The recent growth in technology has led to an enormous amount of data. As a part of this rising technology, different social media platforms are introduced for people to share their insights. Every day, millions of fresh pieces of material are shared on social media platforms like Facebook, Instagram, and Twitter, and these platforms are adding new users on a regular basis. As a consequence, there are also many fake/bot accounts created to gain high traffic and misleading information. Hence, we propose a research work to verify the credibility of the user, whether they are legitimate users or not using machine learning techniques. Social media platforms have some common features like profile name, profile picture, follower count, following count, account created, external URLs, etc., which can be used to understand the authenticity of a user. We apply classification techniques like random forest, logistic regression classifier to analyze the pattern on public datasets take from Kaggle. Compared to other classifiers, experiments on the data using logistic regression with grid search obtained 96.5% of accuracy in detecting the fake accounts.