• Version
• Download 13
• File Size 851.89 KB
• File Count 1
• Create Date 18 January 2026
• Last Updated 18 January 2026

Decoding Review Sentiment Analysis Model

(Mainly Using Bert)

Chehakpreet Kaur, UG Student, BCA, Maharaja Surajmal Institute, Delhi, India

Tashi Singh, UG Student, BCA, Maharaja Surajmal Institute, Delhi, India

Dr. Rimpy, Assistant Professor, Department of BCA, Maharaja Surajmal, Delhi, India

Abstract

Review Sentiment Analysis is performed so that, any business or organization can understand how their audience perceives their products or services. The study of their opinion provides them with extremely valuable information. The analysis helps in deciding where and on what aspects the organization should invest in. This paper reviews and looks over different types of methods of deep learning that can be used and expands on the BERT method. The BERT method is a deep learning model designed and used to improve the accuracy and efficiency of Natural Language Processing tasks.

Introduction

Sentiments are the different thoughts, emotions or attitudes caused by different circumstances or feelings. Every person has their own unique thoughts on any topic of discussion. For example, in context of review analysis, every customer has their own particular experience that shapes their sentiments towards various products/services. Something one person has a pleasant experience with, could be awful for another.

Sentiment Analysis is done to sort the experiences provided by different people. It is the process that is used to identify and categorize the emotions and thoughts of these people.

Different types of Sentimental Analysis :

Emotion Analysis : It is the type of analysis where in which different types of emotions- happiness, sadness, anger, etc. are classified.