Sentiment Analysis using NLP and Machine Learning Techniques on Social Media Data

Internet usage has made social media an integral part of our everyday lives. With the aid of Natural Language Tool Kit (NLTK), sentiment analysis refers to the process of identifying and analyzing a piece of writing in order to determine whether its sentiment, opinions, views, and emotions are positive, negative, or neutral towards a specific issue, item, etc. People today depend on social media to stay connected. Users are allowed to put their ideologies on Twitter, a widely used communication site. People can write short messages and leave comments. An organization can analyze Twitter sentiments to find out how its image is discussed by individuals. With numerous applications for different spaces, there are numerous methods of sentiment analysis. The two main strategies for analyzing opinions are knowledge base and machine learning. In this study, the Twitter data was collected from tweets that were tagged in voting systems. Text mining was used to pre-process Tweets. Then, using the inverse document frequency and term frequency, a vector space model was constructed, and then sentiment analysis was carried out with Random Forest Classifier, Decision Tree Classifier, and Logistic Regression algorithms. Experiments are discussed and conclusions are drawn.