...
首页> 外文期刊>Expert Systems with Application >A novel prediction model based on hierarchical characteristic of web site
【24h】

A novel prediction model based on hierarchical characteristic of web site

机译:基于网站层次特征的新型预测模型

获取原文
获取原文并翻译 | 示例
           

摘要

Internet has developed in a rapid way in the recent 10 years,and tne information or web site has also been increasing fast. Predicting web user's behavior becomes a crucial issue following the purposes like increasing the user's browsing speed efficiently, decreasing the user's latency as well as possible and reducing the loading of web server. In this paper, we propose an efficient prediction model, two-level prediction model (TLPM), using a novel aspect of natural hierarchical property from web log data. TLPM can decrease the size of candidate set of web pages and increase the speed of predicting with adequate accuracy. The experiment results prove that TLPM can highly enhance the performance of prediction when the number of web pages is increasing.
机译:因特网在最近的十年中发展迅速,信息或网站的数量也在快速增长。遵循有效提高用户浏览速度,尽可能减少用户等待时间并减轻Web服务器负载等目的,预测Web用户的行为成为至关重要的问题。在本文中,我们提出了一种有效的预测模型,即两级预测模型(TLPM),它使用了来自Web日志数据的自然分层属性的新颖方面。 TLPM可以减小候选网页集的大小,并以足够的准确性提高预测速度。实验结果证明,当网页数量增加时,TLPM可以大大提高预测性能。

著录项

  • 来源
    《Expert Systems with Application》 |2011年第4期|p.3422-3430|共9页
  • 作者单位

    Department of Information Management, Chaoyang University of Technology, 168 Jifong E. Rd., Wufong, Taichung 41349, Taiwan, ROC;

    Department of Information Management, Chaoyang University of Technology, 168 Jifong E. Rd., Wufong, Taichung 41349, Taiwan, ROC;

    Department of Information Management, Chaoyang University of Technology, 168 Jifong E. Rd., Wufong, Taichung 41349, Taiwan, ROC;

  • 收录信息
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类
  • 关键词

    web usage mining; prediction; data preprocessing; markov model; bayesian theorem;

    机译:Web使用挖掘预测数据预处理Markov模型贝叶斯定理;

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号