehicles and parts and fittings for vehicles, motorcycles,
bicycles, e-bikes, building and agricultural machines, boats
and water vehicles; advertising for automotive vehicle
services and for repair services for vehicles, vehicle
trailers, motorcycles, bicycles, e-bikes, building and
agricultural machines, containers, boats and other water
vehicles; public relations services; product demonstration
and product display services; presentation of services for
sales purposes [for others]; presentations of companies as
well as goods and services on the internet and other analog
and digital media [for others]; conducting, implementation
and organizing of exhibitions, fairs, events and shows for
commercial, promotional and advertising purposes; loyalty,
incentive and bonus program services; provision of
advertising space, time and media, in particular for
classified advertising on global computer and wireless
networks; provision of computerised advertising services;
provision of space in web-sites for advertising goods and
services; direct mail advertising; dissemination of
advertisements and commercial announcements, in particular
for others via the internet and portable communications
devices; advertising, marketing and promotional consultancy,
advisory and assistance services; press advertising
consultancy; marketing assistance; commercial trading and
consumer information services, namely mediation of trade
business for third parties, mediation of agreements
regarding the sale and purchase of goods, bidding quotation,
price comparing services; consumer consultancy in relation
to the selection of goods and services, in particular using
test and researching results and ratings by third parties;
arranging subscription services for others, in particular
subscriptions of vehicles, vehicle trailers, motorcycles,
bicycles, e-bikes, building and agricultural machines,
containers, boats and other water vehicles; auctioneering
services, namely on the internet; provision of commercial
and business contact information; providing user reviews for
commercial or advertising purposes; competition arrangements
for advertising and commercial purposes; commercial trading
and consumer information services, namely commercial
information agency services; import and export agency
services; providing advice to consumers and businesses
relating the conclusion and sale of subscriptions of
vehicles, vehicle trailers, motorcycles, bicycles, e-bikes,
building and agricultural machines, containers, boats and
other water vehicles; online and mail order retail services
relating to new and used land and water vehicles, vehicle
trailers, motorcycles, bicycles, e-bikes, and vehicle,
motorcycle, bicycle and e-bike parts; retail and wholesale
services relating to new and used vehicles, vehicle
trailers, motorcycles, bicycles, e-bikes, building and
agricultural machines, containers, boats and other water
vehicles and parts and fittings for vehicles, motorcycles,
bicycles, e-bikes, boats and water vehicles; administrative
processing and organising of mail order services;
computerized administrative processing and organising of
mail order services; business assistance, management and
administrative services; market reporting services;
accountancy, book-keeping and auditing; computerised
auditing; preparation of accounts; business accounts
management; administrative data processing; computerized
file management; computerised business information
processing services; information services relating to data
processing; automated data processing; consultancy relating
to data processing; on-line data processing services;
systemization of information into computer databases;
processing of business survey results; industrial management
consultation including cost/yield analysis; business
advisory and information services; preparation of documents
relating to business; company record keeping [for others];
operation of businesses [for others]; administration of
business affairs; business consultancy and advisory
services; advice relating to business acquisition; business
management consulting in relation to strategy, marketing,
and public relations; establishing a network of business
contacts (service to assist in -); arranging of business
introductions; arranging of contracts, for others, in
particular for the providing of services; arranging of
contracts, for others, for the purchase and use of wallboxes
or solar panels; negotiation and conclusion of commercial
transactions for third parties; arranging of energy
contracts relating to the charging of electric vehicles,
electric motorcycles, electric boats and e-bikes;
professional business consulting; invoicing; business
analysis, research and information services; market
information services relating to market statistics and sale
of market statistics; provision of on-line business and
commercial information; business information relating to
commercial or industrial enterprises; business expertise
services; provision of statistical information relating to
business; analysis of business statistics; benchmarking
services for business management purposes; conducting of
business appraisals; providing business information via
website; business surveys; provision of computerised
business statistics; provision of sales analyses; cost price
analysis; creation of visitor statistics, in particular for
advertisements on the internet; marketing research; price
analysis services; collection and systematization of
business data; compilation and systematization of
information into computer databases; automated data
processing, systematization and management using artificial
intelligence; data compilation and data management for
others; compilation of statistics; compilation of
directories for publishing on global computer networks on
the internet; business information services provided online
from a computer database or the internet; business analyses
services; rental, hire and leasing of objects in connection
with the provision of the aforesaid services, included in
this class; advice, consultancy and information for the
aforesaid, included in the class; provision of an online
searchable and supplementable marketplace for buyers and
sellers for the search for and the placement of classified
advertisements for the purchase and sale of vehicles,
vehicle trailers, motorcycles, bicycles, e-bikes, building
and agricultural machines, containers, boats and other water
vehicles and parts and fittings for vehicles, motorcycles,
bicycles, e-bikes, building and agricultural machines, boats
and water vehicles, and financing and services offers;
comparison of finance services in particular credits,
hire-purchase contracts, trade-ins and other finance and
funding services; computerised comparison of insurances,
insurance coverages and insurance premiums. Insurance services; consulting and information concerning
insurance; brokerage advisory services relating to
insurance; insurance premium rate computing; computerised
information services relating to insurance; provision of
insurance premium quotations; insurance brokerage and
financial advisory services relating to insurance contracts,
in particular via the internet or portable communications
devices; insurance brokerage; warranty services; guarantee
assurance underwriting; provision of extended warranties;
provision of warranties relating to goods and services, that
can be mediated via a platform on the internet; rental of
office and commercial space, in particular for the use as
showrooms; financial and monetary services, and banking;
banking and financing services; trusteeship representatives;
arranging of financial trade-ins and financial clearing
transactions; trusteeship of money; financial trust
management; maintaining escrow accounts relating to
processing legal transactions; payment and receipt of money
as agents; financial brokerage services; arranging of
finance, credit and finance services; loan and credit, and
lease-finance services; lease-purchase financing; financing
and lease-purchase financing relating to vehicles; debt
recovery and factoring services; financial transfers and
transactions, and payment services; cash, check (cheque) and
money order services; automated payment services; electronic
payment services; payment processing; remote payment
services; accounts payable debiting services; debt recovery;
collection of money owed from settlements; financial
information, data, advice and consultancy services;
computerised financial advisory services; financial rating
and credit reports; financial valuation services; valuations
and financial appraisal services; financial evaluation of
vehicles, vehicle trailers, motorcycles, bicycles, e-bikes,
building and agricultural machines, containers, boats and
other water vehicles; rental, hire and leasing of objects in
connection with the provision of the aforesaid services,
included in this class; advice, consultancy and information
for the aforesaid included in the class. Telecommunication services, in particular computer
communication and internet access services;
telecommunication services via internet portals;
telecommunication services for consumers and public
administration, in particular in connection with the online
registration of vehicles and motorbikes; providing access to
content, websites and portals for the purchase and sale of
vehicles, vehicle trailers, motorcycles, bicycles, e-bikes,
building and agricultural machines, containers, boats and
other water vehicles and parts and fittings for vehicles,
motorcycles, bicycles, e-bikes, building and agricultural
machines, boats and water vehicles; providing access to
portals for the purchase and sale of vehicles, vehicle
trailers, motorcycles, bicycles, e-bikes, building and
agricultural machines, containers, boats and other water
vehicles and parts and fittings for vehicles, motorcycles,
bicycles, e-bikes, building and agricultural machines, boats
and water vehicles via communications devices; providing
access via internet portals and platforms to information
about the purchase and sale of vehicles, vehicle trailers,
motorcycles, bicycles, e-bikes, building and agricultural
machines, containers, boats and other water vehicles and
parts and fittings for vehicles, motorcycles, bicycles,
e-bikes, building and agricultural machines, boats and water
vehicles; providing user access to computer programs in data
networks; provision of access to content, websites and
portals for the provision of services, finance and energy
supplies; providing access to databases on the internet via
telecommunication, in particular providing access to online
databases for placing goods and services with interactive
access and possibility of immediate ordering; audio visual
and digital communications services; providing access to
databases and IT systems; electronic data exchange services;
transmission of digital files, data, voices and images
relating to applications based on artificial intelligence;
electronic bulletin board services [telecommunications
services]; worldwide computer network access services;
providing telecommunications connections to a global
computer network; providing an electronic mailbox; providing
online chat rooms and interactive electronic bulletin boards
on the internet; electronic communication by means of
chatrooms, chat lines and Internet forums; electronic
transmission of messages via chat sessions; providing access
to chatrooms; providing on-line facilities for real-time
interaction with other computer users; computerised
communication services and data transmission; podcast
broadcasting; transmission of podcasts; transmission of
digital files, in particular downloadable digital files
authenticated by non-fungible tokens [NFTs]; internet
service provider services; electronic transmission of data;
electronic transmission of computer programs via the
internet; internet based telecommunications services;
information transmission services via digital networks;
transmission of information online; transmission of messages
and images; streaming of audio material and video material
on the internet; distribution of data or audio visual images
via a global computer network or the internet; providing of
access to telecommunication warehousing services; email
services; message sending, receiving and forwarding;
forwarding messages of all kinds to internet addresses [web
messaging] and to portable communications devices;
transmission of information by data communications for
assisting decision making; transmission of database
information via telecommunications networks; broadcasting of
programs via the internet; providing third party users with
access to telecommunications infrastructure; telephone
communication services provided for hotlines and call
centers; rental, hire and leasing of objects in connection
with the provision of the aforesaid services, included in
this class; advice, consultancy and information for the
aforesaid, included in the class. Transport; storage; transportation of goods; freight and
cargo transportation; delivery services; arranging and
conducting the delivery of new and used vehicles, vehicle
trailers, motorcycles, bicycles, e-bikes, building and
agricultural machines, containers, boats and other water
vehicles and parts and fittings for vehicles, motorcycles,
bicycles, e-bikes, building and agricultural machines, boats
and water vehicles; arranging and conducting of mail order
delivery services; transport information, advice and
reservation services; rental services related to vehicles,
bicycles, e-bikes, boats, planes and vehicle accessories;
reservation services for vehicle rental; arranging car
rentals; rental of bicycles and e-bikes; vehicle
subscription services; bicycle and e-bike sharing services;
car sharing services; car pooling services; computerised
transport information services; booking agency services for
car hire; transport brokerage. Publishing and reporting; provision of electronic
publications and online electronic publications; providing
publications from a global computer network or the internet
which may be browsed; electronic desktop publishing;
publication of electronic books and journals on-line;
publication of the editorial content of sites accessible via
a global computer network; publication of books; writing and
publishing of texts, other than publicity texts; education;
entertainment; electronic library services for the supply of
electronic information (including archive information) in
the form of text, audio and/or video information;
audio-visual display presentation services for entertainment
purposes; provision of on-line entertainment; organization
and conduction of conferences, seminars, workshops,
exhibitions and competitions for cultural, entertainment or
educational purposes; photographic reporting and production
of documentaries; audio and video production, and
photography; providing online electronic publications (not
downloadable) in the nature of news articles relating to
vehicles, vehicle trailers, motorcycles, bicycles, e-bikes,
building and agricultural machines, containers, boats and
other water vehicles and parts and fittings for vehicles,
motorcycles, bicycles, e-bikes, building and agricultural
machines, boats and water vehicles for cultural, educational
and entertainment purposes; electronic library services for
the supply of electronic information (including archive
library services) with respect to information provided in
text, audio and/or video format, particularly regarding
supply and demand for goods and services; providing on-line
videos, not downloadable; providing computer assisted
courses of instruction, training and further training
courses; driving instruction; road safety training;
translation and interpretation; rental, hire and leasing of
objects in connection with the provision of the aforesaid
services, included in this class; advice, consultancy and
information for the aforesaid, included in the class. IT services, namely software development, programming and
implementation; updating and adapting of computer programs
according to user requirements; consultancy with regard to
webpage design; providing computer software from a global
computer network [non-downloadable]; provision of online
support services for computer program users; design,
conception, creation, hosting, administration and
maintenance of websites for others; maintenance of data
bases; design and development of web applications for
portable communications devices (included in the class);
provision of software [hosting] for the creation of
downloadable digital files authenticated by non-fungible
tokens [NFTs]; writing, development, design and maintenance
of computer software and applications for portable
communications devices and for computers to create, edit,
publish, view, search and save classified advertising, in
particular for a vehicle online marketplace for the purchase
and sale of new and used vehicles, vehicle trailers,
motorcycles, bicycles, e-bikes, building and agricultural
machines, containers, boats and other water vehicles and
parts and fittings for vehicles, motorcycles, bicycles,
e-bikes, building and agricultural machines, boats and water
vehicles; writing, development, design, configuration and
maintenance of computer software and applications for
portable communications devices and for computers to create,
edit, publish, view, search and arrange automotive vehicle
services, finance services and repair services for vehicles,
vehicle trailers, motorcycles, bicycles, e-bikes, building
and agricultural machines, containers, boats and other water
vehicles; providing artificial intelligence computer
programs on data networks; platforms for artificial
intelligence as software as a service [SaaS], in particular
for the human-machine interaction; software as a service
[SaaS] services featuring software for machine learning,
deep learning and deep neural networks; software as a
service [SaaS] featuring software for artificial
intelligence; software as a service [SaaS] featuring
software for artificial intelligence assistants, virtual
assistants, chatbots and message sending; hosting services
and software as a service and rental of software; cloud
computing services and software as a service relating to
computer software for enterprise resource planning,
inventory management and inventory monitoring; software as a
service in the field of point-of-sale [POS]; providing
on-line non-downloadable software for arranging dates and
appointments; software as a service in the field of document
management software systems; software as a service for car
dealers, in particular with online interfaces to banks;
providing on-line interfaces by means of non-downloadable
software for the registration of vehicles and motorcycles;
software as a service in the field of online shops; software
as a service for managing customer data and regarding
content management system software; cloud computing services
relating to computer software for managing customer data and
relating to content management system software; platform as
a service [PaaS] services featuring computer software
platforms that utilize artificial intelligence processing to
provide a virtual assistant with generative artificial
intelligence, generative pre-trained transformers, and large
language models for collecting, analyzing, processing,
providing, and summarizing data and business, commercial, or
financial information; providing virtual computer
environments through cloud computing; hosting virtual
environments; user authentication services using single
sign-on technology for online software applications; design
and development of single sign-on software; rental of single
sign-on software; development of and providing temporary use
of on-line non-downloadable software for automated
advertising placement in the form of adverts for customer
products; development of and providing temporary use of
on-line non-downloadable software for automated retrieving
of addresses of others; development of and providing
temporary use of on-line non-downloadable software for
database synchronization; development of and providing
temporary use of on-line non-downloadable software for
information transmission by telematic codes, in particular
QR-codes and bar codes; development of and providing
temporary use of on-line non-downloadable software for
operating online shops; providing temporary use of online
non-downloadable computer software for language translation;
provision of online non-downloadable computer software for
generating the artificial generation of human speech and
texts; provision of temporary use of on-line
non-downloadable chatbot software; hosting an internet
platform for the purchase and sale of vehicles, vehicle
trailers, motorcycles, bicycles, e-bikes, building and
agricultural machines, containers, boats and other water
vehicles and parts and fittings for vehicles, motorcycles,
bicycles, e-bikes, building and agricultural machines, boats
and water vehicles; creation and provision of web pages to
and for third parties; providing temporary use of on-line
non-downloadable software for website development, in
particular for the creation of homepages and websites;
provision of search engines for the internet and portable
communications devices, in particular for the search for and
the placement of classified advertisements for the purchase
and sale of vehicles, vehicle trailers, bicycles, e-bikes,
motorcycles, building and agricultural machines, containers,
boats and other water vehicles and parts and fittings for
vehicles, motorcycles, bicycles, e-bikes, building and
agricultural machines, boats and water vehicles, financing
and service offers; cloud computing; data warehousing;
application service provider [ASP], namely, hosting computer
software applications of others; off-site data backup;
hosting of digital content on the internet; server hosting;
providing back-up computer programs and facilities; rental
of database servers (to third parties); IT consultancy,
advisory and information services; IT consulting in relation
to data processing; IT security services in the nature of
protection and recovery of computer data; data recovery
services; data duplication and conversion services, data
coding services; cross-platform conversion of digital
content into other forms of digital content; IT services,
namely research and development, and implementation of
computer software; computer project management services [IT
services]; data mining; digital watermarking; computer
programming, design and development; technological services
relating to computers; computer network services; updating
of computer database software; data migration services;
updating of websites for others; monitoring of computer
systems by remote access; scientific and technological
services; preparation of technological reports by experts;
testing, authentication and quality control; inspection of
motor vehicles and e-bikes before transport [for
roadworthiness]; inspection of motor vehicles and e-bikes
[for road safety]; design services; design services for
display systems for promotional and presentation purposes;
artwork design; artificial intelligence consultancy; rental,
hire and leasing of objects in connection with the provision
of the aforesaid services, included in this class; advice,
consultancy and information for the aforesaid, included in
the class; providing electronic memory space on the Internet
for advertising goods and services; hosting electronic
memory space on the Internet for advertising goods and
services. Legal services; licensing of computer software, computer
programs and applications for portable communications
devices [legal services]; licensing industrial property
rights; licensing of technology; registration of domain
names [legal services]; information services relating to
consumer rights and other legal affairs; court reporting;
legal information and support services; rental, hire and
leasing of objects in connection with the provision of the
aforesaid services, included in this class; advice,
consultancy and information for the aforesaid, included in
the class.
Rules based override of machine learning output is described. In one or more implementations, the described architecture receives a listing for an item having one or more attributes describing the item. A category of the item, as output by a machine learning model based on the listing, is also received. The machine learning model is trained using a dataset of listings labeled with item categories. A set of deterministic rules is executed on the one or more attributes describing the item and the output of the machine learning model, including identifying the category of the item and applying one or more of the deterministic rules associated with the category. Based on the executing, a different category of the item is output and overrides the category of the item as output by the machine learning model.
Some aspects relate to technologies for a listing platform that facilitates searching multiple items to perform a service and mapping item orders to a service order. In accordance with some aspects, the listing platform receives input indicative of one or more services and a service provider to provide the service(s). The listing platform queries a services data store to identify items for performing the service(s). The listing platform queries an item listings data store to identify item listings for each item. The listing platform provides a user interface presenting an indication of the items and the item listings for each item. The listing platform receives input indicative of an item listing for each item. The listing platform generates at least one service order for the service(s) and item orders for the selected item listings. The listing platform also maps the item orders to the service order(s).
A computer-implemented method and system provides audio and language translation between a speaker at a first computing device and a listener at a second computing device. The first computing device inputs speech in the speaker’s language pre-defined as corresponding to the speaker, translates the speech from the speaker’s language into audio data in the listener’s language predefined as corresponding to the listener. The first computing device superimposes the speaker’s pronunciation as modeled by a speaker pronunciation model onto the audio data in the listener’s language so that the pronounced audio data in the listener’s language will sound as if it is spoken by the speaker. The speaker pronunciation model is trained on the speaker’s voice speaking the speaker’s language and remains stored at the first computing device. The pronounced audio data is streamed to the second computing device while the speaker at the first computing device is speaking.
A system is provided that automatically notifies a user to perform maintenance for one or more parts of an appliance. The system automatically detects that an appliance has been purchased by a user and, in response to automatically detecting that the appliance has been purchased by the user, stores, in a database of an online electronic transaction platform, an appliance profile for a user comprising a list of appliances including the appliance associated with the user, the list of appliances being associated with a plurality of manufacturers. The system selects, in the database of the online electronic transaction platform, an individual appliance from the list of appliances and determines a maintenance schedule for one or more parts associated with the individual appliance. The system automatically notifies the user to perform maintenance for the one or more parts according to the maintenance schedule.
Some aspects relate to technologies for automatic crypto wallet balancing. The crypto wallet balancing can involve rebalancing a first crypto wallet to reduce its balance. This can include monitoring a total value of the first crypto wallet over a time period, where the first crypto wallet is owned by a first user and associated with a first blockchain account on a first blockchain. A rebalance trigger event is detected based on the total value of the first crypto wallet exceeding a maximum threshold value for a threshold amount of time during the time period. Responsive to the rebalance trigger event, a smart contract on the first blockchain is executed, causing: generation of a transaction block transferring digital assets from the first blockchain account to a second blockchain account associated with a second crypto wallet owned by the first user, and addition of the transaction block to the first blockchain.
G06Q 20/36 - Payment architectures, schemes or protocols characterised by the use of specific devices using electronic wallets or electronic money safes
H04L 9/00 - Arrangements for secret or secure communicationsNetwork security protocols
65.
SUMMARY CONTENT GENERATION AND VALIDATION FOR PARTS ON LISTING PLATFORMS
Some aspects relate to technologies for generating and validating content providing summary information for parts on a listing platform, such as compatibility information and item characteristic information for the parts. In some aspects, input data comprising attribute-value pairs based on one or more items is accessed for content generation. The input data and one or more pre-defined generation prompts are provided to a first generative model, which generates content comprising natural language text as output. In some aspects, the content is validated by providing the content and one or more pre-defined validation prompts to a second generative model. If approved, the content can be provided for presentation on a user device. If rejected, the content is not provided for presentation. In further aspects, the rejected content is used by a third generative model for tuning the pre-defined generation prompts.
A request associated with a computing task is received. A machine learning (ML) model is identified based on the request. Metadata preconfigured for the ML model is retrieved. A graphics processing unit (GPU) resource is dynamically allocated based on the metadata preconfigured for the ML model. The computing task associated with the request is processed using the allocated GPU resource.
An artificial intelligence (AI)-based smart actioning system is described. A system receives an activity associated with a website, the activity having one or more attributes. Based on the attributes, the system generates a vector associated with the activity. The system executes a machine learning model that is training using a dataset including a vector space of a set of vectors associated with a set of historical activities associated with the website. This may include executing a vector space similarity search, such as a k-nearest neighbor (k-NN) search, between the vector and the set of vectors. The system generates a prompt for a large language model (LLM) based on one or more vectors identified from the vector space similarity search, and outputs a result received from the LLM that indicates a risk level of the activity.
Systems and methods are directed to pre-processing images and triggering generation of images having natural backgrounds. The imaging system accesses a source image of an item and isolates the item by removing a background from the source image. An item category of the item is identified using an image classification model. Based on the item category, additional information regarding the item are identified including a typical orientation. The imaging system generates a prompt that includes at least some of the additional information and instructions to generate images having a natural background. The imaging system also generates a guidance image that is a combination of the source image with the background removed and a suggested background. Using the prompt and the guidance image, an artificial intelligence (AI) model is triggered to generate one or more images of the item having the natural background and a shadow of the item.
Digital content authentication using a cryptographic possession factor is described. In one or more examples, a navigation request to access a webpage is received and verified, over a plurality of iterations over a session, that an originator of the navigation request maintains a cryptographic token as bound within a browser through use of a public key associated with the cryptographic token. A log describing the verifying is stored and then a determination is made that the webpage involves additional authentication usable to extend the session to access the webpage. A level of the authentication is selected for access to the webpage based on the cryptographic token maintained at the browser and the log. Access to the webpage is controlled based on the selected level of authentication.
Some aspects relate to technologies for software development using a component ecosystem with generative artificial intelligence. In accordance with some aspects, natural language text is received for code generation. Based on the natural language text, one or more component specifications are identified, and a prompt is generated from the natural language text and the component specification(s). Each component specification has a corresponding input, a corresponding output, and one or more corresponding properties. Given the prompt, a generative model generates one or more component instances by hydrating one or more properties of the component specification(s). When multiple component instances are generated, the generative model can also determine a control flow logic specifying an order for the component instances, and the generative model can further generate a translation layer between successive component instances.
Traffic modeling network application services is described. An example system includes a network traffic model that defines traffic relationships for data traffic in a network between a network topology of an application and a plurality of different access points serving network connections through different physical or logical tiers of the network topology with at least one instance of the application based in part on domain-subdomain relationships obtained from a domain name system record model for communicating on the network. A first access point in the system, in response to receiving a first communication between a first endpoint and a first instance of the application, serves the first communication on a first network connection through a first tier of the network topology with the first instance of the application based on a first traffic relationship defined by the network traffic model for the first network connection.
Some aspects of the present technology relate to technologies for performing knowledge graph-enhanced retrieval augmented generation for an e-commerce platform. In accordance with some configurations, a task specification comprising a task and a task input is obtained. This task specification is used to perform entity linking to extract task-aware context. In accordance with some configurations, the task-aware content is concatenated with the task specification to generate a prompt to a large-language model, which uses the prompt as input to generate a response to the task. In some configurations, the response is provided using a user interface.
A method for constructing a point-in-time global consistent graph is described. The method includes accessing a stream of data from a transactional graph database, receiving, at a scalable persistent computational platform that is separate from the transactional graph database, a graph analytics query that indicates a time attribute value, in response to receiving the graph analytics query, constructing, at the scalable persistent computational platform, a point-in-time graph snapshot based on the stream of data and the time attribute value, and processing, at the scalable persistent computational platform, the graph analytics query with the point-in-time graph snapshot.
Transaction data associated with a user is identified. A first embedding representing the transaction data is generated using a deep learning machine learning (ML) model. A second embedding interpretable by a large language model is generated using an adaptor ML model based on the first embedding. A text summary associated with one or more transactions is identified. A third embedding representing the text summary is generated using the large language model. An adaptor ML model is trained based on a similarity between the second embedding and the third embedding.
G06Q 20/40 - Authorisation, e.g. identification of payer or payee, verification of customer or shop credentialsReview and approval of payers, e.g. check of credit lines or negative lists
Query dependent threshold generation for search result filtering is described. In one or more implementations, a user query entered via a search platform is received, and in response, items are retrieved from a storage device based on the user query. Using a first machine learning model, relevance scores are generated for the items, and the relevance scores represent degrees of relevance of respective items with respect to the user query. Using a second machine learning model, a relevance threshold is generated for the user query based on one or more features of the user query. The items are filtered based on the relevance scores and the relevance threshold, and the filtered items are communicated over a network for display in a user interface of the search platform.
A search query is received. A first plurality of items is identified. A reference item is identified based on the search query and the first plurality of items. A plurality of contextual distances is determined using a distance calculation metric. A machine learning model is used to rank a second plurality of items based on the plurality of contextual distances. The second plurality of items is caused to be displayed. The second plurality of items is arranged in an order in accordance with the ranking.
In implementations of systems and procedures for a service provider system, a computing device implements predictive recommendation generation. A registration component is employed to register a specific interaction sequence defining an order of user interactions to be performed by a single user during a session on a platform. The specific interaction sequence is predicted to proceed to a request for a recommendation result by the single user. Additionally, an interaction monitor is configured to monitor, in real-time, interaction sequences of a plurality of users. In response to detecting the specific interaction sequence for a user of the plurality of users, the interaction monitor triggers, in real-time, pre-calculation of the recommendation result for the user, and a cache stores the pre-calculated recommendation result.
Some aspects of the present technology relate to technologies for providing a self-learning apparatus to detect all possible financial fraud behavior within a payment processor. In accordance with some configurations, payment protocols are modeled as a graph and a model simulation exhaustively searches all possible states spaces. Invariants are checked at each step and invariant violations (i.e., exploits) are logged with the sequence of steps that led to a respective exploit. In some aspects, upon determining a fix for the exploit, the graph model is updated and the model simulation is run again to confirm the fix has eliminated the exploit. A policy update or code enhancement may be implemented to modify the payment protocol accordingly.
G06Q 20/40 - Authorisation, e.g. identification of payer or payee, verification of customer or shop credentialsReview and approval of payers, e.g. check of credit lines or negative lists
G06F 9/455 - EmulationInterpretationSoftware simulation, e.g. virtualisation or emulation of application or operating system execution engines
Bitmask encoding-based personalized style generation is described. A personalization system presents different instances of digital content in a user interface and prompts a user to indicate interest or disinterest in an individual digital content items. Based on the input indicating interest or disinterest, a bitmask encoding is generated and returned to the personalization system. The personalization system queries a database to identify respective instances of digital content that correspond to each binary value included in the bitmask encoding, which is then used to generate a prompt that causes a machine learning system to generate a personalized style for the user from which the favorable or unfavorable indications were received. The machine learning system generates a personalized style for an individual and identifies different instances of digital content of interest to the individual, and the personalized style is used to generate a user interface for the individual.
A vehicle diagnosis machine-learning system is described. In one or more examples, the diagnostic service receives a query specifying a vehicle issue, identifying the vehicle corresponding to the query, and obtains the vehicle's history. The diagnostic service then generates a user question using one or more machine-learning models based on the query, the vehicle, and its history, and receives a response to this question via a user interface. A prompt is generated for processing by the machine-learning models, incorporating the query, vehicle, vehicle history, user question, and response. The result of processing of the prompt by a machine-learning model is presented to display in a user interface.
G07C 5/08 - Registering or indicating performance data other than driving, working, idle, or waiting time, with or without registering driving, working, idle, or waiting time
G06F 3/0482 - Interaction with lists of selectable items, e.g. menus
Learning model task performance using a graph is described. A computing system can receive a prompt for performing a task using a learning model, where the task includes intermediate tasks. The computing system generates a graph to represent relationships between at least one term included in the prompt and additional terms associated with performing the task. The graph includes nodes corresponding to the at least one term and the additional terms and edges connecting the nodes. The edges correspond to the relationships between the at least one term and the additional terms. The computing system performs the intermediate tasks based on the graph to obtain a response to the prompt, where the intermediate tasks are associated with the additional terms. In some cases, the computing system removes one or more nodes from the graph responsive to results of the intermediate tasks. The computing system broadcasts the response to the prompt.
Image based attribute generation for item descriptions is described. A computing device receives a digital image depicting an item and encodes one or more embeddings extracted from the digital image using an image encoder. The image encoder is implemented by at least one machine learning model. The one or more embeddings are converted into at least one attribute of the item using a text decoder of the at least one machine learning model that is trained based on a set of attribute training values. A correction for the at least one attribute is generated. Based on the correction and the at least one attribute, item attribute values are extracted including to replace at least one item attribute with a corresponding attribute training value from the set of attribute training values.
In accordance with the described techniques, vector data to be inserted into a vector database is obtained. The vector data is indexed into a mutable index of the vector database that supports real-time vector indexing. Periodically, the vector data is merged from the mutable index into one or more immutable graph-based indexes. A query vector is received, and a vector search is performed in the mutable index and the one or more graph-based immutable indexes based on the query vector.
Some aspects relate to technologies for performing attribute-aware vector searches using adjusted similarity thresholds. In accordance with some aspects, a search query is received. Items are mapped onto an extended embedding space based on encoded attributes for the plurality of items. The search query is also mapped onto the extended embedding space based on encoded attributes for the search query. An adjusted similarity threshold is generated based on a number of attributes to be matched for the search query. Using the adjusted similarity threshold, items that are similar to the search query in the extended embedding space are retrieved. Search results based on the one or more retrieved items can be provided as a response to the search query.
Techniques for creating a user network based on user interest in a product are described. For example, a first search query is received from a first client device. The first client device is associated with a first user. A second search query is received from a second client device. The second client device is associated with a second user. The first search query is matched to a product. The second search query is matched to the product. A user network corresponding to the product is created. The first user is added to the user network based on the matching of the first search query to the product. The second user is added to the user network based on the matching of the second search query to the product.
The technology described herein relates to systems, methods, and computer storage media, among other things, for generating and providing a journey profile for an item (e.g., an item offered on an ecommerce platform) that includes one or more authentication verifiers, and for generating and providing a visualization associated with the journey profile (e.g., a map visualization for item at a plurality of locations corresponding to a plurality of historical location identifiers). For example, the journey profile may be generated based on linking a unique item identifier of the item to the authentication verifier for the item (e.g., the authentication verifier corresponding to a certificate of authenticity or another form of authentication data that is based on a physical inspection of the item).
Retrieval-augmented item attribute generation techniques are described. An attribute generation system receives input describing a target item, such as a title of the target item, an image depicting the target item, or a combination thereof. Given the input, the attribute generation system generates a latent space embedding representation of the target item and identifies similar items based on the latent space embedding representation. The attribute generation system then identifies, for each similar item, one or more aspects that include information describing the similar item. Similar item aspects and the input describing the target item are used to generate a prompt that causes a machine learning system to generate name-value attribute pairs for the target item. The name-value attribute pairs are output for display in a user interface and selectable for inclusion in a digital marketplace listing for the target item.
In the implementation of techniques for generating wait time predictions with machine learning, a system receives time series data and contextual data corresponding to a plurality of communication requests. Based on the time series data and the contextual data, a system trains a machine learning model to generate a wait time prediction for a communication request. The system receives the communication request from a client device and recent time series data and recent contextual data corresponding to recent communication requests. Based on the recent time series data and the recent contextual data, the system updates the machine learning model. Based on the communication request, the system generates the wait time prediction via the updated machine learning model. Upon generation of the wait time prediction, the system broadcasts the wait time prediction to the client device.
Some aspects relate to technologies for automated end-to-end testing using a continuous delivery infrastructure and centralized data storage. In accordance with some aspects, a run identifier is generated for a run of a testing pipeline comprising a sequence of test phases that each tests a corresponding code unit. The run identifier is stored in a data store accessible to the testing pipeline. A first test phase is performed to test a first code unit, resulting in the generation of a first test data that is stored in the data store in association with the run identifier. A second test phase is then performed to test a second code unit by accessing the first test data from the data store for use by the second test phase. A second test data is generated by the second test phase and stored in the data store in association with the run identifier.
Some aspects relate to technologies for graph databases with 1-hop sub-query result caching. In accordance with some aspects, cache entries are generated that comprise key-value pairs, each having a cache key based on a vertex identifier of a vertex from a 1-hop sub-query instance and a value based on results of a 1-hop graph traversal from the vertex based on the 1-hop sub-query instance. A cache key can also be based on a property value of a predicate from a 1-hop sub-query instance. During graph query processing, a 1-hop sub-query instance is obtained based on the graph query. A cache key is generated using a vertex identifier for the vertex from the 1-hop sub-query instance. A lookup is performed in the cache data using the cache key, and an output is provided using one or more leaf vertex identifiers from a value of a cache entry having the cache key.
Systems and methods are directed to managing access to a software instance stored on a virtual machine on a blockchain using a smart contract. The system receives, from a client device of a user, an authorization request to access the software instance. The authorization request can include a token associated with the user. In response to receiving the authorization request, the system accesses the smart contract. Based on the token, a determination is made, through the smart contract, whether the user is an owner of a license to the software instance. In response to determining that the user is the owner, a further determination is made, through the smart contract, whether the license has expired without renewal. Based on the license having expired without renewal, the smart contract autonomously triggers the virtual machine to destroy a container comprising the software instance.
G06F 9/455 - EmulationInterpretationSoftware simulation, e.g. virtualisation or emulation of application or operating system execution engines
G06Q 20/12 - Payment architectures specially adapted for electronic shopping systems
G06Q 20/36 - Payment architectures, schemes or protocols characterised by the use of specific devices using electronic wallets or electronic money safes
H04L 9/00 - Arrangements for secret or secure communicationsNetwork security protocols
H04L 9/32 - Arrangements for secret or secure communicationsNetwork security protocols including means for verifying the identity or authority of a user of the system
92.
SYSTEM AND METHOD FOR DETECTING ABNORMAL ORDER PAYMENT BEHAVIOR USING GRAPH MODEL EMBEDDING AND ANOMALY DETECTION
Some aspects of the present technology relate to technologies for detecting abnormal payment behavior using graph model embedding and anomaly detection. In accordance with some configurations, order payment data is collected from various sources, including e-commerce platforms, financial institutions, and payment processors. The collected payment data is structured as a graph for each order. Nodes represent individual payment transactions related to the order. Graph embedding techniques are applied to transform the payment data graph into a numerical vector space representation. The embedded data is analyzed for a particular interval of time to identify recurring patterns. A baseline for normal patterns is established for the interval of time and any patterns that deviate significantly from the baseline are flagged as potential abnormal payment behaviors. In some aspects, a graph visualization comparison tool aids in the transparent verification of reconciliations and provides intuitive insights for stakeholders.
G06Q 20/40 - Authorisation, e.g. identification of payer or payee, verification of customer or shop credentialsReview and approval of payers, e.g. check of credit lines or negative lists
Methods, systems, and computer storage media for providing an AI-supported fraud detection engine in an item listing system are described. The M-supported fraud detection engine leverages language models (e. g., LLMs) to interpret unstructured data as language model output (i. e., language model output comprising context-free annotation tags including observations, tags, category, case summary, and reasoning annotations). The language model output (i. e., context-free annotation tags) is mapped to context-aware target tags without explicit tagging (i. e., zero-shot approach) to generate structured domain-specific tags. The zero-shot approach allows the AI-supported fraud detection engine to recognize and categorize new fraud pattems dynamically. The structured domain-specific tags can be integrated into traditional fraud detection models that leverage the insights from the structured domain-specific tags to generate fraud detection recommendations. The Al-supported fraud detection engine's ability to generate fraud detection recommendations based on structured domain-specific tags leads to more targeted and accurate responses to potential fraud cases.
A videoconference system is described that generates a video for a room including multiple videoconference participants and outputs the video as part of the videoconference. The videoconference system is configured to generate the video as including a detailed view of one of the multiple videoconference participants located in the room. To do so, the videoconference system detects user devices located in the room capable of capturing video and determines a position of each user device. The videoconference system then detects a user speaking in the room and determines a position of the active speaker. At least one of the user devices is identified as including a camera oriented for capturing the active speaker. Video content captured by one or more user devices is then processed by the videoconference system to generate a detailed view of the active speaker.
Systems and methods are directed to redacting and reinstating personal identifying information (PII) from text data. A PII management system accesses an input text and identifies, using one or more redaction components, PII mentions in the input text to be redacted. A placeholder manager of the PII management system replaces each unique PII string of a final set of PII mentions with a non-PII string to generate redacted text, whereby the non-PII string is generated by the placeholder manager. The placeholder manager also generates a mapping dictionary that maps each unique PII string to the non-PII string that replaces it. The mapping dictionary is used to reinsert one or more unique PII strings after processing of the redacted text. The redacted text is then transmitted to a downstream component for the processing.
Some aspects relate to technologies for using synthetic items generated by a generative model to perform item retrieval for a listing platform based on seed item listings. In some examples, a textual indication of a seed item listing from a listing platform is received. Based on the seed item listing, a textual indication of one or more synthetic items generated by a generative model are obtained. The textual indication of each synthetic item can be generated by the generative model at runtime or previously generated by the generative model and retrieved at runtime using one or more caching techniques. A search is performed on an item listings data store for the listing platform based on the textual indication of the one or more synthetic items to identify one or more item listings. An indication of the one or more item listings is provided for presentation as item listing recommendations.
In the implementation of techniques for distributing digital content with edge devices, a system receives a request including digital content from a client device for uploading the digital content. Based on the request, the system identifies a nearest edge device from a plurality of edge devices. The system transmits instructions to the nearest edge device to generate a unique identifier corresponding to the digital content and to store the digital content. The system initiates one or more requests to a predefined number of one or more edge devices from the plurality of edge devices to store the digital content. The system receives confirmation from the predefined number of one or more edge devices that the digital content is stored. The system generates a response including content corresponding to the unique identifier, which is usable to access the digital content stored. The system broadcasts the response to the client device.
Systems and methods are directed providing a personalized search experience. The system receives a broad term query and identifies a plurality of search results, each having a corresponding embedding. The plurality of search results are clustered into embedding clusters. The system then presents a set of prompts whereby each prompt comprises a display of two or more search results, each chosen from a different embedding cluster. With each selection from a prompt, the system narrows down on one or more embedding clusters. Individual search results from the narrowed-down embedding cluster(s) are then displayed and a user can provide an indication of like or dislike. With each indication, the system dynamically updates a user preference embedding that represents an item that the user desires. The updated user preference embedding is then used to identify updated search results having closest corresponding embeddings to the user preference embedding.
Some aspects of the present technology relate to technologies for context and knowledge sharing abstraction across entities using a relational hierarchical key mapping system. In accordance with some configurations, a first set of one or more data points for a first set of one or more keys is received at a first domain and stored in a schema-less data store. A second set of one or more data points for a second set of one or more keys is received at a second domain and stored in the schema-less data store. Based on at least one shared key, the first set of one or more data points and the second set of one or more data points are linked as a virtual entity such that a request to retrieve data points of the one or more data points is performed as a single seek and a single API call.
