Combining ChatGPT and Recurrent Neural Network to Identify Mobile Misinformation
Abstract The misinformation problem has plagued people for a long time and causes serious trouble. Though there is no method that can solve the problem satisfactorily so far, each method has its own advantages. At the same time, the popular ChatGPT (Chat Generative Pre-trained Transformer) is used extensively and intensively in these days. Misinformation detection could be one of the functions provided by ChatGPT because both of them (misinformation detection and ChatGPT) are related to natural language processing (NLP). However, like other methods, ChatGPT has its own advantages and disadvantages when used for misinformation detection. This research first investigates the effectiveness of the application of mobile misinformation detection using ChatGPT. Subsequently, it combines the predictions from ChatGPT and RNN (recurrent neural network), and sees whether better results will be produced by using this meta approach.

Keywords
Security, misinformation, misinformation identification, ChatGPT, Chat Generative Pre-trained Transformer, RNN, recurrent neural network, NLP, natural language processing

References
GPT models

Conference

International Conference on Recent Advances in Computer Science and Information Technology (ICRACSIT 2023), Houston, Texas, September 03, 2023.