Fake Profile Identification Using Machine Learning

Associate Professor. G Swathi, R Vaishnavi, Shaik Noorus Sabiha, P Rakesh Anand, P Nithish Kumar

Abstract- – The abstract highlights the enormous user engagement of social net-working sites like Twitter and Facebook and explores their substantial influence on modern digital life. It also highlights the influence that user interactions have on day-to-day life and the growing problem of spammers creating phony personas to disseminate unwanted information. The emphasis is the need for more effective techniques to identify and block phony social media profiles and material. The abstract recognizes the limitations of current machine learning-based techniques, pointing to low accuracy and difficult security maintenance. The suggested method uses supervised learning, especially Extreme Gradient Boosting (Xg boost), to distinguish between authentic and fraudulent profiles more accurately. Furthermore, a web page with a WSGI server is built to help recognize and flag these bogus profiles. Our proposed method gives an accuracy of 99%. The urgent problem of phony profiles and material on social networking sites is being ad-dressed by this strategy.

DOI: /10.61463/ijset.vol.11.issue6.107