Social Media Post Classification in Bangla Language:

Abstract

To communicate with others, social media appears to have surpassed all previous methods. Millions of postings are made every day on social media. The Bangla language is widely utilized on Bangladeshi social media platforms. Classifying them based on textual information is often challenging. Social media posts defy easy categorization. Examination becomes more of a chore when the text is written in the Bangla language. Our goal is to classify these social network posts according to their emotional content, making them more accessible for searching, filtering, and organizing. As a means of gauging the persuasiveness of the posts, we conducted an analysis utilizing Sentiment Analysis. Moreover, we attempted to show by contrasting conventional machine learning with the boosting approach. GradientBoostingClassifier and XGBoost Random Forest Classifier were used for boosting, while Support Vector Classifier and Stochastic Gradient Descent were utilized for standard machine learning (SGD). To rate Bangla-language social media postings, we employed an algorithm with the most reliable results.