Abstract: In one embodiment, a method includes receiving a search query from a client system of a first user; parsing the search query into n-grams, and calculating confidence-scores for entities in a set of query-analysis-entities based on the n-grams, each confidence-score representing a probability that one or more of the n-grams are intended to reference a respective entity; determining a set of search results matching the search query, and calculating entity-frequencies corresponding to entities in a set of results-analysis-entities based on a histogram analysis of the set of search results; calculating a filter-score entities in a set of prospective-entities, which includes entities from the set of query-analysis-entities and the set of results-analysis-entities, based on the respective confidence-score and entity-frequency; and sending, to the client system, suggested filters corresponding to entities having a filter-score greater than a threshold filter-score, the suggested filters being selectable to modify th

Abstract: In one embodiment, a method includes receiving unstructured text from a user of a social-networking system, determining whether the unstructured text includes a request for a recommendation, identifying one or more first entity names in the unstructured text, generating a structured query based upon the one or more first entity names, identifying, in the social graph, one or more second entity names corresponding to the structured query, and presenting the one or more second entity names and the unstructured text in a social context of the user. The unstructured text may include text of a post or message generated by the user on a social-networking system. A score may be generated based on the unstructured text to determine whether the text includes a request for recommendation using a machine-learning model based on comparison of the unstructured text to the one or more predetermined words associated with requests for recommendation.

Abstract: In one embodiment, a method includes receiving a search query from a client system of a first user; parsing the search query into n-grams, and calculating confidence-scores for entities in a set of query-analysis-entities based on the n-grams, each confidence-score representing a probability that one or more of the n-grams are intended to reference a respective entity; determining a set of search results matching the search query, and calculating entity-frequencies corresponding to entities in a set of results-analysis-entities based on a histogram analysis of the set of search results; calculating a filter-score entities in a set of prospective-entities, which includes entities from the set of query-analysis-entities and the set of results-analysis-entities, based on the respective confidence-score and entity-frequency; and sending, to the client system, suggested filters corresponding to entities having a filter-score greater than a threshold filter-score, the suggested filters being selectable to modify th

Abstract: In one embodiment, a method includes receiving unstructured text from a user of a social-networking system, determining whether the unstructured text includes a request for a recommendation, identifying one or more first entity names in the unstructured text, generating a structured query based upon the one or more first entity names, identifying, in the social graph, one or more second entity names corresponding to the structured query, and presenting the one or more second entity names and the unstructured text in a social context of the user. The unstructured text may include text of a post or message generated by the user on a social-networking system. A score may be generated based on the unstructured text to determine whether the text includes a request for recommendation using a machine-learning model based on comparison of the unstructured text to the one or more predetermined words associated with requests for recommendation.