为改善用户的Ｗｅｂ页面访问行为、提高访问效率，设计了一种基于贝叶斯网络的网页推荐模型及推荐算

法。通过收集和分析服务器中的描述文件和日志文件，利用Ｂａｙｅｓｉａｎ网络分析页面间的依赖关系，构建了基于贝

叶斯网络的网页推荐模型并产生推荐集。通过在Ｍｉｃｒｏｓｏｆｔ公司提供的网络日志数据集上做的实验，可以获得超

过８０％的准确率和覆盖率。理论分析和实验结果表明：算法能够在线实时向用户做出个性化的推荐，与已有的推

荐算法相比，算法能较快地给出推荐集，并且可以获得更高的准确率和覆盖率-To improve the user s Web page access behavior, improving access efficiency, design a web page recommendation based on Bayesian network model and recommended calculation

Method. Through the collection and analysis of server log files described in the file and, using Bayesian network analysis of dependencies between the pages, constructed based on shell

Bayesian network model and produce recommended web page set of recommendations. Provided by the Microsoft log data sets on the network to do experiments, you can get over

Over 80　accuracy and coverage. Theoretical analysis and experimental results show: online real-time algorithm can make to the user personalized recommendations, with the existing push

Recommended algorithm, given recommendation algorithm can quickly set, and can achieve higher accuracy and coverage