A hybrid approach to multi-document summarization of opinions in reviews

Abstract

We present a hybrid method to generate summaries of product and services reviews by combining natural language generation and salient sentence selection techniques. Our system, STARLET-H, receives as input textual reviews with associated rated topics, and produces as output a natural language document summarizing the opinions expressed in the reviews. STARLET-H operates as a hybrid abstractive/extractive summarizer: using extractive summarization techniques, it selects salient quotes from the input reviews and embeds them into an automatically generated abstractive summary to provide evidence for, exemplify or justify positive or negative opinions. We demonstrate that, compared to extractive methods, summaries generated with abstractive and hybrid summarization approaches are more readable and compact.