staff reporter
Muscat, December 20th
A team at Sultan Qaboos University, led by Dr. Ahmed Shahata from the Faculty of Arts and Social Sciences, has developed an innovative approach to tackling fake news in Arabic media, including social networks.
The team introduced ArabFake, a deep learning algorithm designed to detect misleading news, classify content, and assess the potential risks posed by its spread.
Built on the advanced MARBERTv2 model for multi-dialect Arabic tweets, ArabFake addresses the complexities of the Arabic language while performing three important tasks: identifying fake news, classifying content, and assessing risk.
The algorithm was trained on a validated dataset of 2,495 news items labeled by experts for authenticity and risk, and tested on two large datasets (ANS Corpus and AraNews) containing approximately 200,000 Arabic news articles, both genuine and fabricated.
The results were impressive. ArabFake achieved an accuracy of 94.12 percent in fake news detection, 84.92 percent in content classification, and 88.91 percent in risk assessment, demonstrating its reliability across multiple tasks.
The study also revealed patterns in Arabic fake news. Fabricated articles accounted for 60.4 percent of the dataset, and misleading economic information accounted for 22.4 percent. Nearly two-thirds of fake news is considered highly dangerous to society, highlighting the urgent need for effective detection systems.
Through innovative use of stock recording technology, ArabFake was able to identify linguistic patterns associated with fake news and provide insights into misinformation trends.
The algorithm provides practical opportunities for news organizations, fact-checking efforts, content management systems, and media literacy programs by simultaneously assessing content credibility, estimating risk levels, and prioritizing interventions.
