1 x 2-hour lecture weekly
1 x 1-hour tutorial weekly
Enrolment not permitted
COMP4707 has been successfully completed
Assumed knowledge
Computing skills such as acquired in introductory computing and database topics and programming knowledge and skills such as acquired in second year programming topics. A knowledge of discrete mathematics and/or intelligent systems would also be useful.
Topic description
Data Mining (DM) and Knowledge Discovery (KD) are concerned with the extraction of useful knowledge from large quantities of more or less structured information. With the continued growth in large data sets and the inability of manual analytical techniques to cope with such volumes, data mining algorithms and knowledge discovery processes and frameworks have emerged as potential solutions. Specifically the topic will cover: Introduction - The role of common sense, trends in information management, fundamental ideas, developing data mining algorithms, applications of knowledge discovery, future directions in DMKD. Data mining techniques - association rule mining, clustering algorithms, classification and prediction, sequential pattern mining, graph mining, text mining, higher order data mining, visualisation techniques, spatial data mining, temporal and longitudinal data mining, interestingness, web mining, ethics in data mining, knowledge discovery frameworks, research methods used in data mining and knowledge discovery.
Educational aims
On completion of this topic, students will have gained knowledge in:
  1. The theoretical foundations of induction in data-rich environments
  2. The algorithms commonly used in data mining
  3. The architectural frameworks commonly used in knowledge discovery
  4. Some of the legal and ethics considerations of data mining and knowledge discovery
  5. The research methods used in data mining and knowledge discovery
  6. Current research issues in data mining and knowledge discovery
Expected learning outcomes
At the completion of the topic, students are expected to be able to:
  1. Understand the potential and limitations of given data mining approaches
  2. use common algorithms used in available systems and in the literature
  3. Understand the ethical and legal issues involved
  4. Understand the theoretical fundamentals underpinning the field
  5. Construct rudimentary knowledge discovery systems
  6. Argue the merits of various research issues in the field