Risk Informed Network Threat Response and Analysis

Abstract

With the increasing use of digital platforms for communication, banking, shopping, and government services, the number of online threats has also grown rapidly. Users are frequently exposed to scam messages, phishing links, and misleading content that appear legitimate at first glance. These threats often use urgency, fear, or attractive offers to manipulate users into revealing sensitive information. The major challenge is that most users are not able to clearly distinguish between genuine and malicious content, which leads to a high number of cyber fraud cases. This project presents RINTRA (Risk Informed Network Threat Response and Analysis), a web-based system. The system allows users to input suspicious messages or URLs and evaluates them using a combination of techniques, including keyword-based analysis, statistical pattern detection, dark pattern recognition, and external API-based URL scanning. These methods work together to identify indicators of phishing, financial fraud, identity theft, and manipulative design practices. To provide a more accurate assessment, the system calculates a composite risk score on a scale from 0 to 100 by combining multiple factors such as message content, URL safety, and behavioral patterns. Instead of only presenting technical results, RINTRA also uses an AI-based model to generate a clear and understandable explanation of the detected threat. It explains what is happening, why the content is considered risky, and what actions the user should take, such as avoiding the link or reporting the message. The system is designed with a focus on usability and accessibility, making it suitable for both technical and non-technical users. It bridges the gap between complex cybersecurity tools and everyday user needs by simplifying threat detection and decision-making. While the system performs well in identifying common types of scams and phishing attempts, its effectiveness depends on the quality of input data and external API responses.