基于数据挖掘的开源电子资源访问行为统计模型

为了解决当前图书馆资源个性化推荐过程中存在推荐的准确率、召回率以及效率较低的问题,采用二维距离模型构建用户社区模型,用于描述访问用户与图书馆开源电子资源之间的关系,并对互联网用户需求和访问行为进行模糊规则推理.依据互联网用户属性和图书馆资源访问需求属性之间的模糊规则,建立图书馆开源电子资源访问行为统计模型,并利用该模型向用户提供个性化推荐服务.仿真结果表明,所建模型的推荐召回率高达98. 4%,推荐准确率为99. 2%,运行时间小于0. 04 s.所建模型能够为互联网用户提供准确、高效地图书馆资源个性化推荐服务.

Statistical model for accessing behavior of open source electronic resources based on data mining

In order to solve the problems of low recommendation accuracy, low recall rate and poor efficiency in the process of personalized recommendation of library resources, a user community model was constructed with a two-dimensions distance model to describe the relationship between visiting users and open source electronic resources of libraries. The requirements and accessing behavior of Internet users were inferred with fuzzy rules. According to the fuzzy rules between the attributes of Internet users and the attributes of library resource accessing requirements, a statistical model for accessing behavior of open source electronic resources of libraries was established and applied to provide personalized recommendation service to users. The results show that the recommendation recall rate of the as-proposed model is 98.4%; the recommendation accuracy is 99.2%; the running time is less than 0.04 s. The as-proposed model can provide personalized recommendation service for Internet users accurately and efficiently.