Automatically Evaluating Opinion Prevalence in Opinion Summarization

TL;DR

This paper introduces an automatic metric to evaluate how well opinion summaries reflect the prevalence of opinions in reviews, revealing current methods lag behind human performance but can be improved with preprocessing techniques.

Contribution

The paper proposes a new opinion prevalence metric based on factual consistency scoring, and demonstrates its effectiveness in evaluating and improving opinion summarization methods.

Findings

Human summaries slightly outperform random review extracts.

Existing unsupervised summarization methods underperform compared to humans.

Preprocessing reviews by simplification can significantly improve summarization quality.

Abstract

When faced with a large number of product reviews, it is not clear that a human can remember all of them and weight opinions representatively to write a good reference summary. We propose an automatic metric to test the prevalence of the opinions that a summary expresses, based on counting the number of reviews that are consistent with each statement in the summary, while discrediting trivial or redundant statements. To formulate this opinion prevalence metric, we consider several existing methods to score the factual consistency of a summary statement with respect to each individual source review. On a corpus of Amazon product reviews, we gather multiple human judgments of the opinion consistency, to determine which automatic metric best expresses consistency in product reviews. Using the resulting opinion prevalence metric, we show that a human authored summary has only slightly better opinion prevalence than randomly selected extracts from the source reviews, and previous extractive and abstractive unsupervised opinion summarization methods perform worse than humans. We demonstrate room for improvement with a greedy construction of extractive summaries with twice the opinion prevalence achieved by humans. Finally, we show that preprocessing source reviews by simplification can raise the opinion prevalence achieved by existing abstractive opinion summarization systems to the level of human performance.