Auto-Tagging

Abstract

In this final year project, our aim is to develop an Auto-Tagging System that automatically recognizes the category in which the question falls. Millions of people ask questions at forum websites on a daily basis and a lot of their time is wasted in just selecting categories manually. Hence, we intend to create a model that itself categorizes the questions correctly, so it saves a lot of time and exterminates the risk of getting falsified data which will predict results with more accuracy in terms of data analysis. For data collection, the forum websites that we have chosen are Quora, Stack-overflow, etc. The goal is to collect at least 1.2 million questions to train our model. We will apply different algorithms for training different models like Decision Tree Classifier, Random Forest and Gaussian Naive Bayes Classifier. The algorithm with the best results will be used for test cases.