Sentiment analysis using machine learning with twitter data: Case of Torku

With the development of technology, people have started to share their feelings and thoughts on social media platforms. These posts include opinions, criticisms and suggestions on current issues, companies and products. Thanks to these posts, we have a large dataset that can be analyzed in many areas. This dataset facilitates sentiment analysis using machine learning and models. By processing these datasets with machine learning models, the weights of opinions on any subject or topic can be classified as positive-negative-neutral. In this study, using the posts about the word "Torku" on Twitter, one of the social media platforms, the emotional analyzes of the users were tagged using the TextBlob and VaderSentiment libraries, and the emotion weights were determined by training 4 different machine learning models. Afterwards, these models were compared and the best model was tried to be determined for the determined data set. In this way, the ratios of the general opinions of the users about the word "Torku" were determined.