AI & ML
Protecting Web Applications using a Multimodal AI-based System for Classifying Web Injection Attacks
arXiv cs.CR0 views
arXiv:2611.23660v1 Announce Type: new Abstract: Web applications have been essential to accessing and managing online services. Made available with weaknesses, much due to their rapid development and human error, they are exposed to frequent web injection attacks, leaving security analysts overwhelmed with threats and alerts that often lack critical details for proper risk assessment. Detecting these attacks requires inspecting the payload of network traffic packets, which becomes inaccessible when traffic is encrypted via HTTPS, as encryption hides the content of the traffic. This paper introduces a Multimodal Artificial Intelligence (AI) system to protect online services that rely on web application security by detecting and classifying web injection attacks, and provides security analysts with information to support attack prevention and incident response. The system employs unsupervised Machine Learning (ML) to flag potentially anomalous Network Flows (NetFlows). The corresponding data recorded in Web Server Logs (WebLogs) is then analysed using a heuristic framework based on Natural Language Processing (NLP) and supervised ML to detect and classify web injection attacks. Finally, the resulting classifications are scrutinised using a Similarity Search Engine (SSE) based on Locality Sensitive Hashing (LSH) to validate them. The solution handles encrypted traffic by analysing NetFlows and WebLogs, the latter in their natural, unencrypted form, by leveraging NetFlow-level metadata for anomaly detection and analysing web server logs after TLS termination, thereby avoiding any additional in-network traffic decryption. It was evaluated in the classification of eight web injection attack classes. The results demonstrated its ability to accurately identify attacks and their classes, achieving an F_1 score of 91% and an accuracy of 95%, and offering valuable insights for alert prioritisation.
Source:https://arxiv.org/abs/2611.23660PreviousA-Tag: Cryptographic File Tagging with Enforced Live Human PresenceNextFlowSH: Detecting Web Attacks by Correlating Anomalous Network Flows and Endpoint Logs
Related posts
How to detect user frustration in chatbot conversations
0dev.to
Mình Đã Biến Gemini Thành Trợ Lý Cá Nhân Như Thế Nào (Lịch, Drive, Gmail và Những Lần Suýt Toang)
0dev.to
Im looking fr any free ai models / tools are there?, leaVE UR Lights💡
0dev.to
I checked 10 SEO agency websites for AI-search readiness. The real bugs were the boring ones.
0dev.to

How much GPU memory does an LLM need? The whole sum, worked out on real cards
0dev.to

Routing Upscaling Jobs by Output Size, Not Input Size
0dev.to