Abstract
This paper describes a novel web usage mining approach to discover patterns in the navigation of websites known as Unexpected Navigation Behaviours (UNBs). The approach provides a web designer with a means of identifying and classifying patterns of browsing and, by reviewing these patterns, the designer can then choose to modify the design of their site or redesign it completely. UNB mining is based on the Consecutive Common Subsequence (CCS), a special instance of Common Subsequence (CS), which is used to define a set of expected routes. The predefined expected routes are then treated as rules and stored in a rule base. By using the predefined route and the UNB mining algorithm, interesting navigation behaviours can be discovered. This paper will introduce the format of the expected route and describe the UNB algorithms. It will also describe a tool that a website designer can use to define the expected route more efficiently, which can help the website designer to make decision about where and how the design of website can be improved. The paper concludes with a series of experiments designed to evaluate how well the UNB mining algorithms work and demonstrate how UNB mining can be useful for improving website design.
| Original language | English |
|---|---|
| Pages (from-to) | 71-92 |
| Number of pages | 22 |
| Journal | Journal of web engineering |
| Volume | 8 |
| Issue number | 1 |
| Publication status | Published - Mar 2009 |
Keywords
- Navigation behaviour
- web usage mining
- clickstream data
- sequential mining
- website design
- website designers
- USER NAVIGATION
- WEB
- DISCOVERY
- PATTERNS
- COMMERCE