This work is an implementation of the following paper: Intagram Fake and Automated Account Detection
The dataset contains of 2 set of json files with given features:
user_media_count
- Total number of posts, an account has.user_follower_count
- Total number of followers, an account has.user_following_count
- Total number of followings, an account has.user_has_profil_pic
- Whether an account has a profil picture, or not.user_is_private
- Whether an account is a private profile, or not.user_biography_length
- Number of characters present in account biography.username_length
- Number of characters present in account username.username_digit_count
- Number of digits present in account username.is_fake
- True, if account is a spam/fake account, False otherwise