Social Listening on Budget—A Study of Sentimental Analysis and Prediction of Sentiments Using Text Analytics & Predictive Algorithms

Abstract

Text analytics is a mining technique process applied on the text extracted from various platforms to have an understanding about the context and opinion. And this process of social listening is gaining its importance as it is the direct opinion and feedback from the social class or a consumer pool. Sentimental analysis is process of attaching sentiment to the processed text which is of people’s opinions, attitudes, and judgments on certain context. This study is attempted to understand the sentiment and predict the sentiment on the most debated democratic element of the country, i.e., budget. The data for this study have been collected from Twitter using the APIs on the day of budget. With these data, various machine learning algorithms are applied for the classification problem. For the Twitter sentiment analysis, user sentiments about—#Budget2020 are found out by various sentiment classification techniques. Later, various machine learning models such as SVM, Naïve Bayes, random forest, and XGBoost were modeled. The best model comparison has been made based on evaluation metrics such as accuracy.