摘要
Web预测模型是Web预取技术的核心,由于传统的基于PPM树的预测模型只考虑了用户的浏览序列,因此预测准确度较低。本文通过结合页面内容以及用户的兴趣来调整模型的输出,提出了基于神经网络的Web预测模型。实验表明,该模型能够在一定程度上提高预测的准确率。
The heart of Web prefetching is the predictive model. Because of only considering the user＇s browse sequences,the traditional predictive model which based on PPM tree has a low predictive accuracy. A web predictive model which based on neural networks is proposed by means of integrating web content and the interesting of the users. Experimental results have shown that the neural networks model can improve the predictive accuracy.
出处
《微计算机信息》
2010年第3期202-203,231,共3页
Control & Automation
