Statistical Machine Learning 2013

Course Info

• Syllabus     Navigator: C. Andy Tsao.   Office: SE A411.    Campus Tel: 3520
• Lectures: Mon 1510-1700, Thr 1610-1700@ SE A211
• Office Hours:  Mon 1210-1300, Thr 1210-1300 @ SE A411 or by appointment
• Prerequisites: Statistics, Linear Algebra. Knowledge about regression or General/Generalized Linear Models will be helpful. 
• "Official" computing software: R (original, mirrors @ NTU, PU, Packages), Tinn-R, Rstudio (Editors for R)
  
• Textbook: Hastie, Tibshirani and Friedman (2009). The Elements of Statistical Learning: Data Mining, Inference and Prediction. 2nd Edition. Springer-Verlag.

Class slides, Homework, Data, Themes

• Slides: Week1, Week2, Week 3, Naive Regression and LDA, Consistency
• Homework 1 (To be discussed starting 131014)
• Data: MLBenchmark, MLcomp, Mldata, Resources from Science
• Themes: 
• Average Beauty:  
• Chang and Chou's data set & paper, Langlois Lab, Beauty, Sexappeal and Attraction in The Brain
• http://www.psychologicalscience.org/ Search: beauty;
• BeautyCheck, , faceresearch.org, average faces from the world, 

Supplemental

• Applied Linear Models
  
• S4S, Wiki, Bengio's Lecture, Video Lecture: (Marcus Hutter)
• Opencourse of ML: Learning from data (Caltach); 機器學習的基礎 (Coursera)
• Department of ML @CMU
• Interpreting diagnostic tests (ROC curve)
• Fawcett: ROC101, Intro to ROC curve
• Presentation & Slides: Robin William's *RAP, Basic of Design (The Non-Designer's Design Book); Presentation Zen: Examples at Six Minute, in words, more; Technical Presentation: WikiHow, A Guide to Technical Presentations; NTU: 丙紳隨筆, 台大簡報課

Sample R codes and Freewares

• R bloggers
• Cateogrizing variables: BodyFat, data: bfpa.txt)
• Data Spliting (Training/Testing),
• LIBSVM (by Lin, C.J) and guide (NEW)
• Weka:  Website, Download
• C4.5 (Quilan): Building Classification Models: ID3 and C4.5, Tutorial, C4.5-ofai download
• Boosting: R packages Ada, Adabag, Boosting and Bagging

  ---