• B. Raju
• Dr. B. Selvapriya
• Dr.C. Berin Jones

Sentiment analysis is pivotal in understanding customer emotions and opinions in e-commerce platforms. With the rise of multimodal data, traditional models struggle to extract meaningful insights from text, images, and emojis. This research proposes an advanced deep learning model integrating Fejer Kernel filtering, fuzzy dictionary-based feature extraction, and optimized simulated annealing for feature selection. The model employs BERT for sentiment classification, ensuring high accuracy in sentiment prediction. The proposed methodology is tested on Amazon Customer Reviews and Kaggle datasets, achieving superior performance compared to existing models. This study contributes to the e-commerce sector by providing an efficient sentiment analysis model to improve customer satisfaction and business insights.

Sentiment Analysis, Deep Learning, BERT, Fejer Kernel, Simulated Annealing, E-Commerce

"Optimized Deep Learning Framework for Sentiment Analysis in E-Commerce: Integrating Fejer Kernel Filtering, Fuzzy Semantic Extraction, and BERT", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 9, page no.f626-f637, September 2023, Available :http://www.jetir.org/papers/JETIR2309691.pdf

"Optimized Deep Learning Framework for Sentiment Analysis in E-Commerce: Integrating Fejer Kernel Filtering, Fuzzy Semantic Extraction, and BERT", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 9, page no. ppf626-f637, September 2023, Available at : http://www.jetir.org/papers/JETIR2309691.pdf