Information retrieval based on semantic patterns of queries

摘要

An information retrieval method includes pre-processing a set of historical query information and processing a user query. Pre-processing a set of historical query information includes determining a plurality of semantic patterns based on a plurality of queries in the set of historical query information; establishing correspondence relationships between the plurality of semantic patterns and a plurality of filtering and ranking operations. Processing a user query comprises receiving the user query; retrieve a plurality of results in response to the user query; determining a semantic pattern that corresponds to the user query; determining a set of filtering and ranking operations that corresponds to the semantic pattern based on the correspondence relationships; and performing the set of filtering and ranking operations on the plurality of results to generate a set of filtered and ranked results.

主权项

1. An information retrieval method, comprising:
pre-processing a set of historical query information, comprising:
determining a plurality of semantic patterns based on a plurality of queries in the set of historical query information, wherein the determining of the plurality of semantic patterns comprises:
determining whether a number of times a sematic pattern appears in the set of historical query information exceeds a predetermined threshold; andin the event that the number of times the sematic pattern appears in the set of historical query information exceeds the predetermined threshold, selecting the semantic pattern;establishing correspondence relationships between the plurality of semantic patterns and a plurality of filtering and ranking operations, wherein establishing correspondence relationships comprises:
determining user behaviors corresponding to the plurality of semantic patterns based on the set of historical query information;determining a plurality of user intention attributes of the plurality of semantic patterns based on the user behaviors, wherein the plurality of user intention attributes include a degree of ambiguity attribute, an authority requirement attribute, a time efficiency requirement attribute, a location requirement attribute, a volume attribute, or any combination thereof;determining the plurality of filtering and ranking operations based on the plurality of user intention attributes; andstoring correspondence relationships between the plurality of filtering and ranking operations and semantic patterns based on the user behavior sets and user intention attributes; and processing a user query made by a user, comprising:
receiving the user query;retrieve a plurality of results in response to the user query;determining a semantic pattern that corresponds to the user query; determining a set of filtering and ranking operations that corresponds to the semantic pattern based on the correspondence relationships that are established during the pre-processing; and performing the set of filtering and ranking operations on the plurality of results to generate a set of filtered and ranked results.