Mining Sentiment Using Conversation Ontology

Abstract

The research in the field of opinion mining has been ongoing for several years, and many models and techniques have been proposed. One of the techniques that can address the need for automated information monitoring to help to identify the trends and patterns that matter is sentiment mining. Existing approaches enable the analysis of a large number of text documents, mainly based on their statistical properties and possibly combined with numeric data. Most approaches are limited to simple word counts and largely ignore semantic and structural aspects of content. Conversation plays a vital role in expressing and promoting an opinion. In this chapter, the authors discuss the concept of ontology and propose a framework that allows the incorporation of information on conversation structure in the models for sentiment discovery in text.