EAGLE AI NETWORK TRAFFIC ANALYZER FYP Team Member 1 Name: Asif Mujeeb Roll No. 19I-1991 Member 2 Name: Malik Touseef Husnain Roll No. 19I-2028 Supervisor: Sir Jawad Hassan Co-Supervisor: Sir Shoaib Saleem Khattak S INTRODUCTION • Cyber attacks, including Distributed Denial of Service (DDoS) and crypto jacking, are escalating daily • We introduce an Network Traffic Analyzer with AutoML to detect attacks in short time 7/1/20XX EAGLEAI 2 MOTIVATION & CASE STUDY 7/1/20XX EAGLEAI 3 BACKGROUND In 2017, Equifax, one of the largest credit agencies in the U.S., experienced a massive data breach. Hackers exploited a vulnerability in the Apache Struts framework, compromising the personal information of approximately 147 million people. The breach was exacerbated by delayed detection and response, allowing unauthorized access to sensitive data for over two months. 7/1/20XX EAGLEAI 4 COMPARATIVE ANALYSIS Without Our Solution With Our Solution The delayed detection led to significant financial losses, legal repercussions, and damage to Equifax's reputation. 7/1/20XX The hypothetical early detection and alerting system would have enabled a swift containment response, minimizing the breach's impact and preserving customer trust. EAGLEAI 5 PROBLEM STATEMENT Traditional threat detection systems struggle with the volume, variety, and evolving nature of cyber threats. This leads to high false positives and missed detections, compromising security. 7/1/20XX EAGLEAI 6 SOLUTION Our solution provides a faster Network Analyzer based on AI algorithms. Which is a more accurate, scalable and evolving solution for the Network. Making it a classification type solution. 7/1/20XX Pitch deck title 7 SCOPE Network Based Analysis Machine Learning Algorithms Lambda Architecture Automatic Model Selection Data Processing Frameworks User-Friendly Frontend . 7/1/20XXs EAGLEAI 8 OBJECTIVES Scalable Accurate Automation Self-Learning 7/1/20XX Pitch deck title 9 HIGH-END FEATURES 7/1/20XX Real-Time Threat Detection AutoML for Dynamic Model Selection Scalable Lambda Architecture Advanced Anomaly Detection User-Friendly Interface Self-Optimizing System Comprehensive Data Analysis Customizable Alert System Pitch deck title 10 MODEL DIAGRAM 7/1/20XX EAGLEAI 11 TECHNOLOGIES USED TensorFlow 7/1/20XX Google’s AutoML MongoDB Pitch deck title Apache Spark 12 TECHNOLOGIES USED Python 7/1/20XX Pitch deck title 13 HYPOTHETICAL IMPLEMENTATION Application Outcome Had the AutoML Network Traffic Analyzer been deployed within Equifax's network infrastructure, its advanced machine learning algorithms and real-time processing capabilities could have identified unusual data access patterns early on. The system's ability to rapidly detect and alert on such anomalies could have significantly reduced the breach's window, potentially preventing the extensive exfiltration of personal data. 7/1/20XX EAGLEAI 14 TIMELINE EagleAi 2024 February March May September Capturing Packets Training ML Implementing AutoML Data Processing Implementing ML GUI Implementation December Reporting, Testing & Validation Feature Extracting for ML 7/1/20XX EAGLEAI 15 WORK DISTRIBUTION 7/1/20XX Asif Mujeeb Malik Touseef Husnain Capturing Packets Data Processing Implementing AutoML GUI Implementation Feature Extraction for ML Training ML Implementing ML Pitch deck title 16 CONCLUSION Our AutoML Network Traffic Analyzer represents a significant leap forward in cybersecurity, offering a smarter, more adaptable solution to the ever-growing challenge of cyber threats. We welcome any questions. 7/1/20XX EAGLEAI 17 THANK YOU 7/1/20XX EAGLEAI 18