Abstract: Systems, methods, and non-transitory computer readable media can determine one or more actions that a user is likely to take on a page associated with a social networking system, based on one or more first machine learning models. One or more card types that correspond to the one or more actions can be ranked based on a second machine learning model. One or more cards can be generated based on the ranked card types, and each card can include a recommended action associated with the page.

Abstract: A system identifies unconnected content items of high quality and provides the unconnected content items for display to a user. The method comprises receiving several content items posted on pages of a social networking system. The system then determines a subset of those content items (e.g., high quality content items). A topic is then extracted from each of the subset of content items. The topic is mapped to one or more related pages of the social networking system that represent an expanded set of pages associated with the content item. For each of the related pages, a user is identified who is connected to the related page. Finally, the content item (e.g., a high quality content item) is provided to the user for display in the user's newsfeed.

Abstract: Systems, methods, and non-transitory computer readable media are configured to train a machine learning model. The training can be based on a training set of embeddings of a first type and a training set of embeddings of a second type. The machine learning model can be trained to receive an embedding of a second type and to output a corresponding embedding of the first type. A given embedding of the second type can be provided as input to the machine learning model. An embedding of the first type can be obtained from the machine learning model. The embedding of the first type can correspond to the given embedding of the second type.

Abstract: Technology for media item and user language classification is disclosed. Media item classification may use models for associating language identifiers or probability distributions for multiple languages with linguistic content. User language classification may define user language models for attributing to users indications of languages they speak read, and/or write. The text classifications and user classifications may interact because the probability that given text is in a particular language may depend on a determined likelihood the user who produced the text speaks that language, or conversely, a user interacting with text in a particular language may increase the likelihood they understand that language. Some embodiments use language-tagged social media content to train n-gram classifiers for use with other social media content.

Abstract: Technology for media item and user language classification is disclosed. Media item classification may use models for associating language identifiers or probability distributions for multiple languages with linguistic content. User language classification may define user language models for attributing to users indications of languages they speak read, and/or write. The text classifications and user classifications may interact because the probability that given text is in a particular language may depend on a determined likelihood the user who produced the text speaks that language, or conversely, a user interacting with text in a particular language may increase the likelihood they understand that language. Some embodiments use language-tagged social media content to train n-gram classifiers for use with other social media content.

Abstract: Systems, methods, and non-transitory computer readable media are configured to apply a spectral clustering technique to at least a portion of a similarity graph to generate clusters of geographic sub-regions constituting geographic regions. A tf-idf technique is performed to determine pages of a social networking system associated with a geographic region as potential local suggestions for a user associated with a geographic sub-region in the geographic region. References to at least a portion of the pages are presented as local suggestions to the user.

Abstract: Systems, methods, and non-transitory computer-readable media can calculate user similarity scores for a plurality of users on a social networking system with respect to a first user based on user embeddings for the plurality of users and the first user. A set of similar users comprising a plurality of similar users is determined based on the user similarity scores. Page recommendation scores are calculated for a plurality of pages associated with the plurality of similar users based on the user similarity scores. One or more page recommendations are determined for the first user based on the page recommendation scores.

Abstract: Systems, methods, and non-transitory computer-readable media can receive JOINKEY information identifying a JOINKEY and embedding element information identifying a plurality of embedding elements associated with the JOINKEY. A training instance is created comprising a pre-determined number of embedding elements from the plurality of embedding elements. A plurality of negative samples from a sample cache are added to the training instance. One or more embeddings are updated based on the training instance.

Abstract: A social networking system selects a set of groups for presentation to a user of the social networking system. To select groups, the social networking system determines scores for various groups representing a likelihood of the user interacting with the groups. The social networking system may identify a set of groups based on interactions between the user and various groups occurring during a specific time interval and determine scores for groups in the set. When determining a score for a group, the social networking system accounts for times associated with interactions between the user and various groups. Based on the scores, one or more groups are selected and presented to the user. Additionally, the social networking system may identify a time to present the selected one or more groups to the user based on prior user interactions with various groups.

Abstract: Systems, methods, and non-transitory computer-readable media can select a set of selected pages from a plurality of pages on a social networking system based on page selection criteria. A set of potential stories from the set of selected pages is aggregated. The set of potential stories are ranked based on ranking criteria. An administrator feed associated with a first page is generated, the administrator feed comprising a plurality of stories from the set of potential stories based on the ranking the set of potential stories.

Abstract: In one embodiment, an embedding is determined for each entity in a set of entities that is selected from a plurality of entities. Each embedding corresponds to a point in an embedding space, which includes points corresponding to embeddings of entities. The embeddings of the entities are determined using a deep-learning model. Embeddings are determined for each entity attribute in a set of entity attributes. Each of the entity attributes in the set is of an entity-attribute type and is associated with at least one entity. The entity-attribute embeddings are refined using the deep-learning model. The embeddings of the entities in the set of entities are modified based on the entity-attribute embeddings that are associated with the respective entity to obtain updated embeddings for each entity in the set. The updated embeddings include information regarding the entity attributes that are associated with the respective entities.

Abstract: Systems, methods, and non-transitory computer-readable media can train a machine learning model to determine predictive search recommendation based on search prediction information. Search prediction information associated with a user is provided to the machine learning model. A predictive search recommendation is presented to the user based on the machine learning model and the search prediction information. A search is performed based on the predictive search recommendation for one or more search results associated with entity pages on a social networking system.

Abstract: Systems, methods, and non-transitory computer-readable media can present a service directory landing page comprising a plurality of selectable service category options associated with a plurality of pre-defined service categories. A search results page is presented, including one or more search results based on search criteria. Each of the one or more search results is associated with an entity page of a social networking system. The service directory landing page and the search results page are accessible without logging into the social networking system. Each entity page on the social networking system is accessible only when logged into the social networking system.

Abstract: Systems, methods, and non-transitory computer readable media are configured to receive values associated with features corresponding to an instance involving a page of a social networking system and an administrator of the page. The values associated with the features are applied to a machine learning model. A probability that the administrator of the page will take action on the page in response to receipt of an electronic notification provided to the administrator is determined based on the machine learning model.

Abstract: Systems, methods, and non-transitory computer-readable media can generate layered training data for determining embeddings for entities that are accessible through the social networking system, wherein the layered training data includes layers of data that are organized by a hierarchy, and wherein each layer of data corresponds to entities of a same type. A respective embedding for each entity in a set of entities can be determined, wherein the embeddings are trained iteratively using each layer of data in the layered training data. One or more candidate entities that are related to a first entity can be determined based at least in part on the respective embeddings for the candidate entities and the first entity. At least a first candidate entity from the one or more candidate entities can be provided as a recommendation to a user that formed a connection with the first entity.

Abstract: Systems, methods, and non-transitory computer-readable media can determine a plurality of candidate entities for recommendation to a user of a social networking system based on candidate criteria. A recommendation pace rating is determined for each of the plurality of candidate entities based on historical recommendation data. A first entity of the plurality of candidate entities is selected for recommendation to the user based on the recommendation pace ratings.

Abstract: Systems, methods, and non-transitory computer-readable media can determine a respective latent representation for each entity in a set of entities that are accessible through the 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, each latent representation for an entity having a lower dimensionality than a topic model of the entity. One or more candidate entities that are related to a first entity can be determined based at least in part on the respective latent representations for the candidate entities and the first entity. At least a first candidate entity from the one or more candidate entities can be provided as a recommendation to a user that formed a connection with the first entity.

Abstract: Systems, methods, and non-transitory computer-readable media can determine one or more geographic clusters that each correspond to a respective portion of a geographic region, each geographic cluster representing a neighborhood that includes a set of places which users residing in the neighborhood tend to frequently visit. A determination can be made that a user is located in a first geographic cluster. At least one content item can be provided to be presented to the user, the content item being associated with the first geographic cluster.

Abstract: A social networking system selects a set of groups for presentation to a user of the social networking system. To select groups, the social networking system identifies candidate groups and selects the set of groups from the candidate groups. To identify certain candidate groups, the social networking system determines a location associated with various groups based on locations associated with users included in the group. For example, the social networking system determines a centroid of a group based on locations associated with users included in the group and associates the centroid with the group if at least a threshold percentage of distances between locations associated with users included in the group and the centroid do not exceed a threshold distance. Groups associated with locations within a threshold distance of a location associated with the user are identified as candidate groups.

Abstract: The social networking system monitors implicit interactions between a user and objects of the social networking system with which the user has not established a connection. Based on the implicit interactions between the user and an object, the social networking system identifies a soft connection between the user and the object. The social networking system may then identify soft connections to include in a candidate list of soft connections to recommend to the user. The social networking system may also extract signals from the set of candidate list of soft connections, and may use the extracted signals to rank the soft connections in the list of candidate soft connections. The social networking system may then recommend soft connections to the user based on the rank associated with the soft connections in the candidate list of soft connections.