This article introduces RELIANCE (Reliable Ensemble Learning for Information and News Credibility Evaluation), an innovative ensemble learning system designed for robust information and fake news credibility evaluation. RELIANCE comprises five diverse base models, including Support Vector Machine (SVM), naïve Bayes, logistic regression, random forest, and Bidirectional Long Short Term Memory Networks (BiLSTMs), which collaborate to minimize the generalization error in predictions.
The key highlights of the study are:
Introducing five distinct and diverse methods for news credibility evaluation, including SVM-based, naïve Bayes-based, logistic regression-based, random forest-based, and BiLSTMs-based models.
Enhancing the credibility evaluation accuracy through ensemble learning by integrating the strengths of the five base models. The ensemble learning approach, using a Multi-Layer Perceptron (MLP) as the meta-model, outperforms the individual base models.
Comprehensive experiments demonstrate the superiority of RELIANCE over individual models and baseline approaches in distinguishing between credible and non-credible information sources.
RELIANCE establishes itself as an effective solution for evaluating the reliability of information sources, with potential real-world applications in empowering users, journalists, and fact-checkers to combat misinformation in the digital era.
To Another Language
from source content
arxiv.org
Key Insights Distilled From
by Majid Rameza... at arxiv.org 04-23-2024
https://arxiv.org/pdf/2401.10940.pdfDeeper Inquiries