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{196251,
        author = {Sk. Sameerunnisa and Vellaturi Lakshmi Subhashini and Perli Karthik and Panitapu Venkat Rishi and Mullapudi Koushik},
        title = {Web Vulnerability Detection Using ML Techniques},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {4973-4983},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=196251},
        abstract = {Web application security testing platform which was created to focus on all key elements of modern web system environment and to help penetration testers identify smartly the threats arising from the web application. The stack is implemented in Python, and it includes FastAPI as the HTTP server and LightGBM/Isolation Forest for the machine learning bits, with Nmap feeding into port scanning, and DNS brute-forcing (lookout for subdomains) to get started. The front-end is built in React/TypeScript, and the entire platform runs on Render in Docker, making it easy to scale up and deploy updates. Most of the current vulnerability scanners are not able to detect important problems as they adopt primarily signature-based detection and do not involve real-world introspection. They also have a tendency to produce too many false positives, and they do not typically maintain a history of scans which means that it is difficult for organizations to track their progress. This work addresses these limitations by integrating machine learning techniques with reconnaissance modules maintaining a persistent scan history, and providing an AI assistant to answer questions about payloads, URLs and code snippets. This method not only enhances the coverage of detection, but also facilitates faster and more comprehensible remediation for users. 
This platform’s main strength is that it can find a wide range of vulnerabilities, from common to rare, sometimes which traditional tools often fail to identify. Plugging in port scanning, subdomain enumeration and code analysis delivers a more holistic assessment. The open-source and modular design means that other people are free to adapt and improve it, while the deployment on Render making it available for real-world use. It details how the system can be applicable at the local level through a case study in
Hyderabad and what it means to organizations, given that a vulnerability assessment process is both extensive and easy-to-use.},
        keywords = {Web Security, Vulnerability Assessment, Machine Learning, LightGBM, Isolation Forest, Reconnaissance, Port Scanning, Subdomain Discovery, Explainable AI, AI Assistant, CSRF Detection, Zero-Day Flagging},
        month = {April},
        }