Detection of Propaganda in Information Warfare using Deep Learning

Social media usage has dramatically expanded, which has had a significant impact on the current generation. Online social media platforms are used to disseminate specific propaganda and share information. Because it was created with a specific goal in mind, the news that goes along with a piece of propaganda could be real or fake. It is difficult to manually track every news and determine which reports are true or false. Detecting fake messages is a difficult task because models are required to summarize the messages and compare them to the real messages to classify them as fake networks and deeply structured semantic models. Hence we propose a methodology based on neural network to build a model of propaganda detection. The proposed approach is effective and does not require prior domain knowledge, which is an advantage over other existing approaches. Following dataset training, we achieved an accuracy of 91%. Precision, recall, F1 score and support have been chosen as the performance analysis metrics.