Bag of meta-words: A novel method to represent document for the sentiment classification

作者：

Highlights：

• A framework using meta-words features to represent document.

• Two models to extract meta-words from the word embedding.

• The overall model can effectively address to sentiment classification.

摘要

•A framework using meta-words features to represent document.•Two models to extract meta-words from the word embedding.•The overall model can effectively address to sentiment classification.

论文关键词：Natural language processing,Sentiment analysis,Sentiment classification,Document representation,Word embedding

论文评审过程：Received 11 December 2017, Revised 11 June 2018, Accepted 29 June 2018, Available online 30 June 2018, Version of Record 6 July 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.06.052