Detection of Fake News Using Clustering Algorithms

Due to its potential for causing considerable social and national harm, fake news is a growing problem in today's media landscape, especially on social media. Previously, it has been the focus of much research. A supervised machine learning method for classifying fake news as genuine or false is developed utilising Python scikit-learn and natural language processing (NLP) for textual analysis, as detailed in this paper's research on the identification of fake news. Python scikit-learn module contains utilities like Count Vectorizer and Tiff Vectorizer that may help with text data tokenization and feature extraction. Based on the findings from the confusion matrix, we will use feature selection methods to examine and pick the most accurate features to research and select the best features. Internet and social media have made it easier for anyone to get their hands on a wealth of information. While these tools have made communication and information flow simpler and quicker, they have also threatened the authenticity of the news that is being disseminated. Fake news has had such an effect on society that it even influenced the 2016 USA presidential election. In our model, we compare different models to find out which model is providing highest accuracy for detecting fake news. It can be done by using the sklearn module.