Instructor

Time and Place

Course Information

• Basic course information
• Course syllabus

Textbook

Lecture Slides

• Introduction to Machine Learning (PPT file)
• Inductive Classification (PPT file)
• Decision-Tree Learning (PPT file)
• Ensembles (PPT file)
• Experimental Evaluation (PPT file)
• Computational Learning Theory (PPT file)
• Rule Learning and Inductive Logic Programming (PPT file)
• Neural Network Learning (PPT file)
• Support Vector Machines (PPT file)
• Bayesian Learning: Naive Bayes (PPT file); Other Bayes (PPT file)
• Instance-Based Learning (PPT file)
• Text Categorization (PPT file)
• Clustering (PPT file)
• Natural Language Learning (PPT file)

Assignments and Program Code

The class uses the Weka package of machine learning software in Java. The code for the local version of Weka used in class is in /u/mooney/cs391L-code/weka/. See the guide on running the course version of Weka..

See the instructions on handing in homeworks.

• Homework 1: Active Learning with Version Spaces (due: Sep. 27)
• Homework 2: Transfer Learning with Boosted Decision Trees (due: Oct. 16)
• Homework 3: Computational Learning Theory (due: Oct. 25)
• Homework 4: Rule Learning (due: Nov. 8)

Final Project

• Old Project Resources and Suggestions
• Really Old Project Resources and Suggestions
• Final Report Format. [Talk Version (postscript slides)]
• Sample projects from Fall 2004 that were eventually extended and published at conferences:
  • Brian Kulis, ICML-05
  • Yuk Lai Suen, ECML-05