Authors : S. Rajarajeswari; Ashwin D; Nithish Kumar S; Vishal R; Vishwa N

Volume/Issue : Volume 10 - 2025, Issue 4 - April

Google Scholar : https://tinyurl.com/ym5uku4a

Scribd : https://tinyurl.com/mr47k69d

DOI : https://doi.org/10.38124/ijisrt/25apr1434

Google Scholar

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Note : Google Scholar may take 15 to 20 days to display the article.

Abstract : As the online shopping is growing very fast, business sentiment analysis based on customer reviews is very important for companies. Conventional sentiment analysis tools such as VADER are incapable of picking up high-level language structures, regional idioms, and context-dependent phenomena. This work introduces a state-of-the-art sentiment analysis system using transformer-based architectures such as BERT and RoBERTa, which are fine-tuned over a bespoke e- commerce corpus. The model undertakes Aspect-Based Sentiment Analysis (ABSA) to derive sentiments for particular product features like price, quality, delivery, and customer service. In addition, multilingual support is built using Indic NLP models and Google language detection APIs to support regional languages like Hindi and Tamil. The real-time sentiment stream is built using Apache Kafka, allowing companies to track customer feedback in real-time. Experimental results indicate that the system proposed here is more accurate and relevant than conventional approaches and offers a scalable solution for contemporary e-commerce websites.

Keywords : Sentiment Analysis, BERT, RoBERTa, ABSA, Multilingual NLP, Real-Time Streaming, E-commerce.

References :

1. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova, "BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding," arXiv preprint arXiv:1810.04805, 2018.
2. Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, et al., "RoBERTa: A Robustly Optimized BERT Pretraining Approach," arXiv preprint arXiv:1907.11692, 2019.
3. Hutto, C.J., and Gilbert, E., "VADER: A Parsimonious Rule-based Model for Sentiment Analysis of Social Media Text," Proceedings of the International AAAI Conference on Web and Social Media, vol. 8, no. 1, 2014.
4. Maria Pontiki et al., "SemEval-2014 Task 4: Aspect Based Sentiment Analysis," Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014), 2014.
5. P. Bojanowski, E. Grave, A. Joulin, and T. Mikolov, "Enriching Word Vectors with Subword Information," Transactions of the Association for Computational Linguistics, vol. 5, pp. 135–146, 2017.
6. P. K. Goyal, M. Shrivastava, "Fast and Accurate Sentiment Analysis Using Transformer Models for Code-Mixed Languages," Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2020.
7. R. Zaharia et al., "Apache Kafka: A Distributed Messaging System for Log Processing," Proceedings of the NetDB, 2011.
8. Indic NLP Library, "Resources and Tools for Indian Language Processing," https://anoopkunchukuttan.github.io/indic_nlp_library/, Accessed 2025.
9. T. Wolf et al., "Transformers: State-of-the-Art Natural Language Processing," Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, 2020.
10. P. Rajpurkar et al., "SQuAD: 100,000+ Questions for Machine Comprehension of Text," Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing (EMNLP), 2016.