Predicting Tags for Stack Overflow Questions Using Classifier

Abstract

The adequacy of any online educational related platform is mostly dependent on the user’s experience that he/she experiences on using that platform. Hence it is the fundamental requirement to develop a system that takes care of user’s interest into account while posting content on the platform online . Online platforms like Reddit, Quora, Geeks for Geeks, and Stack Exchange have large amount of data in the form of questions and answers that are posted by the users.Now-a-days very large-scale datasets are available by such websites that can be used in the form of input for designing a system based on tag prediction. We have used stack_overflow_tag_prediction dataset. Each question in Stack Overflow generally contains four segments ID, Title,Body and Tags as illustrated in Fig.1. With the help of text in the title and body our system predicts the tags for the particular question. The predicted tags are vital for the actual working of Stack Overflow.