This paper is published in Volume-4, Issue-3, 2018
Area
Online Security
Author
Vinod Khanapure, Amruta Ayachit, Pooja Gujanatti, Reshma Gajappanavar, Anusha Ashok Shettar
Org/Univ
Angadi Institute of Technology and Management, Savagaon, Karnataka, India
Pub. Date
25 May, 2018
Paper ID
V4I3-1481
Publisher
Keywords
Twitter, Feature extraction, Compromised accounts, Phishing, Online social activities.

Citationsacebook

IEEE
Vinod Khanapure, Amruta Ayachit, Pooja Gujanatti, Reshma Gajappanavar, Anusha Ashok Shettar. A literature survey on detecting compromised accounts on social networks, International Journal of Advance Research, Ideas and Innovations in Technology, www.IJARIIT.com.

APA
Vinod Khanapure, Amruta Ayachit, Pooja Gujanatti, Reshma Gajappanavar, Anusha Ashok Shettar (2018). A literature survey on detecting compromised accounts on social networks. International Journal of Advance Research, Ideas and Innovations in Technology, 4(3) www.IJARIIT.com.

MLA
Vinod Khanapure, Amruta Ayachit, Pooja Gujanatti, Reshma Gajappanavar, Anusha Ashok Shettar. "A literature survey on detecting compromised accounts on social networks." International Journal of Advance Research, Ideas and Innovations in Technology 4.3 (2018). www.IJARIIT.com.

Abstract

Online social networks, such as Facebook, Twitter have become one of the main media to stay in touch with the rest of the world. Over time, social network users build a trust relationship with the account they follow. Unfortunately, when control over an account falls into hands of a cybercriminal, he can easily exploit this trust to further his own malicious agenda. Moreover, the effect of an account compromise can extend well beyond the reputation of a company. This paper presents COMPA, which is designed to detect compromised accounts on online social networks. COMPA is based on a simple observation, social network users develop habit overtime and those habits are fairly stable. Compromised accounts on the online social network are detected with the help of behavioral profile of user and messages sent by him or her. Every time a new post or a message is generated, it is compared to this behavioral profile. If the message significantly deviates from the learned behavioral profile, COMPA flags it as compa. We implemented our approach in a tool called COMPA and evaluated for Twitter and Facebook in this paper. We show that our system is capable of building meaningful behavioral profiles for individual accounts and detect it and send message notification to admin, who is able to block it.