Sadeque,Farig YousufBiswas, TrishaLamia, Tasmim AfrojShykat, Tarikul IslamRafi, Md. Arifin Ahmed2024-04-232024-04-23©20232023-05ID: 20101628ID: 19301190ID: 19301008ID: 19301009http://hdl.handle.net/10361/22652Cataloged from PDF version of thesis.Includes bibliographical references (pages 58-59).This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2023.Development in information and technology has made the communication easier in the recent decades. Easy access of social media is creating restraints amid of differentiating fake and real news. In the recent period the problem has increased drastically and use of image is making the news more impactful. Even though news websites are publishing the news and provide the source of authentication still there are other portals and platform which intentionally spread fake news to exploit an event. In this paper we proposed a hybrid system where we are combining CNN and RNN to detect fake news . We applied two techniques to reduce the model complexity and increase accuracy based on text data and image. With this system, detecting fake news it’ll stop misleading people and creating an unstable situation as well as taking benefits of the situation.69 pagesenBrac University theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission.Natural language processingMultinomial naive bayesBERTBi-GRUBi-LSTMWord embeddingFake newsSupport vector machineDeep learning (Machine learning)Fake news--Prevention--Data processing.Thesis