Enhancing Sentiment Analysis of Twitter Data

Abstract

Because so much work is being put into data mining and information closure, sentiment is a typical technique for people to communicate the current trends in internet-based lives. Using sentiment analysis, a number of potential events can be found, including human trafficking, break-ins, and unclear feelings, and so on. Rapid Miner is a social media mining application we use to get unrefined dataset about cricket betting and matchfixing using hashtags. After cleaning the dataset via stemming, lemmatizing, analysing of sentiment etc we get a refined dataset. This study's primary objective is to use sentiment analysis on texts, evaluating the effectiveness of machine learning algorithms such as KNN, Naïve Bayes, Logistic regression and then checking which model has the highest accuracy and fine tune the model into further increasing the accuracy. Keywords: Rapid miner, Python, Cloud mining, Sentiment, LR, KNN, Mapping, New model