A SCALABLE METHOD OF SOCIAL MEDIA FOR EVENTS CLASSIFICATION AND MINING TWITTER DATA

Abstract

Social media is one of the mainstream sources for information retrieval and furthermore it is being utilized for sharing everyday occasions of our lives and the occurrences which are going on around the globe. Twitter can be utilized as microblogging administration used to find occasions and news progressively from anyplace in the World. Twitter posts are for the most part short and can be produced continually, so we can state that they are appropriate sources of spilling data for supposition mining and
sentiment extremity location. Twitter posts on a specific occasion or subject can assist us with knowing feelings of individuals about that specific occasion or subject. Sentiment examination on twitter posts can enable us to know how individuals respond to a specific occasion and how their assessment can change if something bizarre occurred. Sentiment examination helps in numerous business regions to know surveys of an occasion. While doing any Machine Learning undertaking, the fundamental concern is an exactness of
a model. In the event that the dataset is exact, at that point we can get a higher exactness of Machine Learning model. The goal of this paper is to examine approach which gives more precision of the machine learning assignment to discover sentiment extremity on Twitter data