SenseWorth – A Tweets Classifier

Abstract

A lot of people use tweeter to provide their opinions on various topics including sports, politics, finance, etc. Now the information that generally is present on Twitter does not have any means to check whether the data being twitter is correct or not. So in order to check the authenticity of the data,the data must be classified into true and false. Firstly, the dataset would be created using python by devising the code for the same. The code thus designed would be able to extract the tweets with the amount specified by the user. Hence the CSV would be created. After the creation of the CSV, data pre-processing would be applied to the data such that all the unnecessary data such as emojis, words, and information would be removed automatically, and thus we would receive the tweet without any kind of stop-words. This cleaned data would further be tested and trained with the help of machine learning algorithms. These Machine learning algorithms would generally be used to classify the data according to the domain and provide the user with an authenticated answer whether the tweet is true or not along with its accuracy. This helpsthe userto identify the tweet and thus provide an authenticated answer on which informationto believe and to which we should not