Emotion space model for classifying opinions in stock message board

作者：

Highlights：

• We focus on the investors' opinion in stock message board.

• Extending six basic emotion words increases classifying accuracy.

• Opinion classifying based on ESM is effective and fast.

• We compare ESM with traditional SVM information gain and mutual information.

摘要

•We focus on the investors' opinion in stock message board.•Extending six basic emotion words increases classifying accuracy.•Opinion classifying based on ESM is effective and fast.•We compare ESM with traditional SVM information gain and mutual information.

论文关键词：Opinion classification,Emotion,Stock message

论文评审过程：Received 28 October 2014, Revised 9 August 2015, Accepted 12 August 2015, Available online 16 September 2015, Version of Record 10 November 2015.

论文官网地址：https://doi.org/10.1016/j.eswa.2015.08.023