Machine Learning Basics

What You'll Learn

• What is Machine Learning?
• The ML Workflow
• Key Algorithms (Decision Trees, Random Forests, K-Means)
• Model Evaluation
• Bias and Variance

What is Machine Learning?

Machine Learning (ML) is the field of study that gives computers the ability to learn without being explicitly programmed.

Traditional Programming: Input + Rules -> Output

Machine Learning: Input + Output -> Rules

The ML Workflow

1. Data Collection: Get the data.
2. Data Preparation: Clean, format, and split data.
3. Model Training: Feed data to the algorithm.
4. Model Evaluation: Test how well it learned.
5. Tuning: Adjust settings (hyperparameters) to improve.
6. Prediction: Use the model on new data.

Key Algorithms

1. Decision Trees (Supervised):

• Like a flowchart of "if-then" rules.
• Easy to interpret.
• Prone to overfitting.

2. Random Forests (Supervised):

• A collection (ensemble) of many decision trees.
• More accurate and robust than single trees.
• "Wisdom of the crowd."

3. K-Means Clustering (Unsupervised):

• Groups data points into K clusters based on similarity.
• Used for customer segmentation.

Model Evaluation

How do we know if our model is good?

For Regression (Numbers):

• MAE (Mean Absolute Error): Average error.
• RMSE (Root Mean Squared Error): Penalizes large errors more.

For Classification (Categories):

• Accuracy: % correct.
• Precision: Of those predicted positive, how many were actually positive?
• Recall: Of those actually positive, how many did we find?
• F1 Score: Harmonic mean of Precision and Recall.

Bias vs Variance Trade-off

• Bias: Error from erroneous assumptions (Underfitting).
• Variance: Error from sensitivity to small fluctuations in the training set (Overfitting).
• Goal: Find the sweet spot (low bias, low variance).

Next Steps

Ready for the final challenge? Let's apply everything you've learned in the Mini-Project!