Hossain, Muhammad IqbalAlam, Md. Golam RabiulSraboni, TasnubaUddin, Md. RifatShahriar, FahimRizon, Ruhit AhmedPolock, Shakib Ibna Shameem2021-09-062021-09-0620212021-06ID 17301017ID 17101016ID 21141084ID 17101050ID 17101435http://hdl.handle.net/10361/14979Cataloged from PDF version of thesis.Includes bibliographical references (pages 33-34).This thesis is submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Computer Science and Engineering, 2021.Information is power although fake information can have severe consequences when it gets viral. Living in the era of social media is like always getting influenced by the news of the online world even though it is fake. Moreover, online news portals and social media are becoming standardized for consuming information. It is effortless to spread fake news using these mediums. Fake news is represented as authentic news with the wrapping of inaccurate information. In recent times, the rate of lynching has increased because of the spread of fake news. Besides, COVID19 related false information is affecting people by creating chaos and spreading panic worldwide. Some fake news automation systems exist to tackle this problem. However, they are largely developed for English. There are hundreds of millions of people who speak Bangla worldwide. In this work, we propose a model that can favorably detect fake news in Bangla. We have applied some pre-processing and feature extraction techniques to our dataset. Experimental analysis on real-world data demonstrates that Passive Aggressive Classifier and Support Vector Machine achieves 93.8% and 93.5% accuracy respectively which are higher than the other Machine Learning classifiers.34 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.Fake NewsBangla Fake NewsMachine LearningNLPTf-IDFPassive Aggressive ClassifierMachine learningThesis