Bangla Sentiment Analysis Using Artificial Intelligence
MD. Ismail Hossain Sadhin, Elora Majumder Bandhan · SSRN Electronic Journal · 2024
For understanding and interpreting human emotions expressed in textual data as public opinion, reviews, or customer feedback, sentiment analysis is a vital section of natural language processing. By analyzing these sentiments, any organization can adopt proper strategies to enhance its effectiveness. Despite of being one of the most widely spoken languages globally, Bangla remains relatively underexplored in this domain. Rarely research has been conducted for Bangla Sentiment Analysis. In this work, sentiment analysis has been done based on the Bangla language, using a dual-method approach. Initially, sentiment expressions in Bangla text were analyzed with the sequential model. Following this, a comparative analysis is conducted using Support Vector Machines (SVM), that enhanced the accuracy of sentiment classification. The results reveal a notable improvement in accuracy and performance with the SVM approach.