Learning Opinion Summarizers by Selecting Informative Reviews

09/09/2021
by   Arthur Bražinskas, et al.
15

Opinion summarization has been traditionally approached with unsupervised, weakly-supervised and few-shot learning techniques. In this work, we collect a large dataset of summaries paired with user reviews for over 31,000 products, enabling supervised training. However, the number of reviews per product is large (320 on average), making summarization - and especially training a summarizer - impractical. Moreover, the content of many reviews is not reflected in the human-written summaries, and, thus, the summarizer trained on random review subsets hallucinates. In order to deal with both of these challenges, we formulate the task as jointly learning to select informative subsets of reviews and summarizing the opinions expressed in these subsets. The choice of the review subset is treated as a latent variable, predicted by a small and simple selector. The subset is then fed into a more powerful summarizer. For joint training, we use amortized variational inference and policy gradient methods. Our experiments demonstrate the importance of selecting informative reviews resulting in improved quality of summaries and reduced hallucinations.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
11/16/2022

Noisy Pairing and Partial Supervision for Opinion Summarization

Current opinion summarization systems simply generate summaries reflecti...
research
12/21/2022

OpineSum: Entailment-based self-training for abstractive opinion summarization

A typical product or place often has hundreds of reviews, and summarizat...
research
04/30/2020

Few-Shot Learning for Abstractive Multi-Document Opinion Summarization

Opinion summarization is an automatic creation of text reflecting subjec...
research
12/14/2020

Unsupervised Opinion Summarization with Content Planning

The recent success of deep learning techniques for abstractive summariza...
research
05/19/2023

Attributable and Scalable Opinion Summarization

We propose a method for unsupervised opinion summarization that encodes ...
research
06/03/2022

Beyond Opinion Mining: Summarizing Opinions of Customer Reviews

Customer reviews are vital for making purchasing decisions in the Inform...
research
11/27/2022

Unsupervised Opinion Summarisation in the Wasserstein Space

Opinion summarisation synthesises opinions expressed in a group of docum...

Please sign up or login with your details

Forgot password? Click here to reset