A Survey on Indonesian Hoax Analyzer and Fake News Detection Using Deep Learning Techniques
Jonathan Laimeheriwa, Irene Anindaputri Iswanto, Wiwi Oktriani, Muhammad Fadlan Hidayat · 2024
In today's digital world, where people consume information quickly through social media like Facebook, Instagram, and Twitter, there's a growing concern about fake news. It's crucial to know whether information and its sources are reliable, especially as people may disregard real news if it doesn't align with their beliefs. This survey aims to explore the landscape of fake news in the context of Indonesia, focusing on the development and implementation of hoax analyzers and fake news detection using deep learning techniques. Our review shows that deep learning models such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Transformer models like IndoBERT have demonstrated effectiveness in identifying misinformation within Indonesian content, with promising accuracy and efficiency metrics. However, challenges persist, including the need for extensive data, computational resources, and tailored approaches to handle Indonesian language complexities. This survey highlights the importance of advancing these detection systems to help reduce the spread of misinformation and support informed decision-making in Indonesian society.