A Prospect-Guided global query expansion strategy using word embeddings

作者:

Highlights:

• Global query semantics modeled from the standpoint of prospect vocabulary terms.

• Selective semantic exploration strategy adds new terms related to more relevant topics.

• Disambiguation issues addressed without exogenous resources.

• Significant results improving both recall and precision metrics without relevance feedback.

摘要

•Global query semantics modeled from the standpoint of prospect vocabulary terms.•Selective semantic exploration strategy adds new terms related to more relevant topics.•Disambiguation issues addressed without exogenous resources.•Significant results improving both recall and precision metrics without relevance feedback.

论文关键词:Global query expansion,Word embeddings,Information retrieval,Candidate terms pooling methods

论文评审过程:Received 10 February 2017, Revised 27 June 2017, Accepted 9 September 2017, Available online 20 September 2017, Version of Record 20 September 2017.

论文官网地址:https://doi.org/10.1016/j.ipm.2017.09.001