Pre-requisite: COMP5320 Database Systems
OverviewData mining is a process of extracting, from a large amount of data, interesting patterns that are non-trivial, hidden, new and potentially useful. It is a rapidly growing field and is becoming important because with the increasing quantity and variety of online data collections by many organizations and commercial enterprises, there is a high potential value of patterns discovered in those collections.
This module looks at different data mining techniques and gives you the chance to use a state-of-the-art data-mining tool and evaluate the quality of the discovered knowledge. The topics include: introduction to data mining and knowledge discovery process, data description, , data pre-processing, attribute selection, market basket analysis and association rules, classification, clustering, outlier detection, post-processing, social impact and trend of data mining.
This module appears in:
Total contact hours: 28 hours
Private study hours: 122 hours
Total study hours: 150 hours
Method of assessment
Assessment 1- Class Exercises (10%)
Assessment 2 - Group Project (30%)
Assessment 3 - In Class test (10%)
Berry, M., and Linoff, G. (2012). Data Mining Techniques: For Marketing, Sales and Customer Relationship Management.
Bramer, M. (2007). Principles of Data Mining.
Han, J., and Kamber, M. (2012). Data Mining: concepts and techniques.
Tan, P., Steinbachm, M., and Kumar, V. (2013). Introduction to Data Mining.
Witten, I. H., and Frank, E. (2011). Data Mining: Practical Machine Learning Tools and Techniques.
Understand the motivation for data mining in the context of business and information technology
Know how data mining is used, particularly for marketing, sales and customer relationship management
Understand the concepts and main techniques in data mining
Be able to describe the differences between the major data mining tasks
Have an understanding of the knowledge discovery process
Understand the purpose of the main tasks involved in data preparation for mining
Gain hands-on experience in using a state-of-the-art data mining tool