A Sentimental Education: Sentiment Analysis Using Subjectivity Summarization Based on Minimum Cuts

TL;DR

This paper introduces a novel machine-learning approach for sentiment analysis that focuses on extracting subjective text portions using minimum cut graph techniques, improving the classification of sentiment polarity.

Contribution

It presents a new method combining graph-based extraction of subjective content with text categorization for enhanced sentiment analysis accuracy.

Findings

Effective extraction of subjective text improves sentiment classification.

Graph-based techniques facilitate incorporation of cross-sentence context.

Method demonstrates promising results on sentiment analysis tasks.

Abstract

Sentiment analysis seeks to identify the viewpoint(s) underlying a text span; an example application is classifying a movie review as "thumbs up" or "thumbs down". To determine this sentiment polarity, we propose a novel machine-learning method that applies text-categorization techniques to just the subjective portions of the document. Extracting these portions can be implemented using efficient techniques for finding minimum cuts in graphs; this greatly facilitates incorporation of cross-sentence contextual constraints.