Twitter, análise de sentimento e desenvolvimento de produtos: quanto os usuários estão expressando suas opiniões?

Abstract

This paper aims to perform an investigation about how Twitter can work as a way to make direct contact with products’ clients. Using a machine learning method called SVM, it intends to estimate how much Twitter users are expressing their opinions about a certain product. The operating system called Windows 7® was chosen as a study case and a message gathering from Twitter was done. The SVM was trained by a set of messages manually classified as being neutral or containing opinions. Test with a set built in the same manner was also performed, resulting in an accuracy of 80%. Finally, remaining messages were submitted to the classifier already trained and separated in two distinct classes (opinative and neutral). It was concluded that most part of what is posted on Twitter doesn’t express the user’s opinion, being necessary a considerable amount of messages to perform a poll only with those containing sentiments related to the product in case.