Department of Information Science, 2007
INFO 411 - Advanced Knowledge Engineering
COURSE OUTLINE
Objectives
- Acquiring knowledge on advanced methods and tools for knowledge engineering,
including intelligent data analysis, knowledge extraction and representation,
fuzzy systems, neural networks and other machine learning and computational
intelligence methods.
- Obtaining skills for building ready-to-use intelligent information systems to
solve difficult data analysis problems in science, engineering, and business.
Overview
Modern information systems are nowadays required to incorporate more adaptability
and intelligence so as to accomplish various challenging tasks, such as planning,
classification, optimisation, decision support, diagnosis, text mining, and multimedia data
processing etc. In this course, some advanced methods in machine learning and computational
intelligence will be explored. The interrelationship between these methods will be addressed,
and a systems approach of feature extraction, clustering and classification, mixture of experts,
and classification combination will be presented, along with a methodology of constructing
powerful hybrid knowledge engineering systems.
The course is taught with formal and informal lectures and in-class discussion is encouraged.
Students will give presentations based on selected research papers of interest.
A major component of this full-year course is a project work in the second semester,
aimed at developing an intelligent data analysis system for a case study on a real-world knowledge
engineering problem.
Course Delivery
- Time: Full year
- Lectures: Mon. 17:00-17:50, Thu. 11:00-11:50
- Class labs: Thu. 12:00-13:50 at CO3.01
- Presentations: (first semester): 2
- Assignments: (first semester): 2
- Project: second semester
Textbooks
Essential: E. Alpaydin, Introduction to Machine Learning, MIT Press, 2004.
Although the course is built around the specified textbook, a substantial part of the course will
be based on selected research articles from recent literature.
Assessment package
- Internal assessment (60%): presentations - 20%; assignments - 10%; project - 30%
- Exam - 40%
Teaching staff
- Lecturer: Dr. Da Deng - office CO 9.03, ddeng@infoscience.otago.ac.nz, Ext.: 8090
- Teaching Fellow/Tutor: TBA
Schedule
First Semester
| Week | Topic | Assessment |
|---|
| 1 | Introduction; Data types and similarity measures (Ch.1+) |
| 2 | Multi-Dimension Scaling (Ch.6) |
| 3 | Feature selection. Entropy-based and wrapper methods. (Ch.6+) |
| 4 | Clustering algorithms: k-means, GEM, hierarchical, DBSCAN etc. (Ch.7+) |
| 5 | Online neural learning for clustering and visualisation: SOM, GNG etc. | A1 |
| 6 | Parametric methods for classification. (Ch.4,5) |
| 7 | Nonparametric methods for classification (Ch.8+) |
| 8 | Assessment of learners. Meta-learning (Ch.14) |
| 9 | Decision trees. (Ch.9) | P1 |
| 10 | Associative rule extraction: Apriori and FP-Growth |
| 11 | HMM and Mixture of Experts (Ch.12,13) |
| 12 | Combining multiple learners (Ch.15) | A2 |
| 13 | Text mining and multimedia mining | P2 |
Second Semester
| Week | Schedule |
|---|
| 1 | Systems approach in intelligent data analysis; applications case study. |
| 2 | Students Project Presentation: Topic and Literature Review |
| 3-10 | Project works; Weekly meeting discussing project-related issues. |
| 11 | Project Presentation |
| 12 | Project Report due. |
Reading materials
This list is pentative only and is to be extended during the year.
- SOM-based Data Visualisation Methods,
J. Vesanto, 1999.
- Artificial neural networks for feature extraction and multivariate data projection. IEEE Transactions
on Neural Networks,6(2), 296--317 (Mao, J. & Jain, A. K.1995).
- Random
projection in dimensionality reduction: applications to image and
textdata,E. Bingham and H. Mannila, ACM SIGMOD KDD 2001.
- I. Guyon and A. Elisseeff, An introduction to variable and feature selection,
Journal of Machine Learning Research 3 (2003) 1157-1182.
- Consistency-based
search in feature selection, M. Dashand H. Liu, Artificial Intelligence151(2003) 155-176.
- On Clustering
Validation Techniques, M. Halkidi et al., 2001.
- Feature weighting in k-means clustering,
D. Modha and W.S. Spangler, Machine Learning, 52, 217-237,
2003.
- Web
Mining Research: A Survey, R. Kosala, H. Blockeel, ACM SIGKDD
Exploration, 3(1), 2000, pp.1-15.
- W. Ditto, T. Munakata, Principles
and applications of chaotic systems, Communications of ACM,
38(11),pp.96-102, Nov. 1995.
- Symbolic
Interpretation of Artificial Neural Networks (Ismail A. Taha
and Joydeep Ghosh), IEEE Trans. Knowledge and Data Engineering,
11(3),1999,pp.448-463.
- Strategies for unsupervised multimedia processing: self-organizing trees and forests
Kyan, M.; Jarrah, J.; Muneesawang, P.; Ling Guan
Computational Intelligence Magazine, IEEE
Volume 1, Issue 2, May 2006 Page(s): 27 - 40
- Advanced computational models and learning theories for spoken language processing
Nakamura, A.; Watanabe, S.; Hori, T.; McDermott, E.; Katagiri, S.
Computational Intelligence Magazine, IEEE
Volume 1, Issue 2, May 2006 Page(s): 5 - 9
Last modified: 5/7/2006.