A note for Machine Learning

Reference: Kevin Murphy, Machine Learning A Probability Perspective

http://www.cs.ubc.ca/~murphyk/MLbook/pml-intro-22may12.pdf

• predictive / supervised learning: map x -> y
  • training set D
  • if y categorical : classification, pattern recognition
  • if y real number: regression
• descriptive (unsupervised learning)
  • only have D
  • to find interesting patterns (knowledge discovery)
  • discover clusters
  • latent factors (PCA)
  • matrix completion
• Concepts
  • parametric model vs non-parametric model
  • k nearest neighbor
  • curse of dimensionality

Methods

http://blagrants.blogspot.com/2014/01/machine-learning-with-r-book-review.html

1. Nearest Neighbor
2. naive Bayes
3. Decision Trees
4. Classification Rule Learners
5. Linear Regression
6. Regression Trees
7. Model Trees
8. Neural Networks
9. SVM
10. Association Rules
11. K-means Clustering
12. Random Forest

Check book Machine Learning with R by Brett Lantz