Abstract: The majority of people these days prefer to read the news online via social media. A multitude of websites disseminate news and offer the source of verification. How to validate news and articles shared on social media platforms like Facebook Pages, Twitter, WhatsApp groups, and other microblogs and social networking sites is the question. It is detrimental to society for people to trust rumors and pass them off as news. It’s vital to put an end to the rumors and concentrate on accurate, verified news reports. This project’s goal is to create two models that use machine learning (SVM) and deep learning (LST) algorithms, respectively, to identify bogus news. Utilizing both machine learning and deep learning, an attempt is made to aggregate news and then use Support Vector Machine and Long Short-Term model to determine whether the news is real or fake. First, the dataset is cleaned and pre-processed; next, feature extraction techniques are applied to the pre-processed data, and the model is trained using both algorithms independently to obtain two distinct models using SVM and LSTM, respectively. Confusion Matrix, Classification reports, and accuracies of both models are calculated and compared in order to identify the best model for Fake News Detection. The accuracy of the LSTM classifier was 99.54%, while the SVM classifier yielded 99.29%. While both algorithms produced results with acceptable accuracy, LSTM performed better in classifying the news articles than SVM. This project’s main contribution is comparing the accuracies of SVM and LSTM algorithms to determine which algorithm and techniques match the problem of fake news detection the best.