Popularity Assessment of Cricket Player Based on Bangla Text in Social Media

Abstract

Social media has turned into a center field of Data Mining in the most recent decade, in light of the fact that the utilization of internet based Social media has been expanded in all respects quickly. In this research, the most popular social media Twitter has been focused by us, where users can post their opinion, sentiment or expression within 140 characters with various issue of interests, such as Politics, Entertainment, Sports, Lifestyle etc. Twitter is particularly famous on different fields for example sentiment analysis, emotion detection, trend detection and so on. They give API's to data miner which are exceptionally helpful for picking up user’s data for analysis. In our research, we have collected tweets about some cricket players of Bangladesh National Cricket team written in Bengali language. We construct a model to predict whether a tweet is positive or negative and compute player’s popularity by figuring out the complete number of positive tweets about every player. However, sentiment analysis of Bengali text is very difficult and complicated task due to lack of language resource. We used various Machine Learning algorithms and compared them with each other to get the best accuracy.