Senti-N-Gram: An n-gram lexicon for sentiment analysis

作者：

Highlights：

• An n-gram sentiment dictionary called Senti-N-Gram is developed along with scores.

• An automatic score calculation algorithm is proposed for n-grams.

• A random corpus is used instead of human-annotators to make the dictionary.

• To validate the score of n-grams with that of human annotators used t-test.

• A new ratio-based method is proposed to classify the sentiment of consumer reviews.

摘要

•An n-gram sentiment dictionary called Senti-N-Gram is developed along with scores.•An automatic score calculation algorithm is proposed for n-grams.•A random corpus is used instead of human-annotators to make the dictionary.•To validate the score of n-grams with that of human annotators used t-test.•A new ratio-based method is proposed to classify the sentiment of consumer reviews.

论文关键词：Senti-N-Gram,n-grams,Lexicon,Sentiment analysis,Automatic sentiment score calculation,Automatic n-gram construction

论文评审过程：Received 4 August 2017, Revised 28 February 2018, Accepted 2 March 2018, Available online 6 March 2018, Version of Record 20 March 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.03.004