Highlighting keyphrases using senti-scoring and fuzzy entropy for unsupervised sentiment analysis

作者：

Highlights：

• An unsupervised sentiment classification system using n-grams technique for online reviews

• Formulation and senti-scoring of phrase patterns

• Senti-scores of phrases are computed from SentiWordNet lexicon and fuzzy linguistic hedges.

• Applied Fuzzy Entropy filter and k-means clustering for extracting and highlighting keyphrases

• Results of comparison with other state-of-the-art indicate the higher scores of our system.

摘要

•An unsupervised sentiment classification system using n-grams technique for online reviews•Formulation and senti-scoring of phrase patterns•Senti-scores of phrases are computed from SentiWordNet lexicon and fuzzy linguistic hedges.•Applied Fuzzy Entropy filter and k-means clustering for extracting and highlighting keyphrases•Results of comparison with other state-of-the-art indicate the higher scores of our system.

论文关键词：Sentiment analysis,Social media,Keyphrases,N-grams,Linguistic hedges,Fuzzy entropy

论文评审过程：Received 10 March 2020, Revised 10 August 2020, Accepted 13 November 2020, Available online 21 November 2020, Version of Record 30 December 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114323