Supervised Learning

• Regression: when the variable to predict is numerical
  • using age to predict income
• Classification: when the variable to predict is categorical
  • using age to predicate a category
• Algorithms:
  • Linear Regression
  • Logistic Regression
  • Decision Trees
  • Random Forest
  • Gradient Boosting
  • Artificial neural networks

Unsupervised Learning

• Used to identify structure/pattern in data
• Algorithms:
  • K-Means
  • Clustering

Basic Steps for Machine Learning

• Collect and prepare data
• Choose a training model or algorithm
• Evaluate a model
• Hyperparameter tune
• Make predictions

Linear Regression

Goal: Find the best Fit Line equation that can predict the values based on the independent variables

1. Simple Linear Regression (one independent feature)
   1. One independent variable and dependent variable
   2. $y = β_0 +β_1X$
      1. y = dependent variable
      2. X = independent variable
      3. β_0 = intercept
      4. β_1 = slope
2. Multiple Linear Regression (more than one feature)
   1. 1+ independent variable and 1 dependent variable
   2. $y = β_0 +β_1X +β_2X+...+β_nX$
      1. y = dependent variable
      2. X…X_p = independent variable
      3. β_0 = intercept
      4. β_1… β_n = slope
3. Univariate Linear Regression (one dependent variable)