In the domain of e-commerce, customer reviews wield significant influence over business strategies. Despite the existence of various recommendation methodologies like collaborative filtering and deep learning, they often encounter difficulties in accurately analyzing sentiment and semantics within customer feedback. Addressing these challenges headon, this paper introduces BERTFusionDNN, a novel framework merging BERT for extracting textual features and a Deep Neural Network for integrating numerical features. We assess the efficacy of our approach using a Women Clothing ECommerce dataset, benchmarking it against established techniques. Our method adeptly extracts valuable insights from customer reviews, fortifying e-commerce recommendation systems by surmounting barriers associated with deciphering both textual nuances and numerical intricacies. Through this endeavor, we pave the way for more robust and effective strategies in leveraging customer feedback to optimize e-commerce experiences and drive business success.
Zhao, Z., Zhang, N., Xiong, J., Feng, M., Jiang, C., & Wang, X. (2024). Enhancing E-commerce Recommendations: Unveiling Insights from Customer Reviews with BERTFusionDNN. Journal of Theory and Practice of Engineering Science, 4(02), 38-44.
@article{zhao2024enhancing,
title={Enhancing E-commerce Recommendations: Unveiling Insights from Customer Reviews with BERTFusionDNN},
author={Zhao, Zhiming and Zhang, Ning and Xiong, Jize and Feng, Mingyang and Jiang, Chufeng and Wang, Xiaosong},
journal={Journal of Theory and Practice of Engineering Science},
volume={4},
number={02},
pages={38--44},
year={2024}
}