Research
Mahdieh Zabihimayvan, Assistant professor of Computer Science, Central Connecticut State University
Research
I work on machine learning for Web systems security, with a concentration in Web systems characterization, measurements, and analytics. My research utilizes prescriptive analysis on massive data sets generated by Web server traffic, collected by crawling on Web pages, or feature extraction of websites structure. My current research is in: (i) evaluation of dark Web content ecosystem; (ii) analyzing the navigational pattern of Web crawlers with focus on designing efficient Web robots for privacy enhancing systems such as Tor; and (iii) utilizing machine learning and evolutionary computing techniques to detect phishing attacks on the Internet. Along this theme, I pursue research projects that utilize proactive cyber threat intelligence analysis to enhance cyber resilience.
I'm looking for research collaborators in machine learning, cybersecurity, and network science. Please email me if you are interested!
Evaluation of dark Web ecosystem
The dark Web defines content on the World Wide Web that cannot or has not yet been indexed by search engines. Tor is the most popular dark network in the world that protects the identity of both content providers and their clients against any tracking on the Internet. This protection makes Tor a good platform for censorship circumvention, free speech, and information dissemination.
Let's work together on evaluating the dark Web ecosystem. The goal is to improve understanding of the services provided on the dark Web and how they interact with each other. Our work will use data science and machine learning techniques for both the information and network analysis.
Web Robot detection
Web robots are employed by modern Web-based technologies and services to collect and scrutinize dynamic content repositories contain. However, aggressive robots with specialized functions may find innovative ways to harvest e-mail addresses, perform click fraud, and access the information only authorized human users should be allowed to access.
Let's see how we can detect such malicious Web robots using machine learning methods and help Web servers have a better performance and service!
phishing attack detection
Phishing as one of the most well-known cybercrime activities is a deception of online users to steal their personal or confidential information by impersonating a legitimate website. Several machine learning-based strategies have been proposed to detect phishing websites. These techniques work on features that are extracted from Web pages.
Let's work together on such features to see how we can differentiate phishing websites. We will also learn how to enhance phishing attack detection using different machine learning techniques.
Recent publications:
A Daliri, M Alimoradi, M Zabihimayvan, R Sadeghi (2024), World Hyper-Heuristic: A novel reinforcement learning approach for dynamic exploration and exploitation, Expert Systems with Applications (244, 122931).
B. McConnell, D. Del Monaco, M. Zabihimayvan, F. Abdollahzadeh, S. Hamada (2023), Phishing Attack Detection: An Improved Performance Through Ensemble Learning. In International Conference on Artificial Intelligence and Soft Computing, pp. 145-157, Cham: Springer Nature Switzerland.
M. Alimoradi, A. Daliri, M. Zabihimayvan, R. Sadeghi (2024), Statistic Deviation Mode Balancer (SDMB): A novel sampling algorithm for imbalanced data.
R. Sledzik, M. Zabihimayvan (2022), Focal Loss Improves Performance of High-Sensitivity C-Reactive Protein Imbalanced Classification, In the 35th International Symposium on Computer Based Medical Systems. IEEE (Accepted).
M. Alimoradi, M. Zabihimayvan, A. Daliri, R. Sledzik, and R. Sadeghi (2022), Deep Neural Classification of Darknet Traffic, In the 24th International Conference of the Catalan Association for Artificial Intelligence (Accepted).
M. Zabihimayvan, and D. Doran (2022), A first look at references from the dark to the surface web world: a case study in Tor. International Journal of Information Security (pp. 1-17), Springer.
M. Zabihimayvan , R. Sadeghi, D. Kadariya, D. Doran (2020), Interaction of Structure and Information on Tor. In International Conference on Complex Networks and Their Applications, pp. 296-307. Springer.
W. Romine, N. Schroeder, J. Graft, F. Yang, R. Sadeghi, M. Zabihimayvan, D. Kadariya, and T. Banerjee (2020), Using Machine Learning to Train a Wearable Device for Measuring Students’ Cognitive Load during Problem-Solving Activities Based on Electrodermal Activity, Body Temperature, and Heart Rate: Development of a Cognitive Load Tracker for Both Personal and Classroom Use, Sensors, no. 17: 4833.
M. Zabihimayvan, R. Sadeghi, D. Doran, M. Allahyari (2019), A Broad Evaluation of the Tor English Content Ecosystem, Proceedings of the 10th Conference on Web Science, pp 333-342, ACM.
M. Zabihimayvan, R. Sadeghi, N. Rude, D. Doran (2017), A Soft Computing Approach for Benign and Malicious Web Robot Detection, Expert Systems with applications (pp 129-140), Elsevier .
J. Hamidzadeh, M. Zabihimayvan, R. Sadeghi (2017), "Detection of Web site visitors based on fuzzy rough sets", Soft Computing (pp 2175-2188), Springer.
M. Zabihimayvan, D. Doran (2018), Some (Non-)Universal Features of Web Robot Traffic, In the 52th Annual Conference on Information Sciences and Systems, IEEE.
M. Zabihimayvan and D. Doran (2019), Fuzzy Rough Set Feature Selection to Enhance Phishing Attack Detection, In International Conference on Fuzzy Systems, pp. 1-6.
If you are interested in applied machine learning and data science techniques, network science, and their applications in cybersecurity, please feel free to contact me for more collaboration opportunities!