A Review of Ambiguous News Detection Approaches with Deep Learning, Machine Learning, and Ensemble Paradigms

Abstract

In this modern world, fake and ambiguous news identification and detection is a critical issue in the life of digital and social media. Fake news manipulates the public and gains readership in the wrong sense. Its fast spread and misuse are very harmful to an individual, society, organization, government, and nation. Presently, many automated learning-based detection systems and models have been developed to date. This paper aims to review those existing ambiguous-fake news identification models using deep learning, machine learning, and ensemble learning paradigms. This review compares a large number of such contributions using some key parameters and explores their challenges. Their analytical observations state that most of the works used the Kaggle dataset for the implementation. The accuracy results of DL learning-based systems outperformed the results of both ML-based and ensemble learning-based learning systems.