Document-specific keyphrase candidate search and ranking

作者：

Highlights：

• An efficient method (KCSP) is proposed for computing patterns within text.

• An entropy based method (PF-H) is presented for ranking them.

• A pattern's gap constraint becomes its inherent property, not specified manually.

• PF-H measures the meaningfulness, uncertainty and uselessness of a pattern.

摘要

•An efficient method (KCSP) is proposed for computing patterns within text.•An entropy based method (PF-H) is presented for ranking them.•A pattern's gap constraint becomes its inherent property, not specified manually.•PF-H measures the meaningfulness, uncertainty and uselessness of a pattern.

论文关键词：Keyphrase candidate search,Sequential pattern mining,Keyphrase candidate ranking,Entropy

论文评审过程：Received 30 June 2017, Revised 14 December 2017, Accepted 16 December 2017, Available online 16 December 2017, Version of Record 22 December 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.12.031