Abstract: Example implementations include methods and systems that relate to search results and recommendations in a media content delivery system. An example method includes providing a search query to a multi-task language model associated with a media content delivery system. The method also includes providing user engagement information to the multi-task language model. The user engagement information indicates user engagement activity with the media content delivery system. The method also includes retrieving, using the multi-task language model and based on the search query, one or more candidate media items from a media item database of the media content delivery system. The method also includes identifying, using the multi-task language model and based on the user engagement information, one or more recommended media items from the media item database.

Abstract: Example implementations include methods and systems that relate to providing search results in a media content delivery system. An example method includes receiving a search query input via a user interface and generating, by use of a Language Model (LM), a text-based intermediate summary based on the search query input and information about one or more backend services associated with the media content delivery system. The text-based intermediate summary is indicative of a user's search intent. The method also includes generating, by use of the LM, structured instructions based at least on the text-based intermediate summary and the information about the one or more backend services. The structured instructions are processable by one or more of the backend services. The method additionally includes executing the structured instructions by way of one or more of the backend services so as to generate search results.

Abstract: An electronic device generates training data to train a classifier to classify a respective search query as complete or incomplete, including: obtaining a first search query input by a first user; determining a media content item selected by the first user from the first search query; comparing metadata associated with the media content item with the first search query input by the first user; and labeling the first search query as complete or incomplete based on the comparison. The electronic device trains the classifier, using the generated training data, to classify a respective search query as complete or incomplete and uses the trained classifier to determine whether a second search query is complete or incomplete. The electronic device provides, for display, for a second user, one or more complete search queries as recommendations for a received search query, including the second search query if second search query is complete.

Abstract: A computer system associated with a media-providing service is provided, the media-providing service configured to provide a plurality of media items to a plurality of users of the media-providing service. The computer system is configured to perform operations for providing sets of results of media items to users based on input text provided by the users. The operations include receiving, from a user of the media-providing service, an input that includes a text string. The operations include generating, by applying the text string to a trained machine-learning model, a first set of results from the plurality of media items. The operations include retrieving, by applying the text string to a search algorithm, a second set of results being distinct from the first set of results. And the operations include providing, for playback to the user, a representation of the first set of results and the second set of results.

Abstract: An electronic device provides, to a model trained to generate synthetic queries, information corresponding to a first media content item and a first value for a parameter indicative of query specificity. The device receives, from the model trained to generate synthetic queries, a first query corresponding to the first media content item, wherein the first query is generated in accordance with the first value for the parameter indicative of query specificity. The device provides, to a client device for display at the client device, the first query and receives user selection of the first query. In response to the user selection of the first query, the device provides a first result set of media content items for the first query.

Abstract: An electronic device, for a search session, receives one or more user-input search queries and determines, based on interactions with each user-input search query, whether the search session satisfies success criteria. The electronic device generates a graph that includes, for a plurality of search sessions that satisfy the success criteria: a first set of nodes, each node in the first set of nodes corresponding to a respective search query of the plurality of search queries in a respective search session, and a second set of nodes, each node in the second set of nodes corresponding to a respective content item selected from a respective search query of the plurality of search queries in a respective search session. The electronic device converts the first and the second set of nodes of the graph to a vector space and provides a recommendation based on the nodes in the vector space.

Abstract: An electronic device provides, to a model trained to generate synthetic queries, information corresponding to a first media content item and a first value for a parameter indicative of query specificity. The device receives, from the model trained to generate synthetic queries, a first query corresponding to the first media content item, wherein the first query is generated in accordance with the first value for the parameter indicative of query specificity. The device provides, to a client device for display at the client device, the first query and receives user selection of the first query. In response to the user selection of the first query, the device provides a first result set of media content items for the first query.

Abstract: An electronic device generates training data to train a classifier to classify a respective search query as complete or incomplete, including: obtaining a first search query input by a first user; determining a media content item selected by the first user from the first search query; comparing metadata associated with the media content item with the first search query input by the first user; and labeling the first search query as complete or incomplete based on the comparison. The electronic device trains the classifier, using the generated training data, to classify a respective search query as complete or incomplete and uses the trained classifier to determine whether a second search query is complete or incomplete. The electronic device provides, for display, for a second user, one or more complete search queries as recommendations for a received search query, including the second search query if second search query is complete.

Abstract: Methods, systems and computer program products are provided personalizing recommendations of items with associated explanations. The example embodiments described herein use contextual bandits to personalize explainable recommendations (“recsplanations”) as treatments (“Bart”). Bart learns and predicts satisfaction (e.g., click-through rate, consumption probability) for any combination of item, explanation, and context and, through logging and contextual bandit retraining, can learn from its mistakes in an online setting.

Abstract: Methods, systems and computer program products are provided personalizing recommendations of items with associated explanations. The example embodiments described herein use contextual bandits to personalize explainable recommendations (“recsplanations”) as treatments (“Bart”). Bart learns and predicts satisfaction (e.g., click-through rate, consumption probability) for any combination of item, explanation, and context and, through logging and contextual bandit retraining, can learn from its mistakes in an online setting.

Abstract: Methods, systems and computer program products are provided personalizing recommendations of items with associated explanations. The example embodiments described herein use contextual bandits to personalize explainable recommendations (“recsplanations”) as treatments (“Bart”). Bart learns and predicts satisfaction (e.g., click-through rate, consumption probability) for any combination of item, explanation, and context and, through logging and contextual bandit retraining, can learn from its mistakes in an online setting.

Abstract: Methods, systems and computer program products are provided personalizing recommendations of items with associated explanations. The example embodiments described herein use contextual bandits to personalize explainable recommendations (“recsplanations”) as treatments (“Bart”). Bart learns and predicts satisfaction (e.g., click-through rate, consumption probability) for any combination of item, explanation, and context and, through logging and contextual bandit retraining, can learn from its mistakes in an online setting.

Abstract: Methods, systems and computer program products are provided personalizing recommendations of items with associated explanations. The example embodiments described herein use contextual bandits to personalize explainable recommendations (“recsplanations”) as treatments (“Bart”). Bart learns and predicts satisfaction (e.g., click-through rate, consumption probability) for any combination of item, explanation, and context and, through logging and contextual bandit retraining, can learn from its mistakes in an online setting.

Abstract: Techniques are provided for prediction locations of users that submit search queries. A query is received at a search engine. An inverted index is searched to identify one or more geographical locations associated with one or more terms of the received query. The inverted index lists a plurality of query terms and one or more geographical locations associated with each query term. Each geographic location that is associated with a listed query term in the inverted index is a determined location for at least one user previously having submitted the listed term in a search query. A geographical location is predicted for a user that submitted the received query based on the identified one or more geographical locations. In this manner, a location is predicted for the user based on similar queries previously submitted by users.

Abstract: Example methods, apparatuses, or articles of manufacture are disclosed that may be implemented, in whole or in part, using one or more computing devices to facilitate or otherwise support one or more processes or operations for identifying points of interest in a text, such as in an unstructured text, for example, in connection with bootstrapping points of interest via social media.

Abstract: Techniques are provided for prediction locations of users that submit search queries. A query is received at a search engine. An inverted index is searched to identify one or more geographical locations associated with one or more terms of the received query. The inverted index lists a plurality of query terms and one or more geographical locations associated with each query term. Each geographic location that is associated with a listed query term in the inverted index is a determined location for at least one user previously having submitted the listed term in a search query. A geographical location is predicted for a user that submitted the received query based on the identified one or more geographical locations. In this manner, a location is predicted for the user based on similar queries previously submitted by users.