Large amounts of data are increasingly being collected by public and private organisations, and research projects. The Internet as well provides a very large source of information about almost every aspect of human life and society.

This course provides an overview of the technologies and concepts used for data mining and matching. It focuses on the practical aspects of these techniques rather than the mathematical and statistical foundation.

More information is on the course webpage: http://cs.anu.edu.au/courses/COMP7410/.

On completion of this module, participants should have gained a understanding of the basic concepts and techniques used in data mining and data matching, including:

1. the data mining process, how data mining is defined, application areas, disciplines involved, and the major challenges in data mining;
2. data issues relevant to data mining (size, complexity, types and formats), data warehousing, data cleaning and pre-processing;
3. unsupervised learning techniques like cluster analysis and association rules mining (including the basic methods used);
4. supervised learning techniques (classification and prediction), including the basics of decision tree induction and how to measure classifier accuracy;
5. schema integration, data matching (deterministic and probabilistic linkage), the importance of data cleaning, deduplication and geocoding.

10% Online quizzes

20% Online discussion

40% Exam

30% Case Study Project 

Incompatible with COMP8400 Algorithms and Techniques for Data Mining

Assumed knowledge:

• Basic understanding of spreadsheets and databases (tables, attributes, records).
• Knowledge of working with windows based computer systems.