Loading...
Please wait, while we are loading the content...
Similar Documents
Extraction of useable structures from click stream data (2004).
| Content Provider | CiteSeerX |
|---|---|
| Author | Perry, Andrew |
| Abstract | Number of words = 27,710 as counted by Word’s word count function. This includes the body of the report and excludes the attached appendices. The common methods for determining usage statistics for websites are unreliable as the World Wide Web was never designed with this in mind. In order to improve the site from both a technical and a users perspective there is an increasing need for more meaningful information about site usage. This report describes an extendable architecture that covers all areas of the web usage mining process. To make this application as practical as possible the data analysed is limited to server access logs, which are commonly available but require some pre-processing to make them a suitable source of data. Simple but effective algorithms are used to pre-process the log data and remove any inconsistencies and eliminate any potential source of bias. The results obtained have highlighted new sources of data inconsistency, such as the use of container pages by search engines, and demonstrate the importance of considering pre-processing as whole rather than a |
| File Format | |
| Publisher Date | 2004-01-01 |
| Access Restriction | Open |
| Subject Keyword | Common Method Suitable Source Attached Appendix Data Inconsistency World Wide Web Click Stream Data Web Usage Mining Process Useable Structure Meaningful Information Access Log Container Page New Source Site Usage Log Data Potential Source Search Engine Usage Statistic Extendable Architecture Effective Algorithm Word Word Count Function |
| Content Type | Text |
