Simulating interaction with an object represented in a virtual environment is disclosed. A system includes one or more circuits configured to receive an indicator of a sensed touch in a virtual environment and to determine, based on the indicator, an area of the sensed touch. The one or more circuits are further configured to generate a simulated touch by applying a field to one or more touch simulators, the field actuating the one or more touch simulators by linearly displacing an element of the one or more touch simulators. In implementations, the touch simulators are actuated based on data describing textures and weights of the object represented in the virtual environment.
A user may request a recommendation for an item from other users. Other users may respond to the request by recommending for or against items. Users may up-vote or down-vote the recommendations or responses of other users. The recommendations of the other users may be used to identify items and provide one or more recommendations to the requesting user. The original question and the responses may form a conversation thread. The recommendations may be inserted into the thread as responses, may be presented alongside the thread, or may be presented at the end of the thread. The recommendations may be based on one or more attributes of the user. The weight of the recommendations provided by other users may vary.
G06Q 30/0282 - Notation ou évaluation d’opérateurs commerciaux ou de produits
G06Q 50/00 - Technologies de l’information et de la communication [TIC] spécialement adaptées à la mise en œuvre des procédés d’affaires d’un secteur particulier d’activité économique, p. ex. aux services d’utilité publique ou au tourisme
3.
Digital Content Control Based on Nonfungible Tokens
Techniques are described, as implemented by computing devices, to provide digital content to users through use of nonfungible tokens (NFTs). This is performed by leveraging a blockchain such that digital content associated with an item is made available to encourage the user to interact with NFTs on the service provider platform (e.g., discounts, NFT listing functionality, automatic initiation of NFT transfers, and so forth) based on a user's possession of the NFT.
G06Q 20/38 - Protocoles de paiementArchitectures, schémas ou protocoles de paiement leurs détails
H04L 9/00 - Dispositions pour les communications secrètes ou protégéesProtocoles réseaux de sécurité
H04L 9/32 - Dispositions pour les communications secrètes ou protégéesProtocoles réseaux de sécurité comprenant des moyens pour vérifier l'identité ou l'autorisation d'un utilisateur du système
In accordance with techniques for image to structured workflow diagram conversion, an image of a human-drawn workflow diagram is received by a workflow platform. Using one or more image recognition algorithms, the workflow platform detects shapes and lines in the human-drawn workflow diagram, and in one or more implementations, the workflow platform extends the detected lines. Relationships between the shapes are determined based on relative positionings of the extended lines with respect to the shapes. The workflow platform is configured to generate a structured workflow diagram for display in a user interface, and the structured workflow diagram includes compute nodes representing the shapes that are connected by generated lines representing the relationships. In addition, a structured workflow file is generated representing the structured workflow diagram.
G06T 7/50 - Récupération de la profondeur ou de la forme
G06T 11/20 - Traçage à partir d'éléments de base, p. ex. de lignes ou de cercles
G06V 10/48 - Extraction de caractéristiques d’images ou de vidéos en cartographiant les valeurs caractéristiques du motif en espace paramétrique, p. ex. transformation de Hough
G06V 20/70 - Étiquetage du contenu de scène, p. ex. en tirant des représentations syntaxiques ou sémantiques
Systems and methods are directed to tuning prompt inputs for a large language model (LLM). A prompt tuning system accesses multiple types of unstructured data for use in generating a fusion prompt input embedding. The prompt tuning system then generates a respective embedding for each of the multiple types of unstructured data. The respective embedding for each of the multiple types of unstructured data is provided to a fusion graph neural network (GNN). The fusion GNN generates the fusion prompt input embedding by combining information from the respective embedding for each of the multiple types of unstructured data into a single embedding. The fusion prompt input embedding provides context for a pure text input that comprises instructions for an output from the LLM. The fusion prompt input embedding and a token embedding representing the pure text are inputted to the LLM. An output of the LLM is then displayed. (FIG. 6)
Methods, systems, and computer storage media for providing generative artificial intelligence (AI) security management using a generative AI security engine in an item listing system. A generative AI security engine supports generative AI security management based on security analysis and detection operations for a plurality of generative-AI-supported applications and generative AI models. In operation, a request associated with prompt data is communicated from a generative AI client. Based on communicating the request, a response that is generated based on a redacted version of the prompt data is received at the generative AI client. The prompt data is analyzed using a plurality of security engine operations to cause generation of the redacted version of the prompt data. The redacted version of the prompt data is used to generate the response at a generative AI model. The response is caused to be generated at an interface associated with the generative AI client.
A triggering event is received. A paging entity and an action entity are identified based on a job description associated with the triggering event. Job data is retrieved from a data source based on the paging entity. The pagination engine is integrated with a plurality of data sources for job data retrieval. An action on the job data is performed based on the action entity. The action engine is integrated with a plurality of downstream components for action processing.
Methods, systems, and computer storage media for providing brand-focused listing management in an item listing system are described. Using a brand-focused machine learning model, the brand-focused listing management engine supports listing accuracy compliance with the item listing marketplace standards, and optimization of search visibility and customer experience. The brand-focused listing management engine also supports identifying deliberate bypassing or violating of the item listing system's guidelines for product listings and employs corrective actions to improve listing quality and functionality associated with listing management including seller listing flow and optimizing visibility based on listing quality signals. In operation, an item listing associated with an item listing system is accessed. The listing is analyzed using a brand-focused machine learning model that is trained based on a multi-dimensional authenticity analysis dataset. Based on analyzing the item listing, a brand-focused security notification associated with the item listing is generated. The brand-focused security notification is communicated.
The technology described herein relates to systems, methods, and computer storage media, among other things, for determining whether an electronic transmission (e.g., associated with an electronic payment transaction) should be blocked (e.g., based on being a fraudulent transaction). In embodiments, a policy-based reinforcement learning risk decision agent is used to make these determinations for a plurality of stages associated with the electronic payment transaction (e.g., a pre-authorization stage, a post-authorization stage, and a delay-captured stage). The policy-based reinforcement learning risk decision agent can be trained using previous electronic payment transaction data for previous electronic payment transactions. For example, this particular agent can be trained using pre-authorization electronic payment transaction data, post-authorization electronic payment transaction data, and delay-captured electronic payment transaction data for each of the previous electronic payment transactions.
G06Q 20/40 - Autorisation, p. ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasinExamen et approbation des payeurs, p. ex. contrôle des lignes de crédit ou des listes négatives
Methods, systems, and computer storage media for providing brand-focused machine learning model training in an item listing system are described. The brand-focused machine learning model training engine supports training a brand-focused machine learning model that predicts brands for item listings that do not include brand information. The training can be based on novel training techniques and training features from data associated with the multi-dimensional authenticity analysis dataset (i.e., brand protection and verification data and item listing system data) to cause generation of a brand-focused machine learning model that is subsequently deployed. In operation, a multi-dimensional authenticity analysis dataset associated with a plurality of brands is accessed. A brand-focused machine learning model using the multi-dimensional authenticity analysis dataset is trained. Training the brand-focused machine learning model is based on brand multi-dimensional authenticity features. The brand-focused machine learning model is deployed in an item listing system to support one or more applications.
Methods, systems, and computer storage media for providing brand-focused security administration in an item listing system are described. Using a brand-focused machine learning model, the brand-focused security administrator engine supports generating different comprehensive reports and analytics on brand-related security metrics, including brand compliance rates, infringement incidents, enforcement actions taken, and overall item listing system integrity. The brand-focused security administrator engine also supports visualizing the brand-focused security analytics results data in a manner that enhances usability and decision-making by presenting complex brand-focused security analytics results data in a clear, actionable format. In operation, brand-focused security data is accessed. The brand-focused security data is analyzed using a brand-focused machine learning model that is trained based on a multi-dimensional authenticity analysis dataset. Brand-focused security analytics results data is generated. The brand-focused security analytics results data is communicated to cause display of one or more visualizations based on the brand-focused security analytics results data.
Some aspects relate to technologies for performing item retrieval on a listing platform using clusters of interchangeable parts formed using fitment data. In accordance with some aspects, item data is accessed for each of a plurality of part item listings on a listing platform, where the item data for each part item listing includes fitment data. An item embedding is generated for each part item listing using the item data. The item embeddings are clustered to generate a plurality of clusters, wherein each cluster includes one or more item embeddings. Cluster data is stored for the plurality of clusters. The cluster data for each cluster associates a cluster identifier and an item listing identifier for each item embedding in the cluster. The cluster data can be leveraged to perform item retrieval for the listing platform.
Some aspects of the present technology relate to technologies for progressive decisioning of behavioral risk by dynamically analyzing and synthesizing behavioral fraud pattern. In accordance with some configurations, a signature comprising time series data corresponding to a user of an online transaction platform is received at various checkpoints in a user workflow. These signatures are stored in a knowledge graph. Upon receiving a search query from a business user for a combination of signatures indicative of fraud, the search query is converted, without human intervention, into graph traversal logic of the knowledge graph. The knowledge graph is traversed, in real-time, utilizing the graph traversal logic. Based on the traversing, the user workflow can be dynamically modified to prevent fraudulent activity.
G06Q 20/40 - Autorisation, p. ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasinExamen et approbation des payeurs, p. ex. contrôle des lignes de crédit ou des listes négatives
In an example embodiment, an item characteristic is received, the item characteristic pertaining to an item being listed for sale, by a seller, via an ecommerce service. Then, a plurality of past transactions of items having the item characteristic are analyzed. Based on this analysis, a first set of one or more optimal listing configuration parameters are identified in accordance with a first set of listing criteria. Then, the first set of one or more identified optimal listing configuration parameters to the seller in a user interface that permits the seller to change one or more listing configuration parameters based on the presentation.
Item request management employing secure hashes to queue item requests for processing is leveraged with an online platform supporting item listings. In one or more implementations, item request data associated with an item request for a listed item is received. The item request data includes information such as a recipient for the item and/or an item procurement quantity associated with the item request. A secure hash is generated from the item request data, with the secure hash storable in a secure hash ring to represent the item request data in the secure hash ring. A queued item request is generated in an item request processing queue from the secure hash, and the queued item request is processed to allocate an amount of the item equal to the procurement quantity to the recipient.
09 - Appareils et instruments scientifiques et électriques
35 - Publicité; Affaires commerciales
36 - Services financiers, assurances et affaires immobilières
39 - Services de transport, emballage et entreposage; organisation de voyages
42 - Services scientifiques, technologiques et industriels, recherche et conception
Produits et services
(1) Downloadable computer software for online commerce, online marketplace, and online auction services; Downloadable and recorded computer e-commerce software to allow users to conduct electronic business transactions in online marketplaces via a global computer network; (1) Online commerce services, namely, operating online marketplaces for buyers and sellers of goods and/or services; Providing a website featuring an online marketplace for exchanging goods and services with other users; Online auction services; Provision of an on-line marketplace for buyers and sellers of goods and/or services;
(2) Safe deposit box services;
(3) Storage of high value collectibles; Storage services being vault services for storing high value collectibles;
(4) Providing online non-downloadable computer software for online commerce, online marketplace, and online auction services; Electronic data storage; Providing temporary use of online, non-downloadable e-commerce software to allow users to conduct electronic business transactions in online marketplaces via a global computer network;
API recommendations based on performance metrics are described. In one or more implementations, an API recommendation system receives a request from a client device. Based on a condition of the request, the API recommendation system selects an application programming interface (API) of a plurality of APIs for performance of the request and stores performance metrics related to the performance of the request by the API in a performance index. The API recommendation system then receives a subsequent request and, using a machine learning model, determines a recommendation on calling the API for performance of the subsequent request by analyzing the performance metrics in the performance index. The API recommendation system then outputs instructions for performing the recommendation on calling the API.
Some aspects of the present technology relate to technologies for performing fraud detection on online transaction platforms through user behavior sequence data. In accordance with some configurations, a multi-task convolutional neural network (MTCNN) model is used to predict, in real-time, whether user behavior sequence data is indicative of fraudulent activity. To perform fraud detection in such configurations, a one-layer convolutional neural network architecture with multi-range kernels is employed. The MTCNN model receives a sequence of page browsing signals corresponding to a buyer. Each page browsing signal corresponds to a position in the sequence. One or more portions of the page browsing signals are selected. Each of the one or more portions of the page browsing signals and the corresponding position are embedded in one or more sequence embeddings. A fraud risk for each of the one or more sequence embeddings is predicted utilizing the MTCNN model.
G06Q 20/40 - Autorisation, p. ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasinExamen et approbation des payeurs, p. ex. contrôle des lignes de crédit ou des listes négatives
A system for creating generated descriptive text is provided. A prompt having first facts for an item is received and parsed to extract a first fact in a format. Second facts are generated where the first fact and the second facts are output in the format. A search query is generated that includes the first fact and the second facts and then a search is conducted using the search query. An output is generated based on the results. The output includes a suggested description of the item using at least one first fact of the first facts and the second facts. The output also has a summarization of the plurality of first facts and the second facts along with differences between the first facts and the second facts. A distribution of the plurality of first facts and the second facts in the results is provided in the output.
G06F 16/215 - Amélioration de la qualité des donnéesNettoyage des données, p. ex. déduplication, suppression des entrées non valides ou correction des erreurs typographiques
G06F 16/248 - Présentation des résultats de requêtes
G06F 16/25 - Systèmes d’intégration ou d’interfaçage impliquant les systèmes de gestion de bases de données
20.
MACHINE LEARNING MODEL TRAINING USING A SELF-TRAINING APPROACH FOR KNOWLEDGE DISTILLATION
A plurality of data items associated with user-generated content is identified. A first subset of data items in the plurality of data items is annotated using a first machine learning (ML) model. The first ML model is trained based on the first plurality of labels generated for the first subset of data items. The first ML model is used to annotate a second subset of data items in the plurality of data items. A second ML model is trained based on a second plurality of labels generated based on the annotating of the second subset of data items.
A plurality of data items associated with user-generated content is identified. A first subset of data items in the plurality of data items is annotated using a first ML model. A second ML model is trained based on the first plurality of labels generated for the first subset of data items. A second subset of data items in the plurality of data items is annotated using the second ML model trained. A third ML model is trained based on a second plurality of labels generated for the second subset of data items based on the annotating.
Systems and methods of improving the functioning of a streaming platform system by managing database change stream offsets using a time series database are disclosed. In some example embodiments, a computer system retrieves an offset value from a plurality of offset values stored in a time series database, with the plurality of offset values being indexed in the time series database in time order, and the retrieved offset value being retrieved using a time parameter, and then the computer system transmits a data request to a stream-processing platform, with the data request comprising the retrieved offset value, and the data request being operable to retrieve a data record stored in association with the retrieved offset value in a storage layer of the stream-processing platform using the offset value.
A plurality of nodes is deployed in at least two distributed data centers. The plurality of nodes comprises a leader node in communication with a client device and at least one follower node in communication with the leader node. The leader node locally stores first transaction data and the at least one follower node each locally stores second transaction data. The leader node receives transaction data from the client device and updates the first transaction data based on the received transaction data. The leader node forwards the received transaction data to the at least one follower node and the at least one follower node updates the second transaction data based on the forwarded transaction data. The updated second transaction data stored in the at least one follower node is synchronized with the updated first transaction data stored in the leader node.
In accordance with the described techniques, a provider service receives a request log of a service endpoint including requests sent to the service endpoint by the provider service. The provider service extracts request log data from the request log including a failure count and an average response duration for the requests. The failure count includes the requests that have failed due to the provider service sending too many requests to the service endpoint and the requests that have failed due to server-side errors of the service endpoint. A throughput capacity for the service endpoint is predicted using a machine learning model based on the failure count and the average response duration. Then, the provider service adjusts a rate limiting threshold for the service endpoint based on the throughput capacity, and the rate limiting threshold defines a rate at which the provider service sends requests to the service endpoint.
Technologies are shown for selecting a provider to service a client service request using a consensus protocol and creating a block on a blockchain to service the client service request. In accordance with some aspects, a first miner receives parameters of each proposal transaction from a plurality of proposal transactions for servicing a client service request. The parameters of at least one proposal transaction from the plurality of proposal transactions is received from a second miner. The first miner uses a selection algorithm to select a first proposal transaction from the plurality of proposal transactions based on the parameters of each proposal transaction. The first miner appends a block to a blockchain based on the first proposal transaction.
Various examples described herein support or provide operations including providing a prompt to a large language model (LLM) for generating reward functions. The prompt can include a set of instructions for generating a set of reward functions associated with training a reinforcement learning (RL) agent to predict an objective. The set of reward functions is obtained from the LLM and used to train one or more instances of an RL agent to predict the objective. A score representing accuracy of the predicted objective for the one or more instances of the RL agent is generated and an individual instance of the one or more instances of the RL agent is selected to predict the objective based on the generated score.
A search engine leverages a neural translation model to provide diverse and multiple query reformulations. Two decoders are injected and a diversity inducing optimization function is introduced. After a query input is received from a user, a number of items are retrieved from a database in response to the query input. In response to a determination the query input is a null and low query based on a number of responsive items, a plurality of decoders is injected and a diversity inducing optimization function is leveraged to generate a plurality of diverse reformulated queries. A plurality of query results corresponding to the plurality of diverse reformulated queries is provided as output.
Various methods and systems for providing customized-item-specific interfaces of items on extended interface devices in a search system. A plurality of items—each having an extended interface configuration that indicates customized-item-specific interface instructions for extended presentation of items on extended interface devices—are accessed at a primary interface device associated with a set of extended interface devices including a second-user extended interface device that supports a social mode. Based on the first extended interface configuration of a first item, a determination is made that the first item is extendable to a first extended interface device in the set of extended interface devices. Based on determining that the first item is extendable, automatically causing generation of a customized-item-specific interface of the first item on the first extended interface device. Causing generation of the customized-item-specific interface can be based on communicating item extended interface data to the first extended interface device.
An incident alert associated with a deviation in a metric is received. A plurality of correlated metrics is identified based on contextual data. A plurality of labels for the plurality of correlated metrics is generated in accordance with a plurality of categories of signals. A causal graph is constructed based on the plurality of labels and a topology graph of the plurality of correlated metrics. A root cause of an incident associated with the incident alert is determined using a link analysis algorithm.
In various example embodiments, a system and method for constructing and scoring word vectors between natural language words and generating output to a user in the form of personalized recommendations are presented.
Interactive Augmented Reality Assistants are described. An example computing device captures image data of a physical environment for displaying an augmented reality (AR) of the physical environment on a display as an AR environment. The computing device identifies one or more physical items in the physical environment based at least in part on the image data. The computing device communicates with one or more listing servers over a network to identify one or more listed items related to the one or more physical items. The computing device renders a virtual object in the AR environment as an AR assistant for accessing the one or more listing servers. The computing device generates, in response to detecting a user interaction with the AR assistant in the AR environment, an output using information associated with the one or more listed items.
In implementation of techniques for personalized module arrangement via machine learning, a system receives user interface modules and interaction data corresponding to one or more interaction sessions. Based on the interaction data and the user interface modules, the system generates one or more user history representations via a machine learning model. The system generates, based on the one or more user history representations, interaction likelihood predictions via the machine learning model, wherein each interaction likelihood prediction corresponds to a likelihood of interaction with at least one of the user interface modules. Based on one or more interaction likelihood predictions above a predefined threshold value, the system generates an arrangement of the user interface modules. The system broadcasts the arrangement of the user interface modules for display.
A personalized delivery estimate system is described. A commercial transaction is generated between a seller and a buyer for an item in an online marketplace. Historical transactions of buyers and sellers in the online marketplace are stored in a storage device. A personalized delivery time estimate is computed for the buyer of the commercial transaction using seller information, buyer information, and item information with the historical transactions of buyers and sellers in the online marketplace.
09 - Appareils et instruments scientifiques et électriques
35 - Publicité; Affaires commerciales
41 - Éducation, divertissements, activités sportives et culturelles
42 - Services scientifiques, technologiques et industriels, recherche et conception
Produits et services
Software; e-commerce software; computer e-commerce software to allow users to conduct electronic business transactions in online marketplace websites via a global computer network; data and file management and database software; computer database software featuring information in the field of hobbies, collectibles and a wide variety of consumer products of others; computer software and software development tools for use in developing further software and software applications in the field of e-commerce; software for processing electronic payments to and from others; computer software for the identification and authentication of persons. Online trading services, namely, operating online marketplaces for sellers and buyers of goods and services; online trading services in which sellers post products and services to be auctioned or sold and bidding and purchasing is done via the Internet; providing business information to buyers about sellers and their goods and services for the purpose of facilitating transactions between buyers and sellers; evaluation relating to commercial matters; 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; data processing verification; 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; administrative services relating to intellectual property claims management, namely, processing and administration of claims of intellectual property owners against third party sellers. Entertainment services; entertainment services in the nature of fashion shows; educational services relating to fashion, design, archival history, entertainment and cultural activities; live performance services; organisation and presentation of conferences, exhibitions, competitions and fashion shows; arranging and conducting of workshops; conducting courses, seminars and workshops; audio, video and multimedia production, and photography; provision of non-downloadable electronic publications; digital video, audio and multimedia entertainment publishing services; publication of multimedia material online; production of audio/visual presentations; presentation of live performances; provision of entertainment facilities; provision of online information relating to audio and visual media; conducting and organisation of cultural activities; conducting and organisation of cultural activities in the fields of style and fashion; organising events for cultural purposes; arranging and organising award ceremonies; organising and presenting displays of entertainment relating to style and fashion; organisation of shows; organisation and presentation of fashion shows; ticket reservation and booking services for education, entertainment and sports activities and events; education services relating to modelling; publishing of reviews; education, information and advice on fashion, clothing and trends in the retail sector. Software development, programming and implementation; software as a service [SaaS]; design and development of computer software, mobile application software, and application programming interfaces software; providing temporary use of on-line, non-downloadable computer software and software development tools in the field of e-commerce; providing temporary use of online, non-downloadable e-commerce software to allow users to conduct electronic business transactions in online marketplaces via a global computer network; Maintenance and updating of computer software; development, hosting and maintaining of websites that give users the ability to create customized web pages featuring user-defined profiles in the nature of information in the field of intellectual property rights and intellectual property enforcement policies in an online marketplace; providing temporary use of on-line non-downloadable software for processing electronic payments; Providing user authentication services using biometric hardware and software technology for e-commerce transactions; 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; providing online, non-downloadable computer software for the identification and authentication of persons.
09 - Appareils et instruments scientifiques et électriques
35 - Publicité; Affaires commerciales
41 - Éducation, divertissements, activités sportives et culturelles
42 - Services scientifiques, technologiques et industriels, recherche et conception
Produits et services
Software; E-commerce software; Computer e-commerce software to allow users to conduct electronic business transactions in online marketplace websites via a global computer network; Data and file management and database software; Computer database software featuring information in the field of hobbies, collectibles and a wide variety of consumer products of others; Computer software and software development tools for use in developing further software and software applications in the field of e-commerce; Software for processing electronic payments to and from others; Computer software for the identification and authentication of persons Online trading services, namely, operating online marketplaces for sellers and buyers of goods and services; Online trading services in which sellers post products and services to be auctioned or sold and bidding and purchasing is done via the Internet; Providing business information to buyers about sellers and their goods and services for the purpose of facilitating transactions between buyers and sellers; Evaluation relating to commercial matters; 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; Data processing verification; 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; Administrative services relating to intellectual property claims management, namely, processing and administration of claims of intellectual property owners against third party sellers Entertainment services; Entertainment services in the nature of fashion shows; Educational services relating to fashion, design, archival history, entertainment and cultural activities; Live performance services; Organisation and presentation of conferences, exhibitions, competitions and fashion shows; Arranging and conducting of workshops; Conducting courses, seminars and workshops; Audio, video and multimedia production, and photography; Provision of non-downloadable electronic publications; Digital video, audio and multimedia entertainment publishing services; Publication of multimedia material online; Production of audio/visual presentations; Presentation of live performances; Provision of entertainment facilities; Provision of online information relating to audio and visual media; Conducting and organisation of cultural activities; Conducting and organisation of cultural activities in the fields of style and fashion; Organising events for cultural purposes; Arranging and organising award ceremonies; Organising and presenting displays of entertainment relating to style and fashion; Organisation of shows; Organisation and presentation of fashion shows; Ticket reservation and booking services for education, entertainment and sports activities and events; Education services relating to modelling; Publishing of reviews; Education, information and advice on fashion, clothing and trends in the retail sector Software development, programming and implementation; Software as a service [SaaS]; Design and development of computer software, mobile application software, and application programming interfaces software; Providing temporary use of on-line, non-downloadable computer software and software development tools in the field of e-commerce; Providing temporary use of online, non-downloadable e-commerce software to allow users to conduct electronic business transactions in online marketplaces via a global computer network; Maintenance and updating of computer software; Development, hosting and maintaining of websites that give users the ability to create customized web pages featuring user-defined profiles in the nature of information in the field of intellectual property rights and intellectual property enforcement policies in an online marketplace; Providing temporary use of on-line non-downloadable software for processing electronic payments; Providing user authentication services using biometric hardware and software technology for e-commerce transactions; 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; Providing online, non-downloadable computer software for the identification and authentication of persons
36.
MINIMIZING LARGE LANGUAGE MODEL HALLUCINATIONS IN GENERATED SUMMARIES
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.
Systems and methods are directed to detecting data anomalies. A data analysis system accesses data generated on a data platform. The data analysis system then analyzes the data to detect one or more data anomalies. The analyzing includes generating an optimal coordinate system without reducing a number of dimensions using principal component analysis (PCA), transforming the data into the optimal coordinate system without reducing the number of dimensions, and applying a sigma rule to the transformed data on the optimal coordinate system. The sigma rule can be the 3-sigma rule. In some cases, the data analysis system generates and transmits a notification or alert to a user or downstream component regarding the one or more data anomalies. In some cases, the data analysis system removes the one or more data anomalies to derive updated data and can provide the updated data to downstream systems for use.
A categorization method is provided. Training data is received that relates to assigning a first category to a first item using first item data. The first category is based on a first node and a first sub-node associated with the first node. A machine learning model is trained to categorize the first item using the first item data. Second item data associated with a second item is received. A second category is assigned to the second item with the second item data using a machine learning model trained with the training data. The second category is based on a second node and a second sub-node associated with the second node. Additional training data is iteratively received over time to train the machine learning model over time. The additional training data relates to assigning a third category to a third item using third item data associated with the third item.
A dataset is received that includes items listed via a listing platform, titles of the items, and key phrases. Each of the items are paired with one or more of the key phrases in the dataset. A data structure is constructed that maps tokens of the titles to the items associated with the titles, and maps the items to the key phrases that are paired with the items in the dataset. A seed title of a seed item is received as listed via the listing platform, and the seed title includes seed tokens. One or more similar items to the seed item are identified based on occurrence counts of the one or more seed tokens that map to the one or more similar items in the data structure. At least one recommended key phrase is output that maps to the one or more similar items in the data structure.
Key phrase generation using indefinite sequence learning is described. In accordance with the described techniques, a sequence generation model generates a sequence of key phrases based on an input document. During the generation task, the sequence generation model omits use of a self-generated sequence termination token. Key phrases in the sequence are then output as recommended key phrases for the input document.
A test-oriented architectural framework for determining application test results using adaptive objects is described. A data structure for data communications associated with a system-under-test (SUT) is generated for the test-oriented architectural framework. In a test framework associated with the SUT, services that communicate with the SUT using objects are identified. Each object includes at least one field and data items that correspond to the fields. An object is communicated between the SUT and the service associated with the SUT. Based on this object, the test framework generates a structure that maps the set of fields in the object to the data types of the data items therein. Additional objects communicated between the SUT and the service are then automatically regression tested using the structure.
Some aspects relate to technologies for performing item retrieval on a listing platform using a two tower model that leverages sequential user behavior information and co-action items. An item tower of the model generates an item representation embedding of target items by leverage information regarding co-action items for each target item. A user tower of the model generates item interest embeddings for users based on sequential behavior information for each user. Item retrieval can be performed for a given user by accessing an item interest embedding for the user, determining similarity measures for items on a listing platform based on the item interest embedding and item representation embeddings the items, and providing an indication of one or more items based on the similarity measures.
Some aspects relate to technologies for employing metadata identifiers for part item listings on a listing platform to generate inventory intelligence. In accordance with some aspects, textual analysis of listing data for each of a plurality of part item listings on a listing platform is performed to assign a metadata identifier to each part item listing. The metadata identifier for each part item listing comprises a combination of an inventory segment identifier, a category identifier, and fitment data. For a first metadata identifier, a set of part item listings having the first metadata identifier is identified. Demand data and/or supply data is aggregated for the set of part item listings to provide aggregated data. A user interface with one or more metrics is generated using the aggregated data, and the user interface is communicated over a network to a computing device for presentation.
Methods for enhancing or automating a review process of annotation tags for a set of tokens is described. A system may receive a list of tokens with associated tags for each token for a data set and may output any identified inconsistencies where a token is assigned at least two different tags. For example, instead of a human looking at each token individually or taking a sample set of the tags for review, the described techniques may look at all tokens with the associated tags in a set of data and may leverage reorganizing the tokens and associated tags to highlight errors to be fixed. Accordingly, the system may look across all tokens within an entire data set, while a review (e.g., by a human) of possible errors of the data set is limited to the highlighted errors flagged by the system.
Some aspects relate to technologies for probabilistic account linking, for instance, to perform fraud detection on online transaction platforms. In accordance with some configurations, linking strategies are defined for linking accounts based on account attributes. An average linking probability is generated for each linking strategy using account data for accounts on an online transaction platform, and the average linking probabilities are stored. To determine whether to link two accounts, linking strategies shared by the two accounts are identified, an account linking probability for the two accounts is generated using the average linking probabilities for the linking strategies shared by the two accounts, and the account linking probability is compared against a threshold. If the account linking probability satisfies the threshold, the accounts are linked and an action is taken based on the account linking.
G06Q 40/02 - Opérations bancaires, p. ex. calcul d'intérêts ou tenue de compte
G06Q 20/40 - Autorisation, p. ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasinExamen et approbation des payeurs, p. ex. contrôle des lignes de crédit ou des listes négatives
A system for creating generated descriptive text is provided. A prompt having first facts for an item is received and parsed to extract a first fact in a format. Second facts are generated where the first fact and the second facts are output in the format. A search query is generated that includes the first fact and the second facts and then a search is conducted using the search query. An output is generated based on the results. The output includes a suggested description of the item using at least one first fact of the first facts and the second facts. The output also has a summarization of the plurality of first facts and the second facts along with differences between the first facts and the second facts. A distribution of the plurality of first facts and the second facts in the results is provided in the output.
G06F 16/215 - Amélioration de la qualité des donnéesNettoyage des données, p. ex. déduplication, suppression des entrées non valides ou correction des erreurs typographiques
G06F 16/248 - Présentation des résultats de requêtes
G06F 16/25 - Systèmes d’intégration ou d’interfaçage impliquant les systèmes de gestion de bases de données
Smart merchandise branding is provided by automatically activating and populating smart strips in a store management platform. Initially, active sellers in the store management platform are identified. A seller of the active sellers is determined to be eligible for smart strips (e.g., a designated portion of the electronic storefront corresponding to the seller that highlights particular items the seller has listed for sale). The smart strips are automatically populated on the electronic storefront with data. In various aspects, the smart strips correspond to auctions ending soon, featured categories, and/or newly listed items.
In implementations of systems and procedures for item condition verification, a computing device implements an item condition verification system to compare a marketed condition of an item from an item listing with a delivered condition of the item using one or more machine learning models. Based on the comparison, the item condition verification system outputs a result indicating whether the item is significantly not as described by the item listing.
A system and method for optimizing resource allocation in a real-time online auction system of a publication application is described. The method includes receiving campaign data including a total budget, target resource utilization curve, and maximum bid for each auction opportunity of the publication application, maintaining, in a memory, a dynamic adjustment factor for each campaign, applying a resource conservation algorithm by calculating an adjusted bid using the dynamic adjustment factor, tracking, in real-time, resource utilization for each campaign for the publication application, updating the dynamic adjustment factor based on a difference between target and actual resource utilization, to reduce computational load through adaptive bid adjustments, and outputting, to a network interface, the updated dynamic adjustment factor and the adaptive bid adjustments for use in subsequent auctions, to balance resource utilizations across multiple time periods.
Linked packet tracing techniques for software load balancers are described, as leveraging a client identifier generated to link packet traces between packets of a service provider system. In one example, a client-side packet is received from a client device by a processing system. From this, a client identifier is generated by the processing system based on the client device. In response, a modified client-side packet is generated by the processing system by modifying the client-side packet to include the client identifier. The modified client-side packet is transmitted by the processing system to a server system. A server-side packet is then received by the processing system from the server system. In response, the server-side packet is detected as including the client-identifier. From this, the server-side packet is linked to at least the modified client-side packet based on the detecting, and the processing system outputs a result of the linked packets.
H04L 47/125 - Prévention de la congestionRécupération de la congestion en équilibrant la charge, p. ex. par ingénierie de trafic
H04L 47/193 - Commande de fluxCommande de la congestion au niveau des couches au-dessus de la couche réseau au niveau de la couche de transport, p. ex. liée à TCP
H04L 47/2408 - Trafic caractérisé par des attributs spécifiques, p. ex. la priorité ou QoS pour la prise en charge de différents services, p. ex. services du type services différentiés [DiffServ]
H04L 47/2483 - Trafic caractérisé par des attributs spécifiques, p. ex. la priorité ou QoS en impliquant l’identification des flux individuels
51.
Detecting Compatibility Mismatch by Generative Artificial Intelligence
Detecting compatibility mismatch by generative artificial intelligence is described. Compatibility data is obtained (e.g., by accessing a database). The compatibility data is associated with a compatibility between items (e.g., items and categories of vehicles or an item and another item) and includes a list of recommended compatibilities between the items and a user reported compatibility for at least one item. A machine learning model is generated for detecting a compatibility mismatch between a first item and a second item and/or between an item and a category of vehicle. At least a portion of the compatibility data is provided as input to generative artificial intelligence to generate the machine learning model. An update to the list of recommended compatibilities is determined based on the detected compatibility mismatch.
Some aspects relate to technologies for performing fraud detection on online transaction platforms through user behavior data. In accordance with some configurations, a multimodal model is used to predict whether user behavior data is indicative of fraudulent activity. At least two different types of user behavior data on the online transaction platform are accessed and provided as input to the multimodal model, which generates a fraud prediction output based on the user behavior data. Based on the fraud prediction output, one or more actions are taken for the online transaction platform. In some aspects, a sequence of patch indices based on mouse movement data is used for fraud prediction. In particular, a sequence of x,y positions of a cursor on a screen is obtained, the screen is divided into a grid of patches, and the sequence of x, y positions is converted to a sequence of patch indices.
G06N 20/20 - Techniques d’ensemble en apprentissage automatique
G06Q 20/40 - Autorisation, p. ex. identification du payeur ou du bénéficiaire, vérification des références du client ou du magasinExamen et approbation des payeurs, p. ex. contrôle des lignes de crédit ou des listes négatives
G06F 21/56 - Détection ou gestion de programmes malveillants, p. ex. dispositions anti-virus
A real-time augmented reality item guide is described. A service platform may receive a digital image depicting at least one item. A plurality of listings for the at least one item on the service platform is identified via at least one machine learning model of the service platform based on the digital image, each of the plurality of listings associated with a respective point in time. An aspect is fetched from each of the plurality of listings for the at least one item on the service platform. Time-dependent augmented reality digital content is generated for the at least one item based on the aspect. The time-dependent augmented reality digital content is rendered as an overlay on the digital image.
G06V 10/44 - Extraction de caractéristiques locales par analyse des parties du motif, p. ex. par détection d’arêtes, de contours, de boucles, d’angles, de barres ou d’intersectionsAnalyse de connectivité, p. ex. de composantes connectées
G06V 10/74 - Appariement de motifs d’image ou de vidéoMesures de proximité dans les espaces de caractéristiques
G06V 10/764 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant la classification, p. ex. des objets vidéo
Some aspects relate to technologies for leveraging model output from a generative model to perform item retrieval on a listing platform. In some examples, input for item retrieval is provided to a generative model to produce a model output. A lookup is performed on a key-value store using the model output. If a matching query is found in the key-value store, a query embedding corresponding to the matching query is returned. If a matching query is not found, a query embedding is obtained by generating the query embedding from the model output or using the model output to query a known query index for a known query, which is used to lookup a query embedding in the key-value store. One or more item embeddings are identified based on the query embedding. An output is provided identifying one or more item listings corresponding to the one or more item embeddings.
09 - Appareils et instruments scientifiques et électriques
35 - Publicité; Affaires commerciales
42 - Services scientifiques, technologiques et industriels, recherche et conception
Produits et services
Downloadable and recorded computer e-commerce software to
allow users to conduct electronic business transactions in
online marketplace websites via a global computer network;
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 and downloadable
and recorded software development tools for use in
developing further software and software applications in the
field of e-commerce; 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; downloadable and
recorded computer software for authenticating user
identification in the field of e-commerce transactions and
online shopping websites. Online trading services, namely, operating online
marketplaces for sellers and buyers of goods and services;
online trading services in which sellers post products and
services to be auctioned or sold and bidding and purchasing
is done via the internet; infomediary services, namely,
facilitating transactions between buyers and sellers through
providing buyers with information about sellers, goods, and
services; providing consumer product and services
information in the nature of buyer feedback, seller ratings
and reviews, and product evaluations and reviews for buyers
and sellers via a searchable online evaluation database;
customer service management, namely, confirming authenticity
of consumer products, product manufacturers, and sellers of
goods for the purposes of helping consumers make informed
purchasing decisions; business assistance services in the
field of intellectual property claims management, namely,
processing and administration of claims of intellectual
property owners against third party sellers. Design and development of computer software, mobile
application software, and application programming interfaces
software; providing temporary use of on-line,
non-downloadable computer software and software development
tools for use in developing further software and software
applications in the field of e-commerce; providing temporary
use of online, non-downloadable e-commerce software to allow
users to conduct electronic business transactions in online
marketplaces via a global computer network; maintenance and
updating of computer software for others; hosting a website
that gives users the ability to create customized web pages
featuring user-defined profiles in the nature of information
in the field of intellectual property rights and
intellectual property enforcement policies, in order to
assist program participants with inquiries and requests
regarding use of intellectual property by others in an
online marketplace; providing temporary use of on-line
non-downloadable software for processing electronic payments
in the nature of credit card, debit card, cash card, gift
card, wire transfer, and wallet payments; providing
temporary use of on-line non-downloadable authentication
software for authenticating user identification in the field
of e-commerce transactions and online shopping websites;
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; and providing online, non-downloadable
computer software for authenticating user identification in
the field of e-commerce transactions and online shopping
websites.
56.
MACHINE LEARNING MODEL TRAINING ON RISK PREDICTION USING GRAPH KNOWLEDGE DISTILLATION
Various embodiments described herein support or provide operations including identifying a machine-learning (ML) model associated with an omni-view knowledge graph; generating an embedding vector that represents the omni-view knowledge graph; identifying a ML model associated with a temporal-view knowledge graph; generating an embedding vector that represents the temporal-view knowledge graph; and training a ML model based on the generated embedding vectors.
Systems and methods for organizing structured items in an e-commerce system are presented. In some aspects, the systems and methods may include a back-end structured item organizing system associated with a front-end category mapping system which is configured to process use case requests. In some aspects, a use case request may identify an item for classification by the back-end structured organizing system. In some aspects, the described systems and methods may select a listing bucket from a set of listing buckets of the back-end structured organizing system based on attribute values indicated in the use case request, and may cause a user interface of the front-end category mapping system to display one or more characteristics associated with items categorized within the selected listing bucket.
Multiple agent root cause analysis techniques are described. In an implementation, a processor performs operations to troubleshoot performance of a system operation by a plurality of domains (e.g., of a distributed system). The processor executes a system artificial intelligence (AI) agent to determine which domains of the plurality of domains include domain functions in support of the system operation, and to generate, using machine learning, queries to the determined domains based on the system operation. Domain responses are received from domain AI agents associated with the determined domains responsive to the queries and generated based on domain data associated with respective domains. A system response is generated by the system AI agent using machine learning based on the domain responses.
H04L 41/0631 - Gestion des fautes, des événements, des alarmes ou des notifications en utilisant l’analyse des causes profondesGestion des fautes, des événements, des alarmes ou des notifications en utilisant l’analyse de la corrélation entre les notifications, les alarmes ou les événements en fonction de critères de décision, p. ex. la hiérarchie ou l’analyse temporelle ou arborescente
H04L 41/16 - Dispositions pour la maintenance, l’administration ou la gestion des réseaux de commutation de données, p. ex. des réseaux de commutation de paquets en utilisant l'apprentissage automatique ou l'intelligence artificielle
59.
REAL-TIME INVENTORY MAPPING AND NOTIFICATIONS FOR SAVED SEARCHES
A hierarchical data structure is leveraged to provide real-time inventory mapping and notification. Behavioral data is received from a user indicating an interest in a category or in a seller. The behavioral data may correspond to the user saving a search comprising one or more keywords or to the user following the seller. The user is dynamically associated with the interest in the hierarchical data structure. Upon detecting a configurable trigger, item data corresponding to an item in an inventory of items is retrieved. The configurable trigger may be the item corresponding to the category or seller being listed in an inventory, an incentive being provided for the item by the seller, or the item being identified as a fit for another item associated with the user. In aspects, the item data is communicated to the user in real-time.
The technology described herein relates to systems, methods, and computer storage media, among other things, for generating recommendations corresponding to whether requests (e.g., transmitted by computing devices) are malicious network traffic. For example, the recommendations can be generated using various historical and current network traffic trends (e.g., associated with a particular application programming interface). As another example, the recommendations can be generated based on particular models. Based on the generated recommendation, dynamic rate limiting rules can be applied for determining whether a request is malicious network traffic. Based on determining the request is malicious network traffic, the request (and additional requests associated with that particular request) can be blocked.
H04L 47/25 - Commande de fluxCommande de la congestion le débit étant modifié par la source lors de la détection d'un changement des conditions du réseau
Systems and methods are directed to optimizing partitioning of Hive tables. The system aggregates query data associated with a file system and analyzes data partitioning and query patterns derived from the aggregated query data. The system then generates one or more merge rules for merging partitions of at least some data types that do not satisfy a query threshold based on the query patterns. The one or more merge rules are registered into a Hive metastore such that the one or more rules can be accessed during a write or read process. Based on the one or more rules, the system can then merge partitions of the at least some data types that do not satisfy the query threshold into a single physical partition of the file system.
G06F 16/27 - Réplication, distribution ou synchronisation de données entre bases de données ou dans un système de bases de données distribuéesArchitectures de systèmes de bases de données distribuées à cet effet
62.
Autonomous Technician System For Component Handling and Installation
Autonomous technician system for component handling and installation is described. In one or more implementations, an autonomous technician comprises one or more robots to perform operations including component condition verification, component transport, and component installation. The one or more robots verify compatibility of the component with a different apparatus, where the component is listed by a seller in an item listing and is purchased by a buyer of the apparatus to which the component is installed by the one or more robots. In variations, the one or more robots transport the component from the seller to a location of the apparatus of the buyer. The one or more robots further utilize various installation protocols based on the condition of the component and a condition of the apparatus.
Systems and methods for managing keyword predictions for proposed titles are provided. In example embodiments, a network system receives, from a user during a publication creation process, a proposed title for a publication associated with an item. The proposed title includes a plurality of tokens, whereby the plurality of tokens comprises at least all non-stock words in the proposed title. Based on the proposed title, the network system identifies an importance of each token of the plurality of tokens in the proposed title. The network system then causes presentation of a user interface that visually indicates the importance of each token of the plurality of tokens in the proposed title.
A system automatically maintains a plurality of client connections associated with a plurality of clients, the plurality of client connections including active and idle connections. A first server receives a request from a client of the plurality of clients to access a second server of a plurality of second servers communicatively coupled to the first server, the plurality of second servers having varying communication protocols. The first server then identifies a first communication protocol associated with the second server and activating a link between the first server and the second server using the first communication protocol.
H04L 67/1001 - Protocoles dans lesquels une application est distribuée parmi les nœuds du réseau pour accéder à un serveur parmi une pluralité de serveurs répliqués
H04L 67/1008 - Sélection du serveur pour la répartition de charge basée sur les paramètres des serveurs, p. ex. la mémoire disponible ou la charge de travail
H04L 67/1021 - Sélection du serveur pour la répartition de charge basée sur la localisation du client ou du serveur
H04L 67/56 - Approvisionnement des services mandataires
H04L 67/564 - Amélioration de la commande des applications basée sur des données interceptées des applications
H04L 69/14 - Protocoles multicanaux ou multi-liaisons
In accordance with techniques for controlling hallucinations in generated images, a generative image model receives an input image depicting an object and having a first background, and the generative image model produces a generated image depicting the object by replacing the first background with a second background. Further, a salient object detection model generates a first object mask and a second object mask. The first object mask defines a first positioning of the object within the input image, while the second object mask defines a second positioning of the object within the generated image. A hallucination metric capturing an amount of deformation introduced into the object by the generative image model is determined based on a comparison of the first object mask and the second object mask. In one or more implementations, the generated image is output based on the hallucination metric meeting a threshold.
G06V 10/56 - Extraction de caractéristiques d’images ou de vidéos relative à la couleur
G06V 10/764 - Dispositions pour la reconnaissance ou la compréhension d’images ou de vidéos utilisant la reconnaissance de formes ou l’apprentissage automatique utilisant la classification, p. ex. des objets vidéo
H04N 5/272 - Moyens pour insérer une image de premier plan dans une image d'arrière plan, c.-à-d. incrustation, effet inverse
66.
AUTOMATIC METHOD TO DETERMINE THE AUTHENTICITY OF A PRODUCT
A method that comprises receiving at a network connected server from a first client terminal, a message comprising, an user application ID of a user selecting a media object using a user interface presented on a display of the first client terminal and the media object, generating a web document which presents a browser user interface and the media object when accessed by a browser, the web document having a network accessible storage address, sending the network accessible storage address from the network connected server to allow a browser installed in a second client terminal to use of the network accessible storage address to display the media object the browser user interface, identifying a usage of the browser user interface for inputting a reaction to the media object by a user of the second client terminal, and forwarding the reaction to the first client terminal using the sender user ID.
Interactive forums for specific regions of interest in a digital image are described. An example computing device receives data associated with a digital image of an item from a server. The data indicates a tagged region of the digital image and a plurality of comments associated with the tagged region. The plurality of comments is submitted by a plurality of users at a plurality of client devices. The computing device displays the digital image of the item. The computing device detects input selecting the tagged region of the digital image. The computing device displays, in response to detecting the input selecting the tagged region, a user interface that includes an aggregated view of the plurality of comments and overlays at least part of the user interface on at least part of the digital image outside the tagged region.
A system comprising a computer-readable storage medium storing at least one program and a computer-implemented method for providing personalized content sharing service is presented. Consistent with some embodiments, the method may include receiving a request to share a content item with a member of the social network. The method may further include determining an interest of the member of the social network based on social data about the member of the social network. A customized posting to share the content item with the member of the social network is then generated using the interest of the user.
G06Q 50/00 - Technologies de l’information et de la communication [TIC] spécialement adaptées à la mise en œuvre des procédés d’affaires d’un secteur particulier d’activité économique, p. ex. aux services d’utilité publique ou au tourisme
H04L 65/1063 - Serveurs d'applications fournissant des services réseau
H04L 65/612 - Diffusion en flux de paquets multimédias pour la prise en charge des services de diffusion par flux unidirectionnel, p. ex. radio sur Internet pour monodiffusion [unicast]
H04L 65/75 - Gestion des paquets du réseau multimédia
H04L 69/08 - Protocoles d’interopérabilitéConversion de protocole
A recommendation system leverages multi-target search to provide item listing recommendations and/or query suggestions. For a given input image with multiple objects, multi-target search uses object detection to detect each object, and stores complementary object data associating each object from the image. Additionally, a search of an item listing datastore is performed using each object from the image as a search query. Based on item listings returned as search results, complementary item listings data associating item listings is stored. In some configurations, the complementary item listings data is also used to train a machine learning model to predict complementary item listings for a given item listing. When an input item listing is received, item listing recommendations and/or query suggestions are determined for the input item listing using the complementary object data, the complementary item listing data, and/or the machine learning model.
A machine may be configured to perform A/B testing on mobile applications. For example, the machine receives an identifier of a user from a mobile device that requests a layout description of a user interface, determines the layout description that provides a configuration of one or more elements of the user interface according to an experiment variant associated with an experiment, and causes display of the one or more elements of the user interface on the mobile device based on the layout description.
G06F 9/451 - Dispositions d’exécution pour interfaces utilisateur
H04W 4/18 - Conversion de format ou de contenu d'informations, p. ex. adaptation, par le réseau, des informations reçues ou transmises pour une distribution sans fil aux utilisateurs ou aux terminaux
Generating an access point certificate based on a graph that defines relationships between an access point and at least one domain is described. A computing device may generate a graph that defines relationships between a domain and multiple different access points including a serving access point and one or more fallback access points. A device may generate a certificate for accessing data via the domain based on the relationships between the domain and the one or more fallback access points. A device may control access to the data using the certificate. The graph may define a relationship between a network address, an access point, and one or more records associated with at least one domain. Further, the graph may define a first relationship between a network address and a record associated with a domain and a second relationship between the network address and an alias record associated with the domain.
The technology described herein relates to systems, methods, and computer storage media, among other things, for providing search query intent specificity. Embodiments may include identifying a search query performed using a search engine and generating a query vector for the search query by aggregating search result embeddings (e.g., item listing vectors) of search results from the search query. Further, in some embodiments, similarities (e.g., cosine similarities) between the query vector and the item listing vectors can be determined. As such, an intent specificity of the search query can be determined. Further, in some embodiments, the intent specificity can be used to train an intent specificity machine learning model for generating intent specificity scores for other search queries. Based on the intent specificity scores determined using the one or more trained intent specificity machine learning models, determinations can be made with respect to precision and recall, etc.
Artificial Intelligence profile generation techniques are described. In an implementation, a profile manager service is configured to generate a profile, automatically and without user intervention. To do so, the profile manager service utilized a prompt generator module and a profile generation module. The prompt generator module is implemented by a machine-learning model using generative artificial intelligence to generate prompts to guide input of characteristics to be used as part of profile generation. The profile generator module is implemented by a machine-learning model using generative artificial intelligence. The profile generator module is configured to generate a candidate profile based on the characteristics received over a series of iterations in response to the prompts generated by the prompt generator module.
Some aspects relate to technologies for generating user segments for users of a listing platform using a neural network trained using contrastive learning augmented by user-user scores. In accordance with some aspects, user vectors are generated for users of a listing platform based on user data tracked by the listing platform. Additionally, user-user scores are generated for pairs of users based on user interactions with items on the listing platform. A neural network is trained using the user vectors and the user-user scores as training data. Once trained, the neural network is used to generate user embeddings by providing user vectors as input to the neural network. User segmentation is performed by clustering user embeddings into clusters with each cluster providing a user segment.
Taxonomy-based image generation is used for item searching, and enhances the quality and personalization of search results. Prior interacted items are classified into a categorical taxonomy. A generative AI model can be used to classify the prior interacted items, by generating categories or assigning to existing categories. A set of prior interacted items is selected from one of the categories and provided to an image model that generates a photo-realistic image in response. The photo-realistic image includes item renderings that are rendered illustrations of items. An item search using a search engine can be performed based on the generated photo-realistic image. For instance, the photo-realistic image or a portion thereof could be provided as a search query using an image-based search or described to perform a text-based search. Search results are identified for the search query and are provided in response.
G06F 16/9535 - Adaptation de la recherche basée sur les profils des utilisateurs et la personnalisation
G06F 16/583 - Recherche caractérisée par l’utilisation de métadonnées, p. ex. de métadonnées ne provenant pas du contenu ou de métadonnées générées manuellement utilisant des métadonnées provenant automatiquement du contenu
A mediator system, which serves as a conduit between a customer and merchants, includes customer profile data. The mediator system analyzes the customer profile data, and assigns an indication to a customer represented by the customer profile data. The mediator system then provides the indication to the merchants, and then receives bids from the merchants. The bids are for establishing a connection between the merchants and the customer. The mediator system provides to a portion of the merchants, based on the bids, a connection to the customer.
The technology described herein relates to systems, methods, and computer storage media, among other things, for providing an enhanced conceptualized search based on enriched categorizations of item listings. For example, the enhanced conceptualized search can include generating a search vector, using one or more generative artificial intelligence (AI) models, based on a received search query. Some embodiments of generating the search vector may include applying one or more generative AI models to one or more sets of prior user interactions. Based on the search vector, a particular item listing vector can be determined. The item listing vector can be generated by applying one or more generative AI models to each of an image of the item and a textual description of an item listing, and the item listing vector can be generated based on enhanced categories determined (e.g., using one or more generative AI models) for the item listing.
Visual search query intent extraction and search refinement is described. In one or more implementations, a visual search query system receives a search query for items listed on an online marketplace, the search query including an image. Using one or more machine learning models, the visual search query system analyzes the image to determine characteristics of an object in the image. Based on the characteristics of the object in the image, the visual search query system automatically generates one or more search terms and searches the online marketplace to locate items matching the one or more search terms. The visual search query system then displays visual indications of the located items matching the one or more search terms in a user interface of the online marketplace.
Message personalization for an electromechanical device is described. Data is received, the system information data associated with a change in state of one or more components of an electromechanical device. A message is generated by inputting the system information data as a prompt to generative artificial intelligence representative of one or more characteristics of the electromechanical device and trained to analyze the system information data. The message indicates feedback responsive to the change in state of the one or more components. The message is output, where one or more language characteristics of the message are associated with the one or more characteristics of the electromechanical device.
Methods, systems, and computer storage media for providing a knowledge graph using a generative AI knowledge graph (KG) engine “generative AI KG engine” in an item listing system. The generative AI KG engine supports generating the knowledge graph using a generative AI model (e.g., an LLM). In operation, a seed product is accessed in a product listing system. Using a product knowledge graph, a plurality of candidate products associated with the seed product are identified. The product knowledge graph comprises a plurality of products as nodes and a plurality of relationships as edges. The product knowledge graph is associated with a generative AI model. Using a ranker of the product listing system, a plurality of recommended products are identified. The plurality of recommended products are a subset of the plurality of candidate products. The plurality of recommended products are communicated and caused to be generated on a graphical user interface.
In accordance with the described techniques, a configuration update is received changing configuration parameters of a data stream processing application from a source configuration to a target configuration. In addition, the configuration update specifies incremental update parameters indicating how different partitions of the data stream processing application are to be incrementally updated. The computing device is configured to incrementally update the data stream processing application. During a time interval indicated by the incremental update parameters, for instance, the target configuration is applied to a first partition, and the source configuration is applied to a second partition. Thus, during the time interval, the first partition processes events of a data stream in accordance with the target configuration, and the second partition processes events of the data stream in accordance with the source configuration. After the time interval, the target configuration is applied to the second partition.
Generative artificial intelligence (AI) is leveraged to generate a maintenance schedule for a vehicle and identify a list of recommended products to perform tasks described by the maintenance schedule. Initially, information corresponding to a vehicle of a user is received. Based on the information, a generative AI model is utilized to generate a maintenance schedule corresponding to the vehicle. The user is provided the maintenance schedule and a list of recommended products to perform tasks described by the maintenance schedule. In some aspects, the generative AI model generates the list of recommended products. Additionally, the user may be enabled to purchase items from the list of recommended products via an electronic marketplace.
Similarity sensitive diversity is utilized to measure variation in a distribution of item listings along one or more categories. A cosine similarity between category vectors of each category pair in a set of categories is determined and utilized to generate a pairwise similarity matrix. The pairwise similarity matrix may be pruned to remove category pairs below a threshold. Utilizing the pairwise similarity matrix, similarity sensitive diversity between one or more items of a plurality of items may be determined. In various aspects, the similarity sensitive diversity may be utilized to: generate a list of relevant items in an appropriate distribution, suggest refinements of a search query; generate navigation modules; categorize or recategorize the plurality of items; or generate autosuggestions.
Systems and methods for conducting a test on a third-party testing platform are provided. A networked system causes presentation of a setup user interface to a third-party user, whereby the setup user interface includes a field for indicating an attribute of a publication to be tested. The networked system receives, via the setup user interface, an indication of the attribute, a subject to be tested, and one or more test parameters. The networked system applies the attribute change to a first version of the publication to generate a second version of the publication. The first version is presented to a first subset of potential users and the second version is presented to a second subset of potential users. Interactions with both the first version and the second version are monitored and analyzed to determine results of the test. The results are then presented to the third-party user.
G06Q 10/0637 - Gestion ou analyse stratégiques, p. ex. définition d’un objectif ou d’une cible pour une organisationPlanification des actions en fonction des objectifsAnalyse ou évaluation de l’efficacité des objectifs
G06Q 30/0242 - Détermination de l’efficacité des publicités
A query modality recommendation system provides recommendations to use a particular query modality based on one or more categories of search results for a search query. Upon receiving a search query in a first query modality at a search engine, the query modality recommendation system determines to recommend use of a second query modality based on one or more categories of the search results. For example, the first query modality may be a textual query and the second query modality may be an image query. In aspects, recommending use of the second query modality comprises comparing a first search performance of the one or more categories for the first query modality in historical search queries to a second search performance of the one or more categories for the second query modality in the historical search queries.
Systems and methods for public transport infrastructure facilitated drone delivery are provided. In example embodiments, a request to deliver a package to a drop-off destination using a drone is received. Public infrastructure information is accessed. A public infrastructure terminal from which the drone delivers the package is identified based on the public infrastructure information. An instruction is communicated to transport the package to the identified public. A drone delivery route from the identified public infrastructure terminal to the drop-off destination is determined based on the public infrastructure information. An instruction to deliver the package using the drone delivery route is communicated to the drone.
Some aspects relate to technologies for using a generative model to customize homepages. In accordance with some aspects, webpage code for a home page is accessed. Additionally, a prompt is received to modify the homepage to provide a customized homepage. A generative model is caused to generate modified webpage code using the prompt and the webpage code. The modified webpage code is transmitted, over a network to a client device, for rendering the customized homepage on the client device.
A method for suggesting feedback is provided. The method includes determining an interaction between a first and second users, generating a feedback response element, and providing an interactive user interface that lists the feedback response element and a selectable machine learning feedback response element. The method also includes receiving a selection of the selectable machine learning feedback response element and accessing a first feedback associated with the first user where the first feedback has characteristics unique to the first user. Moreover, the method includes automatically generating a second feedback using the first feedback, the second feedback incorporating the characteristics unique to the first user without input from the first user.
Methods, systems, and computer storage media for providing generative artificial intelligence (AI) recommendation management using an artificial intelligence system in an item listing system. A generative AI recommendation engine supports generative AI recommendation management based on a review-based recommendation platform including offline generative AI operations, review-based recommendation guides for items and a review-based recommendation logic. Using generative AI techniques, review-based recommendation guides are generated for a plurality of items. The review-based recommendation logic supports identifying review-based recommended items for users based on review data and the review-based recommendation guides. In operation, review data-associated with a user-for a first item, is accessed. Based on the review data, a review-based recommendation guide feature of the first item is identified. The review-based recommendation guide feature of the first item is mapped to a review-based recommendation guide feature of a second item. The second item is communicated as a review-based recommended item.
The technology disclosed herein relates to the generation and regeneration of three-dimensional environments for users to visualize particular items within the three-dimensional environment. For example, an indication for generating a three-dimensional environment associated with a search query can be received. Previous user interaction data associated with a user providing the indication can be identified and provided to a generative artificial intelligence model. The generative artificial intelligence model can be trained using prior user interaction data from a plurality of other users. In some embodiments, the generative artificial intelligence model can also be trained using a plurality of item features for an item within an item corpus to identify particular images of items associated with a particular style that corresponds to the search query and the previous user interaction data of the user. In embodiments, three-dimensional environments can be generated with items that are associated with the particular style.
In implementations of systems and procedures for automatic image composition for item promotions, a computing device implements the display of graphical promotion emblems associated with items listed for sale on an electronic commerce platform. An emblem is composited with a digital image of an item offered for sale, with the emblem overlapping the digital image. The emblem includes encoding that, when processed, populates a graphical promotion data field of the emblem with promotion data of the item. The display of the emblem is controlled based on the promotion data such that the emblem is displayed while the promotion is active and is hidden while the promotion is inactive.
Methods, systems, and computer storage media for providing generative AI presentation management using a generative AI presentation engine in an item listing system. A generative AI presentation engine supports generative AI presentation management based on a generative-AI-data presentation platform including presentation training operations and a presentation data structure for composite image data including images or text. The composite image data and presentation logic are generated using generative AI models. The presentation logic supports mapping and rotating composite image data on item listing system interfaces. In operation, a request associated with image data in an item listing system is accessed. Composite image data associated with a generative AI model and user data is accessed. The composite image data comprises a generative AI image element and a generative AI item listing interface element. The composite image data is communicated to an item listing system client, causing display of the composite image data.
A bad change detector for operational features of digital platforms is described. Metric data from a website is acquired by the bad change detector while implementing a first version of an operational feature of the website and while implementing a second version of the operational feature. A bad change to the website is detected by measuring an inequality among values of a frequency distribution defined by a time series of the metric data. The operational feature is reverted to the first version, automatically and without user intervention, in response to detecting the bad change, thereby improving operation of computing devices that implement the website.
Various embodiments described herein support or provide operations including receiving a natural language text from a device; using a first machine learning model to generate a first embedding vector that represents the natural language text; matching the first embedding vector with a second embedding vector that represents a data table; generating, based on the natural language text, a prompt as an input to a second machine learning model; using the second machine learning model to generate a response based on the prompt; and causing display of the SQL query in a user interface of the device.
Identity verification services and techniques are described. In one or more implementations, the identity verification service is employed to control resource access based on identity verification. In a first example, an action depicted as performed by a user in verification input data (e.g., head movement, gesture, and so on) is compared by the identity verification service to determine whether the action corresponds to an action specified by a selected verification prompt. In a second example, a likelihood is ascertained by the identity verification service as to whether the user depicted in the verification input data corresponds to a human being through use of a machine-learning model. In a third example, a depiction of the user's face as captured by the verification input data is compared with a digital image of the user maintained by a verification system. Resource access is controlled by the identity verification service.
Catalog-based item listing enhancement is described. A matching item from a collection of items may be selected to match an item listing by at least one of an aspect matching model trained using a catalog of the collection of items and a language model trained using a database of item listings. The item listing may be updated based on an entry for the matching item in the catalog.
Systems and methods are provided for automatically generating a thumbnail for a video on an online shopping site. The disclosed technology automatically generates a thumbnail for a video, where the thumbnail represents an item but not necessarily content of the video. A thumbnail generator receives a video that describes the item and an ordered list of item images associated with the item used in an item listing. The thumbnail generator extracts video frames from the video based on sampling rules and determines similarity scores for the sampled video frames. A similarity score indicates a degree of similarity between content of a video frame and an item image. The thumbnail generator determines weighted similarity scores based item images and occurrences of sampled video frames in the video. The disclosed technology generates a thumbnail for the video by selecting a sample video frame based on the weighted similarity scores.
G06F 16/783 - Recherche de données caractérisée par l’utilisation de métadonnées, p. ex. de métadonnées ne provenant pas du contenu ou de métadonnées générées manuellement utilisant des métadonnées provenant automatiquement du contenu
G06V 10/75 - Organisation de procédés de l’appariement, p. ex. comparaisons simultanées ou séquentielles des caractéristiques d’images ou de vidéosApproches-approximative-fine, p. ex. approches multi-échellesAppariement de motifs d’image ou de vidéoMesures de proximité dans les espaces de caractéristiques utilisant l’analyse de contexteSélection des dictionnaires
