Summarizing Opinions: Aspect Extraction Meets Sentiment Prediction and They Are Both Weakly Supervised

TL;DR

This paper introduces a neural, weakly supervised framework for opinion summarization from online reviews, combining aspect extraction and sentiment prediction to produce extractive summaries with minimal supervision.

Contribution

It presents a novel neural approach that jointly performs aspect extraction and sentiment prediction using weak supervision, along with a new dataset for opinion summarization.

Findings

Significant improvements over baseline methods in automatic evaluation

Human judges prefer the proposed summaries across multiple criteria

The framework effectively identifies salient opinions with minimal supervision

Abstract

We present a neural framework for opinion summarization from online product reviews which is knowledge-lean and only requires light supervision (e.g., in the form of product domain labels and user-provided ratings). Our method combines two weakly supervised components to identify salient opinions and form extractive summaries from multiple reviews: an aspect extractor trained under a multi-task objective, and a sentiment predictor based on multiple instance learning. We introduce an opinion summarization dataset that includes a training set of product reviews from six diverse domains and human-annotated development and test sets with gold standard aspect annotations, salience labels, and opinion summaries. Automatic evaluation shows significant improvements over baselines, and a large-scale study indicates that our opinion summaries are preferred by human judges according to multiple criteria.