Multi-Class Sentiment Analysis with Clustering and Score Representation

Sentiment analysis or opinion mining is the field of computational study of people's opinion expressed in written language or text. Sentiment analysis brings together various research areas such as natural language processing, data mining and text mining, and is fast becoming of major importance to organizations as they integrate online commerce into their operations. This paper proposes improved methods for aspect-level sentiment analysis. We propose to utilize bag of nouns instead of bog of words to improve the clustering results for aspect identification and a new feature set, score representation, that leads to more accurate sentiment identification. This scheme is based upon the three scores (positive ness, neutral ness and negative ness) that are learned from the data for each term. Using this new score representation scheme, we improve the performance of 3-class sentiment analysis on sentences by 20% in terms of f1-measure, as compared to previously published research. We demonstrate the usefulness of the methodology using data from the popular online travel information site TripAdvisor.com.