Machine Learning Primer Feature Selection and Feature Engineering

This will help learn how tech companies like Meta, Twitter, and Airbnb design their feature selection and feature engineering to serve billions of users every day

We will cover the following

1. One hot encoding

• Common problems
• Best practices
• One hot encoding in tech companies

2. Feature hashing

• Benefits
• Feature hashing example
• Feature hashing in tech companies

3. Crossed feature

• Crossed feature in tech companies

4. Embedding

• Benefits
• How to generate/learn embedding vector?
• Embedding in tech companies

5. Numeric features

• Normalization
• Standardization

for full course for system design

1. One hot encoding

One hot encoding is a very common technique in feature engineering. It converts categorical variables into a one-hot numeric array.