DomBERT: Domain-oriented Language Model for Aspect-based Sentiment Analysis

TL;DR

DomBERT is a domain-oriented language model built on BERT that effectively learns from both in-domain and relevant domain corpora, improving aspect-based sentiment analysis especially in low-resource settings.

Contribution

It introduces DomBERT, a novel extension of BERT that integrates domain-specific data to enhance aspect-based sentiment analysis performance.

Findings

Promising results on aspect-based sentiment analysis tasks

Effective learning from low-resource domain data

Improved domain-specific language understanding

Abstract

This paper focuses on learning domain-oriented language models driven by end tasks, which aims to combine the worlds of both general-purpose language models (such as ELMo and BERT) and domain-specific language understanding. We propose DomBERT, an extension of BERT to learn from both in-domain corpus and relevant domain corpora. This helps in learning domain language models with low-resources. Experiments are conducted on an assortment of tasks in aspect-based sentiment analysis, demonstrating promising results.