[UML in Knowledge Modelling] [Return to Research Page]

Closing the KDD Loop to Improve Website Design

The Research

The KDD (Knowledge Discovery in Database) process for website design is an approach which uses the clickstream data for analyzing the browsing behaviour of users and improving the design of websites. There are four steps in the KDD process: Data Collection and Pre-processing, Pattern Discovery and Analysis, Recommendation, and Action. The contributions of the work fell under four headings.

Closing the KDD loop.
Developing techniques, which are suitable for analyzing browsing behaviour and discovering potential website design problems.
Developing an approach to improve the design of websites based on these problems.
Proposing guidelines for use of the KDD process for website design improvement.

The Findings

Three empirical studies are included in the thesis to show how the KDD process can be applied in practice, how the browsing patterns can be identified from the clickstream data, and how the design of websites can be improved.

This project also involved members of the Artificial Intelligence Group.

Presentations and publications associated with this research


Authors	Reference	ISBN	DOI	Google
I. Ting, C. Kimble and D. Kudenko.	Applying Web Usage Mining Techniques to Discover Potential Browsing Problems of Users. Proceedings of the 7th IEEE International Conference on Advanced Learning Techniques (2nd International Workshop on Machine-Mediated Multimodal Communication), (July 2007), Niigata, Japan, IEEE Computer Society, 2007a, pp. 929 - 930.	ISBN:	DOI: 10.1109/ICALT.2007.71	Citations
I.H. Ting, L. Clark, C. Kimble, D. Kudenko and P. Wright.	APD-A Tool for Identifying Behavioural Patterns Automatically from Clickstream Data. Knowledge-Based Intelligent Information and Engineering Systems. Berlin / Heidelberg, Springer. 4693, pp 66 - 73, 2007b.	ISBN: 3540748267	DOI: 10.1007/978-3-540-74827-4_9	Citations
L. Clark, I. Ting, C. Kimble, P. Wright and D. Kudenko.	Combining Ethnographic and Clickstream Data to Identify User Web Browsing Strategies. Information Research, 11(2), 2006.	ISSN: 1368-1613	DOI:	Citations
I. Ting, C. Kimble and D. Kudenko.	UBB Mining: Finding Unexpected Browsing Behaviour in Clickstream Data to Improve a Web Site's Design. Proceedings of the 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI 2005), (September 2005), Compiègne, France, IEEE Computer Society, 2005a, pp. 179 - 185.	ISBN: 076952415X	DOI: 10.1109/WI.2005.153	Citations
I. Ting, C. Kimble and D. Kudenko.	A Pattern Restore Method for Restoring Missing Patterns in Server Side Clickstream Data. Proceedings of Web Technologies Research and Development (APWeb 2005): 7th Asia-Pacific Web Conference, (March 2005), Shanghai, China, 2005b.	ISBN:	DOI:	Citations
I.H. Ting, C. Kimble and D. Kudenko.	A Pattern Restore Method for Restoring Missing Patterns in Server Side Clickstream Data. Lecture Notes in Computer Science. Z. Yanchun, T. Katsumi, Y. Jeffrey Xu, W. Shan and L. Minglu. Berlin/Heidelberg, Springer-Verlag. 3399: APWeb 2005, pp 501 - 512, 2005c.	ISBN: 354025207X	DOI: 10.1007/b106936	Citations
I. Ting, C. Kimble and D. Kudenko.	Visualizing and Classifying the Pattern of User's Browsing Behaviour for Website Design Recommendation. Proceedings of the First International Workshop on Knowledge Discovery in Data Stream, (September 2004), Pisa, Italy, 2004, pp. 101 - 102.	ISBN:	DOI:	Citations

Ongoing work

Derrick has now returned to work as a Lecturer at the National University of Kaohsiung, Tiwan

[UML in Knowledge Modelling] [Return to Research Page]