Fake Profile Detection on social media platforms using ML
TL;DR
This study proposes a machine learning based approach to detect fake profiles in social media that checks a number of profile attributes automatically to decide whether an account is real or fake, instead of relying on human verification.
Abstract
Social media platforms are one of the most common ways people interact, exchange information and build online communities. Their rapid growth has been accompanied by a dramatic increase in the number of fraudulent user accounts. These fake profiles are frequently created for purposes such as identity theft, online fraud, phishing, spamming, spreading misinformation and influencing public opinion. Such activities threaten user confidence and raise security issues for users and social media companies. Hence the development of an automatic approach to detect fraudulent accounts has become increasingly important. In this study, we propose a machine learning based approach to detect fake profiles in social media. The proposed solution checks a number of profile attributes automatically to decide whether an account is real or fake, instead of relying on human verification. The key features such as number of followers, total number of posts, profile description, age of the account, the availability of the profile photo and other variables related to the account are used as input to the prediction process.