- Lecturer
- Murata Tsuyoshi
- Place
- Mon5-6(W833)
- Credits
- Lecture2 Exercise0 Experiment0
- Code
- 76017
- Syllabus updated
- 2010/9/20
- Lecture notes updated
- 2011/1/17
- Access Index
-
- Semester
- Fall Semester
Outline of lecture
This lecture covers many of the important techniques of machine learning (ML) and data mining (DM), such as decision tree learning, classification rule learning, association rule learning, Bayesian learning, regressions, instance-based learning, and clustering.
Purpose of lecture
The goal of this lecture is to learn basic data mining methods using Weka (a data mining tool).
Plan of lecture
1. Machine Learning and Data Mining, Weka
2. Concept descriptions, applications
3. Concept space, bias
4. Input data format, classification, association
5. Clustering, numerical prediction
6. Types of attributes and their transformations
7. Knowledge representation, decision trees, classification rules
8. Association rules, instance-based representation, clusters
9. Basic learning algorithms, Naive Bayes
10. Decision trees, information gain, gain ratio
11. Covering algorithms, rules and decision trees
12. Evaluation of learning, cross validation
13. t-statistic, minimum description length
14. Summary
Textbook and reference
Data Mining: Practical Machine Learning Tools and Techniques (Second Edition)
I. H. Witten, E. Frank, Morgan Kaufmann, 2005.
http://www.cs.waikato.ac.nz/~ml/weka/book.html
Related and/or prerequisite courses
Advanced Artificial Intelligence
Evaluation
Based on the assignments you submit
Supplement
Please visit the following site for more information about this lecture.
http://www.ai.cs.titech.ac.jp/lecture/ml/
