Classifying Agreement and Disagreement Level in Online Conversation

DOI: 10.5176/978-981-08-8227-3_cgat11-27

Authors: Vanichphat and Vishnu Kotrajaras

Abstract:

Online conversation logs from web boards or chat programs can be used for many purposes. For example, the information can be used for analyzing agreement or disagreement about goods or services, articles, or even opinions. In this paper, we propose a novel yet easy to implement approach for capturing the level of opinions from online chat conversation. Recorded chat conversation logs are first processed by Opinion Finder, which gives polarity opinions as agree/disagree. This information is then used with additional information to train Artificial Neural Network to further recognize the level of agreement/disagreement. The results of our experiment show that agreement/disagreement level can be classified accurately.

Keywords: Agreement/Disagreement Detection; Neural Network; Classification; Online Chat