AI-Enhanced Identity Threat Detection and Automated Response: An AITIR Framework for Optimized Cybersecurity Operations

Md Sazzad Hossain; Kiran Kumar Parvatha Reddy

doi:10.71097/IJSAT.v15.i2.11300

AI-Enhanced Identity Threat Detection and Automated Response: An AITIR Framework for Optimized Cybersecurity Operations

Author(s)	Mr. Md Sazzad Hossain, Mr. Kiran Kumar Parvatha Reddy
Country	United States
Abstract	The rapid proliferation of identity-targeted assaults in contemporary corporate contexts demands intelligent real-time detection and response systems that exceed the constraints of conventional rule-driven security frameworks. This paper presents the AITIR (AI-Assisted Identity Threat Detection and Automated Response) framework, which constitutes a novel six-layer pipeline engineered to transform heterogeneous identity data such as Active Directory events, MFA logs, PAM records, and VPN sessions into actionable security outcomes. The preprocessing layer normalizes raw log streams and extracts structured feature vectors, which are then fed into a tripartite AI analytics engine. We propose a hybrid architecture that joins unsupervised anomaly detection via an Isolation Forest algorithm with a Long Short-Term Memory (LSTM) network for sequential behavioral modeling and a supervised XGBoost classifier for threat categorization. A composite Identity Risk Score is generated by merging the outputs of these three models as a weighted aggregation of individual confidences, and this score is evaluated against a dynamic threshold to flag confirmed threats. This score subsequently drives threat intelligence correlation against the MITRE ATT&CK framework and triggers automated mitigation actions such as session termination or mandatory re-authentication. A central achievement of our research is the eradication of human involvement in security operations centers via closed-loop continuous learning, where analyst input and response results retrain the underlying models to adjust to shifting attack patterns. We validate the framework’s efficacy via experimental evaluation on real-world identity log datasets, and the results show marked improvements in detection accuracy and response latency relative to existing baseline methods. Consequently, the AITIR framework presents a scalable and adaptive resolution for identity threat management within intricate organizational networks.
Keywords	Identity Threat Detection, Identity and Access Management, AI-Assisted Cybersecurity, Identity Threat Intelligence, Automated Incident Response, Identity Risk Scoring, Behavioral Analytics, Anomaly Detection, Isolation Forest, LSTM, XGBoost, Machine Learning for Cybersecurity, Identity Security, MITRE ATT&CK, Active Directory Security, MFA Security, PAM Security, Real-Time Threat Detection, Security Automation, Enterprise Cybersecurity.
Field	Computer > Network / Security
Published In	Volume 15, Issue 2, April-June 2024
Published On	2024-06-06
DOI	https://doi.org/10.71097/IJSAT.v15.i2.11300

View / Download PDF File

About IJSAT Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics	Join as a Reviewer Editors & Reviewers Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us	Message on WhatsApp	+91-9687-182-185	editor@ijsat.org

International Journal on Science and Technology

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

AI-Enhanced Identity Threat Detection and Automated Response: An AITIR Framework for Optimized Cybersecurity Operations

Share this