Machine Learning is the branch of data science where algorithms are used to learn from historical data and is used to predict on future/unseen data. It impacts our daily lives in a lot of ways : The recommended links we get when we use the search engine, making portfolio decisions for the financial market, identifying different kinds of prospective customers etc.

This workshop will take a hands-on approach. The following topics will be covered

Basics

• What is Machine Learning

• Different kinds of Machine Learning

  • Supervised Learning
  • Unsupervised Learning
  • Reinforcement Learning

• Some types of ML problems

  • Classification
  • Regression
  • Dimensionality reduction
  • Clustering

Introduction to scikit-learn

Introductory Model - Linear Models & Model Evaluation

• Linear Regression
• Logistic Regression
• Model evaluation
  • Accuracy metrics (ROC, AUC, Precision, Recall F-score)

Handling overfitting

• What is overfitting?
• Motivation to bias-variance
• Cross-validation
• Regularization (high-level concept - no math here)

Tree-based model

• Decision Tree
• Ensemble models
• Bagging and Boosting
• Random Forest
• Gradient Boosting Machines

The repo for the workshop will be: https://github.com/rouseguy/intro-to-machine-learning