Systems and methods for recommending content items

A system comprising:

• at least one processor;
• and
• a memory storing instructions that, when executed by the at least one processor, cause the system to perform:
• determining a respective latent representation for each entity in a set of entities that are accessible through a social networking system, wherein a latent representation for an entity is determined based at least in part on a topic model associated with the entity, wherein the topic model is represented as a data structure in which each element corresponds to a particular topic associated with an input node of at least one trainable model, each latent representation for an entity having a lower dimensionality than a topic model of the entity, wherein determining the respective latent representation for each entity in the set of entities further comprises:
• obtaining, by the social networking system, a respective topic model for each entity in the set of entities;
• and
• training, by the social networking system, the at least one trainable model using the respective topic model to output a respective latent representation that corresponds to the respective topic model,
• wherein training comprises providing the respective latent representation to the at least one trainable model to determine activation of input nodes, including the input node, of the at least one trainable model;
• determining one or more candidate entities that are related to a first entity based at least in part on the respective latent representations for the candidate entities and the first entity;
• and
• providing at least a first candidate entity from the one or more candidate entities as a recommendation to a user that formed a connection with the first entity.

A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computing system, cause the computing system to perform a method comprising:

• determining a respective latent representation for each entity in a set of entities that are accessible through a social networking system, wherein a latent representation for an entity is determined based at least in part on a topic model associated with the entity, wherein the topic model is represented as a data structure in which each element corresponds to a particular topic associated with an input node of at least one trainable model, each latent representation for an entity having a lower dimensionality than a topic model of the entity, wherein determining the respective latent representation for each entity in the set of entities further comprises:
• obtaining, by the social networking system, a respective topic model for each entity in the set of entities;
• and
• training, by the social networking system, the at least one trainable model using the respective topic model to output a respective latent representation that corresponds to the respective topic model,
• wherein training comprises providing the respective latent representation to the at least one trainable model to determine activation of input nodes, including the input node, of the at least one trainable model;
• determining one or more candidate entities that are related to a first entity based at least in part on the respective latent representations for the candidate entities and the first entity;
• and
• providing at least a first candidate entity from the one or more candidate entities as a recommendation to a user that formed a connection with the first entity.