Overview
Learn how to build a predictive pipeline for rapid phishing domain detection in this 40-minute conference talk from BSidesCharm 2018. Explore the Certificate Transparency Log Network and its suitability for machine learning applications. Dive into supervised learning basics and the machine learning lifecycle. Discover data acquisition techniques, understand the logistic regression algorithm, and learn how to measure performance metrics. Gain insights into the practical limitations of this approach and access the project on Github for further exploration and implementation.
Syllabus
Introduction
Certificate Transparency Log Network
Suitable for Machine Learning?
Supervised Learning 101
Machine Learning Lifecycle
Data Acquisition: Where To Start
Algorithm: Logistic Regression
Metrics: Measuring Performance
Not a Unicorn / Magic / Silver Bullet
Available on Github