Authors:
D. Suraj, S. Dinesh, R. Balaji, P. Deepika, Freddy Ajila

Addresses:
1,2,3,4Department of Computer Science and Engineering, SRM Institute of Science and Technology, Ramapuram, Chennai, Tamil Nadu, India. 5Department of Information Technology, Faculty of Computer Science and Electronics, Higher Polytechnic School of Chimborazo (ESPOCH), Orellana Headquarters, El Coca, Ecuador. dr0604@srmist.edu.in1, ds3511@srmist.edu.in2, br1371@srmist.edu.in3, deepikap2@srmist.edu.in4, freddy.ajila@espoch.edu.ec5

Abstract:

Using BERT and TensorFlow in analyzing product reviews is important in understanding customer sentiment. After careful fine-tuning with a large dataset of reviews, BERT becomes adept at categorizing sentiments as positive or negative. This breakthrough benefits businesses by improving their marketing strategy and product development and empowers consumers to make better decisions. The sentiment analysis pipeline integrates seamlessly with a refined BERT model. The BERT model is the key to predicting sentiment polarity. Based on its sensitive insights, BERT classifies reviews by providing a binary difference between positive and negative sentiments. In order to rigorously evaluate the model’s performance, we use several key metrics. These metrics comprehensively evaluate the model’s effectiveness, ensuring reliable results. This research project cannot be overemphasized. It changes the way businesses gauge customer opinions. It serves as a compass to fine-tune marketing strategies and improve product quality. This improves customer satisfaction, strengthening the relationship between businesses and consumers. This research bridges the gap between NLP and sentiment analysis, demonstrating the huge potential for improving sentiment analysis performance and ushering in a new age of cutting-edge applications in NLP. The outcome of this project will be revolutionary. It will revolutionize how businesses understand customer feedback, translate it into more informed decisions, build stronger customer relationships, and deliver better products. In short, it will revolutionize the customer-centric landscape and take the business-consumer relationship to the next level.

Keywords: Sentiment Analysis; Machine Learning; Deep Learning; Artificial Intelligence; Product Reviews; Long Short-Term Memory; Neural Networks; Natural Language Processing (NLP); Recurrent Neural Networks.

Received on: 16/12/2022, Revised on: 21/02/2023, Accepted on: 26/03/2023, Published on: 17/04/2023

FMDB Transactions on Sustainable Computing Systems, 2023 Vol. 1 No. 2, Pages: 77-88

• Views : 250
• Downloads : 10

Download PDF