Modelling SO-CAL in an Inheritance-based Sentiment Analysis Framework

Abstract

Sentiment analysis is the computational study of people's opinions, as expressed in text. This is an active area of research in Natural Language Processing with many applications in social media. There are two main approaches to sentiment analysis: machine learning and lexicon-based. The machine learning approach uses statistical modelling techniques, whereas the lexicon-based approach uses 'sentiment lexicons' containing explicit sentiment values for individual words to calculate sentiment scores for documents. In this paper we present a novel method for modelling lexicon-based sentiment analysis using a lexical inheritance network. Further, we present a case study of applying inheritance-based modelling to an existing sentiment analysis system as proof of concept, before developing the ideas further in future work.