Data Science Tools and Techniques for the Blue Team

Explore data science tools and techniques for cybersecurity in this 40-minute conference talk from NorthSec. Delve into the data scientist methodology and statistical and machine learning techniques available to defenders of corporate infrastructure. Learn about the strengths and weaknesses of different approaches, analyze real data, and understand how to scale Python code from hundreds of thousands to tens of millions of data points. Discover how to make sense of increasing volumes of data from various log sources in corporate environments, going beyond traditional SIEM capabilities. Gain insights into modern data science, feature engineering, model training, and unsupervised learning techniques. Includes practical examples, code explanations, and discussions on measuring similarity and interpreting results.

Intro
What is Data Science
MODERN DATA SCIENTIST
Why You Should Care
Warning
DBIR 2015
Academia
Our Methodology
Docker
Let's Look at the Data
Vectors, Features & The Curse Of Dimensionality
Train, Test, Cross Validate
Experiment
Model Persistence
One More Example Unsupervised Learning
Measuring Similarity
Does it Make Sense?
Take A Way