Comparing Apples to Apples: Generating Aspect-Aware Comparative Sentences from User Reviews

TL;DR

This paper presents a transformer-based model that generates personalized, fluent, and diverse comparative sentences from user reviews to help consumers identify the best products by highlighting key features.

Contribution

It introduces a novel three-component transformer pipeline with a unique decoding method for personalized comparison sentence generation from reviews.

Findings

Generated sentences are fluent and diverse.

Human evaluation shows high relevance and truthfulness.

Model effectively highlights important product features.

Abstract

It is time-consuming to find the best product among many similar alternatives. Comparative sentences can help to contrast one item from others in a way that highlights important features of an item that stand out. Given reviews of one or multiple items and relevant item features, we generate comparative review sentences to aid users to find the best fit. Specifically, our model consists of three successive components in a transformer: (i) an item encoding module to encode an item for comparison, (ii) a comparison generation module that generates comparative sentences in an autoregressive manner, (iii) a novel decoding method for user personalization. We show that our pipeline generates fluent and diverse comparative sentences. We run experiments on the relevance and fidelity of our generated sentences in a human evaluation study and find that our algorithm creates comparative review sentences that are relevant and truthful.