Instruct-DeBERTa: A Hybrid Approach for Aspect-based Sentiment Analysis on Textual Reviews

TL;DR

The paper addresses fine-grained ABSA by reviewing the evolution from lexicon-based to Transformer-based methods and proposing a hybrid pipeline, Instruct-DeBERTa, that combines InstructABSA for ATE with DeBERTa-V3-base-absa-V1 for ASC. It systematically compares multiple baselines (including LLaMA 2-7B and Mistral-7B fine-tuned with QLoRA, SetFit, and various BERT/RoBERTa/DeBERTa models) on SemEval Res-14 and Lap-14 datasets. The hybrid model achieves state-of-the-art joint ABSA performance, with ATE around 91.4–91.6 and ASC around 88.6–89.7 on Res-14/Lap-14, and joint F1 scores of 80.78% and 80.94% respectively, indicating robust cross-domain applicability. These results highlight the value of integrating task-specific transformers in a pipelined ABSA framework and have practical implications for extracting precise consumer insights across domains.

Abstract

Aspect-based Sentiment Analysis (ABSA) is a critical task in Natural Language Processing (NLP) that focuses on extracting sentiments related to specific aspects within a text, offering deep insights into customer opinions. Traditional sentiment analysis methods, while useful for determining overall sentiment, often miss the implicit opinions about particular product or service features. This paper presents a comprehensive review of the evolution of ABSA methodologies, from lexicon-based approaches to machine learning and deep learning techniques. We emphasize the recent advancements in Transformer-based models, particularly Bidirectional Encoder Representations from Transformers (BERT) and its variants, which have set new benchmarks in ABSA tasks. We focused on finetuning Llama and Mistral models, building hybrid models using the SetFit framework, and developing our own model by exploiting the strengths of state-of-the-art (SOTA) Transformer-based models for aspect term extraction (ATE) and aspect sentiment classification (ASC). Our hybrid model Instruct - DeBERTa uses SOTA InstructABSA for aspect extraction and DeBERTa-V3-baseabsa-V1 for aspect sentiment classification. We utilize datasets from different domains to evaluate our model's performance. Our experiments indicate that the proposed hybrid model significantly improves the accuracy and reliability of sentiment analysis across all experimented domains. As per our findings, our hybrid model Instruct - DeBERTa is the best-performing model for the joint task of ATE and ASC for both SemEval restaurant 2014 and SemEval laptop 2014 datasets separately. By addressing the limitations of existing methodologies, our approach provides a robust solution for understanding detailed consumer feedback, thus offering valuable insights for businesses aiming to enhance customer satisfaction and product development.

Figures (5)

• Figure 1: Proposed model
• Figure 2: F1 scores of models for the joint task of ASC and ATE for lap-14
• Figure 3: F1 scores of models for the joint task of ASC and ATE for res-14
• Figure 4: Aspect extraction accuracy of models
• Figure 5: Sentiment polarity percentage of models