Real-Time Threat Detection Using Machine Learning and Apache Kafka

Explore real-time threat detection using machine learning and Apache Kafka in this 29-minute podcast episode featuring Géraud Dugé de Bernonville, a Data Consultant at Zenika Bordeaux. Learn about the ZIEM project, a network mapping and intrusion detection platform developed during a Confluent Hackathon. Discover how the team leveraged TensorFlow, Neo4j, and ksqlDB to analyze and visualize network traffic data in real-time. Gain insights into the potential applications of this technology in banking and security sectors. Understand the process of capturing network packets, processing data with ksqlDB, and generating instant network diagrams using Neo4j. Explore the future plans for ZIEM, including more robust visualizations and pattern detection capabilities. Get tips on getting started with TensorFlow and learn about the broader implications of using Kafka for data processing and manipulation in various industries.

- Intro
- What is the Ziem Project?
- How do you use ksqlDB?
- Creating network visualizations with Neo4j and Neovis.js
- Machine learning plans with Ziem
- Supervised vs. non-supervised machine learning
- Future use cases for Ziem
- How to get started with TensorFlow
- It's a wrap!