Fake News Detection Using Ensemble Learning and Machine Learning Algorithms

Abstract

Digital news becomes widely accessible to a large community of users with the advancement of several channels of communication and the progression of technology and thus, contributes to the increase of spreading of fake news. The current study experiments and investigates machine learning models that classify news as either fake or real. Five classifiers were implemented using Random Forest, Support Vector Machine, Gradient Boosting, Logistic Regression, and Naïve Bayes algorithms. Models were trained using merged open-source datasets extracted from online sources covering different domains. Text lemmatization, vectorization, and tokenization were applied to extract useful information from news text and to improve the generalization capabilities and the performance of fake news classification models. The impact of the voting strategy on the performance of ensemble learning models were explored. The performance of the five classifiers was evaluated using the accuracy, the F1-Score, the recall, and the precision. The attained results are promising. The ensemble classifier trained using random forest algorithm and gradient boosting algorithm outperform the other classifiers and thus it might be used effectively against fake news spreading.

Fake News Detection in Different Applications: State-of-the-art and Future Directions

“…Text characterization removes individual text fragments and groups them together. Sanaa elyassami et al [31] presented a research machine learning model that determines if messages are fraudulent or real. Five classi ers were built using Naive Bayes techniques, Logistic Regression, Gradient Boosting, Support Vector Machine, and Random Forest.…”

Disinformation detection using graph neural networks: a survey

“…;Briscoe et al 2014;Giasemidis et al 2016;Elyassami et al 2022;Kini 2022), random forest (RF)(Briscoe et al 2014;Giasemidis et al 2016;Elyassami et al 2022;Sharma et al 2023), decision tree (DT)(Castillo et al 2011;Giasemidis et al 2016;Kishwar and Zafar 2023) and logistic regression (LR)(Giasemidis et al 2016;Elyassami et al 2022;Kishwar and Zafar 2023). 2.…”