1Asst. Professor, Vasavi College of Engineering, Hyderabad, India
2Asst. Professor, Institute of Advanced Management and Research, Ghaziabad, Uttar Pradesh, India
Unprecedented information distribution throughout human history was made possible with the introduction of the advent of the Internet and the quick uptake of social media platforms such as Facebook and Twitter. Customers are creating and sharing excessively details than they did previously since social media is so widely used, some of which are deceptive with nothing to do with reality. It is difficult to automatically classify a written article as either disinformation or misinformation. Even a subject-matter expert must consider several factors before determining whether an item is true. In this study, we suggest classifying news articles automatically using an ensemble machine learning approach. Our research investigates many textual characteristics that might be applied to differentiate authentic from fraudulent content. Using those characteristics, we train a variety of machine learning approaches using different ensemble techniques and evaluate their performance on four real-world datasets. The higher performance of our suggested ensemble learner technique over individual learners is confirmed by experimental evaluation.
Machine Learning, Deep Learning, Artificial Neural Networks, LSTM (RNN), Deep Learning and Tensorflow
