Abstract: In an example, first entities are extracted from user profiles. Second entities are extracted from content information associated with content item. User-associated metrics associated with the first entities are determined based upon the user profiles and/or content events. First vector representations of the first entities and second vector representations of the second entities are processed to generate an attention distribution array. Each value of the attention distribution array represents, for a user interested in an entity of the first entities, a proportion of (i) entity-specific activity, of the user, related to an entity of the second entities relative to (ii) an entirety of activity of the user. An inferred activity distribution array is generated by applying the user-associated metrics to the attention distribution array. A filtered subset of activity distribution values is generated by pruning values from the inferred activity distribution array.

Abstract: Techniques for automatic intelligent information extraction from electronic messages are disclosed. In one embodiment, a computerized method is disclosed comprising obtaining a corpus of electronic messages, generating training data using the corpus of electronic messages, training an attribute generation model using the training data, analyzing an electronic message from a message folder and generating model input based on the analysis, obtaining model output from the attribute generation model based on the model input, the model output comprising, in connection with a respective type of information, a set of attribute values for a set of attributes corresponding to the respective type of information, and generating a presentation, for display at a user computing device, the presentation comprising information based at least in part on the set of attribute values associated with the set of attributes.

Abstract: The present teaching relates to method, system, medium, and implementations for characterizing data. A location feature is first received. A distance-aware embedding for the received location feature is obtained, where the distance-aware embedding for the location feature is learned based on distances between different pairs of locations. A representation of the location feature is then generated based on the embedding for location related predictions.

Abstract: The disclosed systems and methods provide a novel action prediction framework that performs personalized action prediction. According to an embodiment, the disclosed framework is able to dynamically predict which action (if any) a user might perform in response to receiving a given message. In some embodiments, for a given message, the action prediction framework can determine the probability that a user (e.g., sender, recipient) associated with the message may perform an action or set of action actions (e.g., open, forward, delete, reply, archive) related to the message. In some embodiments, the framework may be used to suggest a predicted action to the user. In some embodiments, a computing device may use the predicted actions to automatically perform the action. According to an embodiment, the action prediction framework includes a multi-label or multi-class model using a neural network.

Abstract: The present teaching relates to personalized content recommendation. A webpage is contrasted for a user having a plurality of slots each of which is to be allocated with a content item. For each of the plurality of slots, a plurality of content items in a plurality of types of content are accessed. For each of the plurality of types of content, a personalized score is predicted for each content item in the type of content, wherein the personalized score is obtained based on a trained model trained. A recommended content item of the type of content is selected based on personalized scores. An overall recommended content item is selected and allocated to a slot based on criteria associated with the personalized scores of the recommended content items and a business rule. The webpage with the plurality of slots allocated with content items is provided to the user.

Abstract: The disclosed systems and methods provide a novel framework that provides mechanisms for a Deep & Cross Network (DCN) framework that performs distilled deep prediction for personalized stream ranking on portal websites. The disclosed framework is scalable to satisfy the much more stringent latency and computational requirements required by current network operating environments. The disclosed framework is able to dynamically evaluate and leverage live traffic on network sites in order to provide, update and maintain current recommendations for users as they traverse to a portal and when they navigate within the portal. The disclosed framework implements a DCN model(s) that is capable of being compressed into a model size for a unified optimization within a live traffic environment by combining knowledge distillation and model compression techniques. The disclosed framework is built as a light-weight deep learning model that can be served in production and perform on par with large models.

Abstract: One or more computing devices, systems, and/or methods for debiasing training data based upon information seeking behaviors are provided. Users associated with a set of training data are segmented into information seeking behavior groups corresponding to varying degrees of information seeking behaviors of the users. Biases for the information seeking behavior groups may be estimated based upon information seeking behaviors of users within the information seeking behavior groups. The training set of data is debiased using the biases to generate a debiased training set of data. A model may be trained to perform a task based upon the debiased training set of data.

Abstract: The present teaching relates to personalized content recommendation. A webpage is contrasted for a user having a plurality of slots each of which is to be allocated with a content item. For each of the plurality of slots, a plurality of content items in a plurality of types of content are accessed. For each of the plurality of types of content, a personalized score is predicted for each content item in the type of content, wherein the personalized score is obtained based on a trained model trained. A recommended content item of the type of content is selected based on personalized scores. An overall recommended content item is selected and allocated to a slot based on criteria associated with the personalized scores of the recommended content items and a business rule. The webpage with the plurality of slots allocated with content items is provided to the user.

Abstract: The instant system and methods solves the cold start problem through various systems and methods directed to aggregating user interaction data associated with a user over a period of time, scoring the user interaction data to determine at least one user interest relevance score and/or at least one surfacing user interest score for each of the plurality of user interaction types, wherein the scoring includes a time sensitive weighting scheme, and generating a user interest profile partition for each of the plurality of user interaction types based on the at least one user interest relevance score and/or the at least one surfacing user interest score.

Abstract: Systems and methods are disclosed for discovering precursors associated with a current user interaction event. One method comprises receiving a selection of a new item by a user and determining a plurality of similarities between the new item selected by the user and a plurality of historical items, the plurality of historical items being associated with prior user activity. Then a plurality of importance weights associated with the plurality of historical items are determined. Based on the plurality of similarities and the plurality of importance weights, at least one enhanced importance matrix is generated. Further, prior interactions of the user with the plurality of historical items are determined. Based on the enhanced importance matrix and the prior interactions of the user with the plurality of historical items, precursors for the new item selected by the user are identified and provided to a display.

Abstract: Systems and methods are disclosed for discovering precursors associated with a current user interaction event. One method comprises receiving a selection of a new item by a user and determining a plurality of similarities between the new item selected by the user and a plurality of historical items, the plurality of historical items being associated with prior user activity. Then a plurality of importance weights associated with the plurality of historical items are determined. Based on the plurality of similarities and the plurality of importance weights, at least one enhanced importance matrix is generated. Further, prior interactions of the user with the plurality of historical items are determined. Based on the enhanced importance matrix and the prior interactions of the user with the plurality of historical items, precursors for the new item selected by the user are identified and provided to a display.

Abstract: The present teaching relates to personalized content recommendation. A webpage is contrasted for a user having a plurality of slots each of which is to be allocated with a content item. For each of the plurality of slots, a plurality of content items in a plurality of types of content are accessed. For each of the plurality of types of content, a personalized score is predicted for each content item in the type of content, wherein the personalized score is obtained based on a trained model trained. A recommended content item of the type of content is selected based on personalized scores. An overall recommended content item is selected and allocated to a slot based on criteria associated with the personalized scores of the recommended content items and a business rule. The webpage with the plurality of slots allocated with content items is provided to the user.

Abstract: A method is provided, including: detecting interactions by a plurality of users with a plurality of content items, each content item having an associated content item vector; for a given user, identifying interactions occurring during a current time period, including identifying positive interactions with a first set of the content items, and negative interactions with a second set of the content items; processing a first set of the content item vectors that are associated with the first set of the content items to determine a positive interaction vector; processing a second set of the content item vectors that are associated to the second set of the content items to determine a negative interaction vector; for the given user, generating a current user profile vector for the current time period, using the positive interaction vector, the negative interaction vector, and a prior user profile vector for a prior time period.

Abstract: Systems and methods are disclosed for discovering precursors associated with a current user interaction event. One method comprises receiving a selection of a new item by a user and determining a plurality of similarities between the new item selected by the user and a plurality of historical items, the plurality of historical items being associated with prior user activity. Then a plurality of importance weights associated with the plurality of historical items are determined. Based on the plurality of similarities and the plurality of importance weights, at least one enhanced importance matrix is generated. Further, prior interactions of the user with the plurality of historical items are determined. Based on the enhanced importance matrix and the prior interactions of the user with the plurality of historical items, precursors for the new item selected by the user are identified and provided to a display.

Abstract: A method is provided, including: detecting interactions by a plurality of users with a plurality of content items, each content item having an associated content item vector; for a given user, identifying interactions occurring during a current time period, including identifying positive interactions with a first set of the content items, and negative interactions with a second set of the content items; processing a first set of the content item vectors that are associated with the first set of the content items to determine a positive interaction vector; processing a second set of the content item vectors that are associated to the second set of the content items to determine a negative interaction vector; for the given user, generating a current user profile vector for the current time period, using the positive interaction vector, the negative interaction vector, and a prior user profile vector for a prior time period.