Sentiment Regression: Using Real-Valued Scores to Summarize Overall Document Sentiment

In this paper, we consider a sentiment regression problem: summarizing the overall sentiment of a review with a real-valued score. Empirical results on a set of labeled reviews show that real-valued sentiment modeling is feasible, as several algorithms improve upon baseline performance. We also analyze performance as the granularity of the classification problem moves from two-class (positive vs. negative) towards infinite-class (real-valued).

• Type: Conference
• Year: 2008
• Where: SEMCO
• Authors: Adam Drake, Eric K. Ringger, Dan Ventura
• Keywords: Overall Sentiment, Real-valued Sentiment Modeling, Semantic Computing, SEMCO 2008, Sentiment Regression Problem