CMT108: Pattern Recognition and Data Mining
| School | Cardiff School of Computer Science and Informatics |
| Department Code | COMSC |
| Module Code | CMT108 |
| External Subject Code | 100358 |
| Number of Credits | 20 |
| Level | L7 |
| Language of Delivery | English |
| Module Leader | Dr Jianhua Shao |
| Semester | Spring Semester |
| Academic Year | 2023/4 |
Outline Description of Module
This module introduces some classic and modern Pattern Recognition, Machine Learning and Data Mining techniques and acquaints students with the application of these techniques.
On completion of the module a student should be able to
1. Understand the power and limitations of present pattern recognition, machine learning and data mining algorithms and techniques.
2. Demonstrate an understanding of data properties and data preparation necessary for machine learning and data mining.
3. Understand how software tools such as WEKA and R can be used to solve practical problems.
4. Demonstrate an understanding of the various techniques available and be able to justify their selection.
5. Understand how these techniques can be applied to application areas such as Bioinformatics.
How the module will be delivered
The module employs a combination of theoretical and practical interactive contact sessions, using information from a selection of papers and textbooks. Students will be expected to undertake some directed self-studies.
Skills that will be practised and developed
Please see Learning Outcomes.
How the module will be assessed
There will be three assessments for this module.
Two courseworks will allow the student to demonstrate their knowledge and practical skills and to apply the principles taught in lectures.
Coursework 1 will be assessing the knowledge and understanding in the area of data mining (LO1, LO2, LO3)
Coursework 2 will be assessing the knowledge and understanding in the area of machine learning (LO1, LO2, LO3)
Exam: A written exam (2 h) will test the student's knowledge and understanding as elaborated under the learning outcomes. It will cover all learning outcomes.
The potential for reassessment in this module is a 100% resit examination during the summer.
Assessment Breakdown
| Type | % | Title | Duration(hrs) |
|---|---|---|---|
| Written Assessment | 70 | Pattern Recognition And Data Mining | N/A |
| Written Assessment | 15 | Data Mining Assignment | N/A |
| Written Assessment | 15 | Pattern Recognition Assignment | N/A |
Syllabus content
Introduction to Pattern Recognition, Machine Learning and Data Mining.
Clustering and Cluster Analysis.
Association Rule Mining
Classification Mining.
Artificial Neural Networks.
Support Vector Machines.
Hidden Markov Models.
Use of Software Tools.
Data Mining and Machine Learning Applications