Template-based Abstractive Microblog Opinion Summarisation

TL;DR

This paper presents a new dataset and task for microblog opinion summarisation, emphasizing abstractive summaries created by journalists, and benchmarks various models to evaluate their effectiveness.

Contribution

Introduces a large, high-quality dataset for microblog opinion summarisation and demonstrates the advantages of abstractive over extractive models through comprehensive benchmarking.

Findings

Abstractive models outperform extractive models in summarisation tasks.

Fine-tuning significantly improves model performance.

Using different sample sizes affects summarisation quality.

Abstract

We introduce the task of microblog opinion summarisation (MOS) and share a dataset of 3100 gold-standard opinion summaries to facilitate research in this domain. The dataset contains summaries of tweets spanning a 2-year period and covers more topics than any other public Twitter summarisation dataset. Summaries are abstractive in nature and have been created by journalists skilled in summarising news articles following a template separating factual information (main story) from author opinions. Our method differs from previous work on generating gold-standard summaries from social media, which usually involves selecting representative posts and thus favours extractive summarisation models. To showcase the dataset's utility and challenges, we benchmark a range of abstractive and extractive state-of-the-art summarisation models and achieve good performance, with the former outperforming the latter. We also show that fine-tuning is necessary to improve performance and investigate the benefits of using different sample sizes.