Feature extraction-based classification model plays a vital role in the news classification process due to large number of features space and data size. However, most of the conventional models have problem in selecting key for fake news classification process. Also, these models have high true negative rate and error rate. In order to overcome these, issues a hybrid feature extraction-based classification model is proposed on the fake news classification problem. Experimental results proved that the proposed fake news classifier has better recall, precision and accuracy than the traditional models.

Fake news, classification, feature extraction, Decision Tree (DT) classification, Random Forest Algorithm and Extra Tree (ET) classification