Fake News Detection in Social Networks Using Machine Learning and Deep Learning: Performance Evaluation

The problems related to fake news are growing rapidly which results in misleading views on some information. Social media networks are one of the fastest medium to spread information by creating a huge impact on manipulating information by influencing readers in positive and negative aspects. This paper aims at evaluating and comparing different approaches that are used to mitigate this issue including some traditional machine learning approaches, such as Naive Bayes, and the popular deep learning approaches, such as hybrid CNN and RNN. The comparison is not only within traditional methods or within deep learning methods, but also across traditional and non-traditional methods. This paper lays a foundation for selecting a machine learning or deep learning method for problem solving regarding the balance between accuracy and lightweightness.