AX-MABSA: A Framework for Extremely Weakly Supervised Multi-label Aspect Based Sentiment Analysis

TL;DR

This paper introduces AX-MABSA, a novel framework for multi-label aspect-based sentiment analysis that operates with extremely weak supervision, relying on minimal initial information and unsupervised techniques to outperform existing methods.

Contribution

It presents a new weakly supervised multi-label ABSA framework that requires no labeled data and introduces an automatic seed word selection and multi-label generation approach.

Findings

Outperforms weakly supervised baselines on four datasets

Uses only a single word per class as initial seed

Employs unsupervised language model post-training for improvement

Abstract

Aspect Based Sentiment Analysis is a dominant research area with potential applications in social media analytics, business, finance, and health. Prior works in this area are primarily based on supervised methods, with a few techniques using weak supervision limited to predicting a single aspect category per review sentence. In this paper, we present an extremely weakly supervised multi-label Aspect Category Sentiment Analysis framework which does not use any labelled data. We only rely on a single word per class as an initial indicative information. We further propose an automatic word selection technique to choose these seed categories and sentiment words. We explore unsupervised language model post-training to improve the overall performance, and propose a multi-label generator model to generate multiple aspect category-sentiment pairs per review sentence. Experiments conducted on four benchmark datasets showcase our method to outperform other weakly supervised baselines by a significant margin.