碩士 / 靜宜大學 / 資訊工程學系碩士班 / 99 / Through the rapid development of the World Wide Web in past decades, data mining focusing on web usage becomes an important topic of research in database technologies. Web usage mining is used mainly to uncover hidden but useful information of web pages from user navigation patterns recorded in web log. In contrast, utility mining used in the traditional database can identify the profit value of items. With the advantages of both web usage mining and utility mining, web utility mining is not only identifies the frequent web navigation patterns, but also identifies the profit of web pages.
To enhance the web utility mining, this study proposes a novel Preference Utility(PU) mining algorithms, which joining selection preference and time preference to define the profit value of web pages. The PU also contains the Two-Phases algorithms to make the research more complete. Through utilizing the mining results, the web servers are able to predict the following web access page to pre-fetch the web page content, thereby improve the performance. Furthermore, the mining results could provide web designers and decision makers realize that more relevant and meaningful information regarding the internet, with which to upgrade their web pages.
| Identifer | oai:union.ndltd.org:TW/099PU005396002 |
| Date | January 2010 |
| Creators | Yu-cheng Chen, 陳又誠 |
| Contributors | Jieh-shan Yeh, 葉介山 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | en_US |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 66 |
Page generated in 0.0084 seconds