Twitter Location Prediction usnig Machine Learning Algorithms

Abstract

Predicting a user's location based on their social media profiles is an active area of study right now. For many years, researchers have tried to figure out how to automatically recognise a location based on its association with or mention in a record. However, it might be difficult to determine where a user is located because many accounts do not include this information or because people provide data that does not correlate to their actual locations. The location of tweets written in English has been the subject of several related works. There are now millions of active Twitter users who tweet many times every day. With its global user base and constant stream of messages, location prediction on Twitter has garnered a lot of attention recently. The suggested approach investigates the big picture of leveraging tweets for location prediction. using machine learning techniques such as naive bayes, support vector machines, and decision trees to deduce the user's location from the tweet's text.