【摘要】 随着WWW的迅速扩张和网络用户的急剧增加,准确预测Web用户的访问行为对减少用户的感知延时、提高网络服务质量具有重要作用。文章分析了传统Markov模型的优缺点,针对低阶Markov模型准确率不高,而高阶Markov模型状态空间复杂度较高的缺点,提出了一种新的K步Markov模型(K-Step Markov Model,K-SMM),并引入了偏爱度的概念,建立了基于偏爱度的多步Markov网页预测模型(Preferred K-Step Markov Model,PKSMM)。实验结果表明,该模型可以获得更高的预测准确率与覆盖率,并能有效地降低存储复杂度。更多还原

【Abstract】 With the rapid growth of World Wide Web and the sharp increase of Internet users,accurately predicting the visit behavior of Web users plays an important role in reducing user’s sense delay and improving network service quality. Analyzing the advantages and disadvantages of traditional Markov model,aiming at the low accuracy rate of low-order Markov model and the high space complexity of high-order Markov model,a new K-step Markov model (K-SMM) is presented and preferred K-Step Markov model (PKSMM) for webpage prediction is built by introducing the concept of preference. The experiment result shows that the model can achieve higher accuracy rate and coverage rate,and lower space complexity effectively.更多还原