Abstract: To generate an embedding model for entities in an online system, a first set of partitions is generated. Each partition of the first set of partitions includes a subset of entities of the online system. Each partition of at least a subset of partitions of the first set of partitions is assigned to embedding workers. Each of the embedding worker determines embedding vectors for each entity in the partition assigned to the embedding worker. A second set of partitions is generated. Each partition of at least a subset of partitions of the second set of partitions are assigned to embedding workers. Each embedding worker retrieves embedding vectors for the entities in the partition assigned to embedding worker, and determines updated embedding vectors for each of the entities based on the retrieved embedding vectors and information about interaction between the entities.

Abstract: In one embodiment, a computing device receives text input from a user. The computing device determines an action by the user indicated in the text, and at least a part of a phrase in the text that indicates one or more topics associated with the action. The computing device determines whether the at least part of a phrase matches one or more phrases stored in an indexed database. When the at least part of a phrase does not match one or more phrases of the indexed database, then the computing device determines a complete phrase input by the user, and determines one or more topics to be associated with the complete phrase. The computing device stores the complete phrase as an instance of the complete phrase. The computing device determines whether at least a threshold number of instances of the complete phrase has been stored.

Abstract: To generate an embedding model for entities in an online system, a first set of partitions is generated. Each partition of the first set of partitions includes a subset of entities of the online system. Each partition of at least a subset of partitions of the first set of partitions is assigned to embedding workers. Each of the embedding worker determines embedding vectors for each entity in the partition assigned to the embedding worker. A second set of partitions is generated. Each partition of at least a subset of partitions of the second set of partitions are assigned to embedding workers. Each embedding worker retrieves embedding vectors for the entities in the partition assigned to embedding worker, and determines updated embedding vectors for each of the entities based on the retrieved embedding vectors and information about interaction between the entities.

Abstract: In one embodiment, a computing device receives text input from a user. The computing device determines an action by the user indicated in the text, and at least a part of a phrase in the text that indicates one or more topics associated with the action. The computing device determines whether the at least part of a phrase matches one or more phrases stored in an indexed database. When the at least part of a phrase does not match one or more phrases of the indexed database, then the computing device determines a complete phrase input by the user, and determines one or more topics to be associated with the complete phrase. The computing device stores the complete phrase as an instance of the complete phrase. The computing device determines whether at least a threshold number of instances of the complete phrase has been stored.