Title
Designing Security User Profiles via Anomaly Detection for User Authentication
Abstract
The ability to detect the anomalous user behavior automatically and create user profiles, storing fresh and accurate security aspect user information, is important for systems administration, security, and development. This paper describes the best utilization of machine learning-based anomaly detection analysis, which is capable of distinguishing data that has security/identification potentials. Thereby, a novel technique for generating dynamic security user profiles is proposed. The real-time analytical outcomes of the anomaly detection methods are encapsulated into structured user profile records. These records store the sudden changing of the user's data, along with the real-time uniquely identifiable users' information. Each record is a unique entity describing a rear users' behavior, which have a substantial influence on user's identity verification. The verification process is in the form of user challenging questions generated from these stored records. The natural of the generated user profiles guarantee that these questions would be chosen such that the security and usability requirements are maintained. "Security" because each question is issued only once to protect the users' responses from being compromised. "Usability" because the data is fresh (real-time data) to help the legitimate user to easily remember and successfully complete the challenge. Real-world scenarios have been given showing the benefits of these challenging questions in building secure knowledge-based user authentication systems.
Year
DOI
Venue
2020
10.1109/ISNCC49221.2020.9297252
2020 International Symposium on Networks, Computers and Communications (ISNCC)
Keywords
DocType
ISBN
Machine Learning,Security User Profiles,Big Data,Anomaly Detection,User Database,User Authentication
Conference
978-1-7281-5629-3
Citations 
PageRank 
References 
2
0.46
2
Authors
2
Name
Order
Citations
PageRank
Iman I. M. Abu Sulayman141.25
Abdelkader H. Ouda2939.26