Populating query suggestion database using chains of related search queries

摘要

A system, computer-readable storage medium storing at least one program, and a computer-implemented method for identifying chains of related search queries is presented. Historical search query data is obtained. Chains of related search queries issued by users and search results corresponding to last related search queries in the chains of related search queries that were selected by the users are identified from the historical search query data, where each related search query in a chain of related search queries except for a last related search query in the chain of related search queries violates a search result selection criterion. The chains of related search queries are aggregated into groups, where a respective group has a common first search query and a common search result corresponding to at least one last related search query that was selected by the users. Aggregate data for the groups are stored in a query database.

主权项

1. A computer-implemented method for identifying chains of related search queries, performed on a server having at least one processor and memory storing at least one program for execution by the at least one processor to perform the method, comprising:
receiving a search query from a user; identifying documents satisfying the search query; ranking the documents based at least in part on a frequency that the search query led to a selection of each of the documents to produce ranked search results; selecting a subset of the ranked search results to produce a set of ranked search results; obtaining historical search query data, the historical search query data including data relating to search queries issued by users and search results selected by the users; identifying, from the historical search query data, chains of related search queries issued by users and search results corresponding to last related search queries in the chains of related search queries that were selected by the users, each respective related search query in a respective chain of related search queries, except for a respective last related search query in the respective chain of related search queries, violating timing criterion with respect to user-selection of search results,
wherein a respective chain of related search queries is a sequence of consecutive search queries that are issued by a respective user and that include an initial search query that is successively refined; aggregating the chains of related search queries into groups, a respective group having a common initial search query and a common search result that was selected by the users from search results returned by the last related search queries in the chains of related search queries in the respective group; storing, in a query database, aggregate data for the groups; identifying, using the query database, at least one last related search query in at least one chain of related search queries that is related to the search query and that includes at least one search result that was selected by users who issued the search query; and returning the set of ranked search results and the at least one related search query to the user.