Social Media Fake Information Identification Method Based on LSTM

Abstract

False information is spreading like wildfire on social media platforms, causing unnecessary confusion and mistrust. Most people do not have the ability to judge for themselves whether information is true or false that is why it is important to use machine learning models to help people identify false information on the web. LSTM model is a good choice for dealing with identifying social media disinformation. The key steps in identifying social media disinformation using LSTM include data preparation and preprocessing, building a model structure adapted to LSTM, model training and evaluation, and subsequent hyper-parameter tuning and model deployment. These steps constitute a complete process that enables LSTM to effectively identify false information and provide credible prediction results through the processing of text data and model training. After model training and validation, LSTM has an accuracy of 99%, AUC of 100% and successfully identifies false information on social media, providing reliable classification results for both true and false information.