CONTENT BASED TWEET CLASSIFICATION ON TWITTER
L. A. Lalitha, P. Sumitha, P. Snavaja Krishnan, Shanu Sushmita, R M Vinaya · International Journal of Advanced Research in Computer Science · 2020
Today, Social Media Networks are more powerful and popular than any other forms of media that exist and due to this global nature of social media, the amount of information available and being shared online by the users is tremendous. This large data that is available can be used for different purposes like marketing, data analysis, community detection, fraud detection, sentiment analysis, etc. In this work, we present a model to classify tweets in Twitter and therefore offer a solution to process large amounts of data and derive meaningful conclusions from the same. Here, we first collect tweets from different communities on twitter and process this raw dataset. This processed data is then converted into a vector form so that the textual information is converted to a numeric form for the machine to implement and then a text classification algorithm is applied to this dataset. Finally, after training the machine using this dataset, the working of the model and its accuracy is evaluated by using a dataset of test tweets where the machine predicts the category to which the test tweet belongs. With this model, we have been able to classify tweets into different categories and have achieved satisfactory results.