Abstract: Some implementations relate to methods, systems, and computer-readable media to generate text tags for games. In some implementations, a computer-implemented method to generate one or more text tags includes obtaining a plurality of chat transcripts, each chat transcript associated with a respective gameplay session of a respective game of a plurality of games. Each chat transcript includes content provided by participants in the gameplay session. The method further includes programmatically analyzing the plurality of chat transcripts to determine one or more characteristics for each game of the plurality of games, and generating a text tag for at least one game of the plurality of games based on the one or more characteristics of the at least one game.

Abstract: Implementations relate to providing search suggestions on a collaboration platform. In some implementations, a method includes generating a dictionary of terms comprising terms associated with games of the collaboration platform by identifying a set of the games that have a frequency of play that meets a threshold frequency and identifying a plurality of search terms associated with previous search queries of the collaboration platform. The method further includes comparing the plurality of search terms with the dictionary of terms, identifying a first set of the search terms that are candidate search terms and a second set of the search terms that are search suggestions, and generating a record that associates candidate search terms of the first set of the search terms with respective search suggestions wherein the record is used to provide a suggestion for one or more new search terms of a new search query.

Abstract: Implementations relate to systems, methods, and computer-readable media to generate text tags for games. A computer-implemented method is provided to generate one or more text tags for a game using a trained machine learning model. Data that includes a game identifier of the game and a set of digital assets associated with the game are provided as input to the trained machine learning model. Predicted text tags are generated using the trained machine learning model based on the set of digital assets associated with the game. The text tags are associated with a respective prediction score. One or more text tags are selected from the plurality of predicted text tags based on the respective prediction score.

Abstract: Some implementations relate to methods and computer-readable media to automatically generate search suggestions. In some implementations, a method includes receiving search session data for a plurality of search sessions, each search session comprising a plurality of search terms input by a respective user. The method further includes, for each search session, identifying that a particular search term of the plurality of search terms is followed by a gameplay session of a particular game of a plurality of games of an online gaming platform, determining that the particular search term is a successful term when the particular search term is a last search term in the sequence, and in response to the determining, adding the search session data to a training corpus. The method further includes applying a machine learning algorithm to the training corpus to generate a plurality of embeddings of search terms.

Abstract: Implementations relate to systems, methods, and computer-readable media to generate text tags for games. A computer-implemented method is provided to generate one or more text tags for a game using a trained machine learning model. Data that includes a game identifier of the game and a set of digital assets associated with the game are provided as input to the trained machine learning model. Predicted text tags are generated using the trained machine learning model based on the set of digital assets associated with the game. The text tags are associated with a respective prediction score. One or more text tags are selected from the plurality of predicted text tags based on the respective prediction score.

Abstract: A method includes identifying a search term and obtaining historical data indicative of a number of times a game was selected from search results corresponding to search queries using the search term and indicative of a number of times the selected game was played by one or more users for at least a threshold amount of time. The method further includes generating a score based on the historical data. The score is associated with the search term and the selected game.

Abstract: Some implementations relate to methods, systems, and a computing device to generate predicted scores for games based on gameplay duration. In some implementations, a method includes obtaining game session data that includes a respective session duration for each game session of a plurality of game sessions, each game session associated with a respective game and a respective game player. The method further includes grouping the plurality of games game clusters based on the game session data, each game cluster including one or more of the games such that no game is in more than one cluster. The method further includes generating a predicted score for one or more of the games and identifying at least one game as a high quality game based on the predicted score. In some implementations, generating the predicted store includes determining a distance between each game and the decision boundary.

Abstract: A user query, such as a user query processed by the online game system, is provided as input into a trained machine learning model. The machine learning model predicts candidate languages of the user query and outputs language confidence scores for the candidate languages. The user query is also matched with stored query data associated with game information in individual language databases for the respective candidate languages. A match scores may be determined based on a certainty of the respective response matches. The match scores and the language confidence scores may be weighted to determine a weighted score. The weighted scores of the response matches are applied to decide which game information retrieved from the identified database is used in forming a response of search results to the user.

Abstract: A system and method includes using a queue with reduce operations. A method can include, responsive to generation of one or more markers by a first node or a second node, causing a submission of one or more markers to a queue associated with a computing cluster. Additionally, responsive to a determination that the first node has completed a first reduce operation, directing the first node to perform a first copy operation to copy first data identified by a first marker of the one or more markers in the queue, where the first copy operation is performed concurrently with the second reduce operation.

Abstract: A method and system is disclosed that generates a dictionary of terms comprising terms associated with games of a collaboration platform. The method further identifies search terms associated with previous search queries. The method compares the search terms with the terms of the dictionary to determine whether the search terms associated with the previous search queries are in the dictionary of terms. The method identifies a first set of the search terms that are candidates to receive search suggestions and a second set of the search terms that are to be the search suggestions based on the comparison. The method also generates a record that associates candidate search terms of the first set of the search terms with respective search suggestions of the second set of the search terms. The record is used to provide a search suggestion for a search term of a new search query.

Abstract: A user query, such as a user query processed by the online game system, is provided as input into a trained machine learning model. The machine learning model predicts candidate languages of the user query and outputs language confidence scores for the candidate languages. The user query is also matched with stored query data associated with game information in individual language databases for the respective candidate languages. A match scores may be determined based on a certainty of the respective response matches. The match scores and the language confidence scores may be weighted to determine a weighted score. The weighted scores of the response matches are applied to decide which game information retrieved from the identified database is used in forming a response of search results to the user.

Abstract: Some implementations relate to methods and computer-readable media to automatically generate search suggestions. In some implementations, a method includes receiving search session data for a plurality of search sessions, each search session comprising a plurality of search terms input by a respective user. The method further includes, for each search session, identifying that a particular search term of the plurality of search terms is followed by a gameplay session of a particular game of a plurality of games of an online gaming platform, determining that the particular search term is a successful term when the particular search term is a last search term in the sequence, and in response to the determining, adding the search session data to a training corpus. The method further includes applying a machine learning algorithm to the training corpus to generate a plurality of embeddings of search terms.

Abstract: A method and system is disclosed that generates a dictionary of terms comprising terms associated with games of a collaboration platform. The method further identifies search terms associated with previous search queries. The method compares the search terms with the terms of the dictionary to determine whether the search terms associated with the previous search queries are in the dictionary of terms. The method identifies a first set of the search terms that are candidates to receive search suggestions and a second set of the search terms that are to be the search suggestions based on the comparison. The method also generates a record that associates candidate search terms of the first set of the search terms with respective search suggestions of the second set of the search terms. The record is used to provide a search suggestion for a search term of a new search query.

Abstract: Some implementations relate to methods and computer-readable media to automatically generate search suggestions. In some implementations, a method includes receiving search session data for a plurality of search sessions, each search session comprising a plurality of search terms input by a respective user. The method further includes, for each search session, identifying that a particular search term of the plurality of search terms is followed by a gameplay session of a particular game of a plurality of games of an online gaming platform, determining that the particular search term is a successful term when the particular search term is a last search term in the sequence, and in response to the determining, adding the search session data to a training corpus. The method further includes applying a machine learning algorithm to the training corpus to generate a plurality of embeddings of search terms.

Abstract: A method and system is disclosed that provides search with improved discoverability within an online gaming platform.

Abstract: Prominent display of a targeted game in search results from a search within an online gaming platform. In some implementations, a server receives a search query from a device. The search query is executed against a data store of available games to obtain search results indicating a set of games. Based on one or more characteristics of the set of games, it is determined whether the search query is directed to a narrow search for a particular game in the set of games. In response to determining such a narrow search, the particular game is designated with a highlight indication, and the search results and highlight indication are transmitted to the device, causing display of a user interface that includes at least a portion of the search results. The highlight indication causes the device to display an identifier of the particular game as highlighted relative to other search results.

Abstract: Implementations described herein relate to methods, systems, and computer-readable media to personalize recommendations. In some implementations, a method includes computing a user-based similarity matrix and a game-based similarity matrix, obtaining a set of user scores from a first machine learning model and the user-based similarity matrix, and obtaining a set of game scores from a second machine learning model and the game-based similarity matrix. The method also includes combining the set of user scores and the set of game scores to generate a normalized set of scores, identifying a set of personalized recommendations for the user associated with a subset of the normalized set of scores, assigning a rank to each of the plurality of items, generating a list of items wherein items in the list of items are ordered based on respective ranks, and providing a user interface to the user that includes the list of items.

Abstract: A request to perform a set of map reduce jobs that generate a plurality of cross products from a dataset based on a join field is received. The join field indicates that each of the plurality of cross products is to be generated from a corresponding subset of the dataset. Responsive to receiving the request to perform the set of map reduce jobs, the set of map reduce jobs on the dataset to generate the plurality of cross products is performed. The final output data of the set of map reduce jobs is stored. The final output data comprises the plurality of cross products.

Abstract: Some implementations relate to methods, systems, and a computing device to generate predicted scores for games based on gameplay duration. In some implementations, a method includes obtaining game session data that includes a respective session duration for each game session of a plurality of game sessions, each game session associated with a respective game and a respective game player. The method further includes grouping the plurality of games game clusters based on the game session data, each game cluster including one or more of the games such that no game is in more than one cluster. The method further includes generating a predicted score for one or more of the games and identifying at least one game as a high quality game based on the predicted score. In some implementations, generating the predicted store includes determining a distance between each game and the decision boundary.

Abstract: Implementations described herein relate to methods, systems, and computer-readable media to dynamically customize rankings. In some implementations, a method includes receiving, from a user, a request for a list of items available on an online gaming platform, identifying a particular trained machine learning model based at least in part on user data associated with the user, providing an input feature vector to the particular trained machine learning model, obtaining an output feature vector from the particular trained machine learning model, assigning a rank for each of the plurality of items based on the output feature vector, generating the list of items, wherein items in the list of items are ordered based on respective ranks, and providing a user interface to the user that includes the list of items.