Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{187676,
        author = {Jatoth Hari Charan and Kodavath Chintu and Sampathi Dhanush},
        title = {Multimodal Sentiment Analysis Framework Using BERT and GPT-4 for Product Feedback Intelligence},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {6},
        pages = {6335-6340},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=187676},
        abstract = {Contemporary digital platforms host user-generated content where sentiment is expressed through multiple modalities including text, audio, emojis, and numerical ratings. Traditional unimodal sentiment analysis approaches fail to capture the comprehensive emotional context distributed across these diverse data types. This paper presents a novel multimodal sentiment analysis framework that unifies heterogeneous inputs into a cohesive textual representation for comprehensive analysis. Our methodology leverages Bidirectional Encoder Representations from Transformers (BERT) for deep contextual sentiment classi- fication, achieving superior performance in categorizing reviews into positive, neutral, and negative sentiments. Furthermore, we integrate generative artificial intelligence through GPT-4 to synthesize analyzed sentiments into actionable product improve- ment strategies and marketing recommendations. Experimental results demonstrate that our BERT-based classifier achieves 89% accuracy, significantly outperforming conventional sequential models. The integration of GPT-4 enables the generation of coherent, contextually relevant business intelligence, providing enterprises with data-driven insights for product enhancement and customer engagement optimization. This research establishes an effective paradigm for combining state-of-the-art transformer architectures for both sentiment comprehension and strategic business application.},
        keywords = {Multimodal Sentiment Analysis, BERT, GPT-4, Transformer Models, Product Improvement, Customer Feedback Analysis},
        month = {November},
        }