![]() | Only 14 pages are availabe for public view |
Abstract In this thesis, we categorized the types of Spam found on Pinterest. We extensively have studied the problem of fake accounts and drawn some characteristics of such accounts. We have been able to create two systems. The first was built for crawling data from the Pinterest website in order to build our dataset as Pinterest does not provide any APIs (Application Program Interface) for developers till the time our dataset has been constructed, and the other is a detection system that classifies an account of being either a fake account or a real user account. After running our system using 886444 pins in 3920 accounts, it found that 1503 are fake accounts containing 345000 pins. We tested the system using four different test data sets and the result came with an average of 90.25% precision of classification. Although it missed some fake accounts, but still produces zero false positives. |