Detecting fake accounts in social Networks / Sarah Khaled Mostafa ; Supervised Hoda M. O. Mokhtar , Neamat Eltazi

Thesis (M.Sc.) - Cairo University - Faculty of Computers and Information - Department of Information Systems

On-line social Networks (OSNs) have become increasingly popular. People{u2019}s social lives become more associated with those networks. They use on-line social networks (OSNs) to keep in touch with each other, share news, organize events, and even run their own e-business. The rapid growth of OSNs and the massive amount of personal data of their subscribers attract attackers and imposters to steal personal data, share false news, and spread malicious activities. Hence, researchers investigate e{uFB03}cient techniques to detect abnormal activities and fake accounts based on account{u2019}s features and classi{uFB01}cation algorithms. However, some of the accounts features that have been studied and exploited have no impact towards fake accounts detection or even show negative contribution, and using standalone classi{uFB01}cation algorithms does not always achieve satisfactory results. In this thesis, a new algorithm SVM-NN, is proposed to provide e{uFB03}cient detection of fake Twitter accounts and bots, feature selection and dimension reduction techniques are applied. Machine learning classi{uFB01}cation algorithms are used to decide the target accounts identity, and label the account as real or fake. We compare our results with other algorithms that are frequently used for account classi{uFB01}cation including support vector machines, and neural networks (NN). Results show that the proposed algorithm (SVM-NN) uses less number of features, while still correctly classify about 98% of the accounts of the dataset

Issued also as CD