Modification on the algorithm provided by the paper: Where is Your Evidence: Improving Fact-Checking by Justification Modeling by Tariq Alhindi, Savvas Petridis and Smaranda Muresan.
We have used the dataset present in this link https://github.com/Tariq60/LIAR-PLUS/tree/master/dataset
We applied GBM and SVM algorithms to classify the news We were able to achieve 71.58% accuracy in binary classification and 43.454 % accuracy in 6-way classification which is more than the current paper. We used unigram features and some tokenization methods with word embeddings formedd from google news word embeddings