Laboratories and exercises
Laboratories and exercises give students an opportunity to carry out experiments that illustrate topics in a realistic setting and at the same time learn the specifics of the software used. Students may also be assigned to work on projects too large to be completed during a single class period. Laboratories can provide time for independent project work and programming assignments with reporting similar to that done in other topics in computer science.
The lab projects can be categorized into several categories, and more innovative ideas and suggestions are encouraged.
- Learn to use data mining systems by using some data mining and data warehousing softwares. Typical such softwares may include Microsoft SQLServer 2005 (Analysis manager), Oracle 10g (data mining part), IBM Intelligent-Miner, and statistics analysis software tools.
- Implement some data mining functions, including association mining, classification, clustering, sequential pattern mining, text-mining, Web mining, bio-mining, spatial data mining packages. Some open or partially open source data mining systems, such as Weka, IlliMine, and so on can be used for data mining algorithm extension and data mining application exploration.
- Implementation, refinement, and performance comparison of several different data mining methods.
- Proposal, implementation and testing of new data mining algorithms and functions.
- Using some sample data sets to implement and test data mining functions, such as KDD CUP data sets, UC-Irvine Machine Learning/KDD Repository, DBLP database, and other selected Web data sets.