Instruct-DeBERTa: A Hybrid Approach for Enhanced Aspect-Based Sentiment Analysis with Category Extraction

Dineth Jayakody, Koshila Isuranda, Malkith AVA, Nisansa de Silva, G G N Sandamali, K L K Sudheera, Sachintha Rajith

Keywords: Aspect-Based Sentiment Analysis, Aspect Ex- traction, DeBERTaV3, Hybrid Model, InstructABSA, Natural Language Processing, Sentiment Classification, Textual Reviews

TL;DR: Instruct-DeBERTa combines InstructABSA for aspect extraction and DeBERTa-V3 for sentiment classification, optimized for the hospitality industry, improving aspect-based sentiment analysis performance.

Abstract: Aspect-based sentiment Analysis (ABSA) is an advanced NLP task that identifies sentiments related to specific aspects of a product or service, offering more detailed consumer insights than general sentiment analysis. The proposed research introduces a hybrid model, Instruct-DeBERTa, which combines InstructABSA for aspect term extraction (ATE) and DeBERTa-V3-baseabsa-V1 for aspect sentiment classification (ASC). Using datasets from SemEval Restaurant 2014,2015,2016 and SemEval Laptop 2014, the model demonstrated improved performance across domains. Further enhancements included category classification using cosine similarity, linear layers with ReLU activation, regularization methods, and optimized attention heads for the hospitality domain. These improvements address existing model limitations, providing a comprehensive solution for analyzing consumer feedback, valuable for enhancing customer satisfaction and product development.

Submission Number: 45