Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{195146,
        author = {Kalaiselvi S and Subi s and Karthika S and Kalpana devi P},
        title = {AI-Based Cyber Threat Detection Using Isolation Forest},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {10},
        pages = {7397-7401},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=195146},
        abstract = {The growing complexity of cyber threats demands intelligent, scalable detection systems capable of identifying both known and novel attack patterns in real time. This paper presents an AI-driven cyber threat detection system built on the Isolation Forest (IF) algorithm an unsupervised anomaly detection technique that requires no pre-labeled training data. The system ingests raw network traffic in CSV format, extracts numeric features, computes anomaly scores, and classifies each record into one of three interpretable threat tiers: High-Risk Anomaly, Mild Suspicious Behavior, and Normal Traffic. Deployed as an interactive Streamlit web application on Streamlit Community Cloud, the system is accessible to security analysts without specialized infrastructure. Evaluation on the KDD Cup 1999 and CICIDS 2017 benchmark datasets yields a precision of 94.7%, recall of 92.3%, F1-score of 93.5%, accuracy of 95.1%, and sub-millisecond per-sample inference times. The paper presents the full system architecture, algorithmic design, experimental results, comparative analysis, and future research directions.},
        keywords = {Cyber Threat Detection; Isolation Forest; Anomaly Detection; Intrusion Detection System; Unsupervised Learning; Network Security; Streamlit; Zero-Day Attacks; Real-Time Analytics},
        month = {March},
        }