Twitter Sentiment Analysis
Chirag Kariya, P. A. Khodke
Abstract
In this era of growing social media users, Twitter has significantly large number of daily users who post their opinions in the form of tweets. This paper presents an idea of extracting sentiments out of the tweet and an approach towards classifying a tweet into positive, negative or neutral. This approach can be in many ways useful to any organization, who gets mentioned or tagged in a tweet. Generally the tweets being unstructured in format, first of all the tweet needs to be converted into the structured format. In this paper, tweets are resolved using pre-processing phase and access of tweets has been accomplished via libraries using Twitter API. The datasets need to be trained using algorithms in a way, such that, it becomes capable of testing the tweets and it releases the required sentiments out of the feeded tweets.