Detection of Malicious Profiles and Protecting Users in Online Social Networks

Everyone today actively uses online social networks to get in touch with their friends, for career opportunities, and business also. Some of the most popular social networks are Facebook, Instagram, LinkedIn, and Twitter. But the question is how much their authentication systems are built in a secured way. The authentication systems mostly depend on the user's general details such as name, photo, and location. In such systems, mischievous persons can easily create fake profiles by cloning the user’s identity to abuse the user’s information. Using fake profiles they can misuse original user’s photos, contacts, and videos. This kind of mischievous person can be identified by the use of privacy detection mechanisms. So, this research emphasizes some data mining techniques for protecting the original user’s information and to identify the malicious accounts in social network sites. By the assessment of 3PS (Publically Privacy Protection System), this work employs the malicious account detection method in OSN depending upon the mischievous person’s uncountable shared posts in a day and latest activity and behaviors. Examining the network similarity and comparison of attributes threshold values referred to the original user’s profile can be used to identify the malicious accounts. For this E_SVM-NN classifier is used based on the feature reduction techniques. This work involves in creating OSN accounts for experiments and investigates the latest updates, posts, comments, photos, and performing online search etc. which are used to evaluate the effectiveness and significance of the proposed work in contrast to the previous works.