论文部分内容阅读
针对当前E-learning推荐系统存在的问题,引入多Agent(MAS)系统,提出一种基于Web使用挖掘的集成MAS与Web services的分布式智能推荐系统模型。该模型能动态生成基于用户使用信息的个性化链接页面,有效地帮助学员找到所需的资源信息。提出了一种基于最近最少使用策略的系统推荐算法,包括系统整体实现算法、系统聚类算法及推荐算法。实时性能分析显示该系统运行性能良好。
Aiming at the problems existing in the current E-learning recommendation system, a multi-agent (MAS) system is introduced and a distributed intelligent recommendation system model based on Web usage mining and integrated MAS and Web services is proposed. The model can dynamically generate personalized links page based on user’s usage information, which can effectively help trainees to find the required resource information. This paper proposes a system recommendation algorithm based on the least-recently-used strategy, which includes the whole system realization algorithm, the system clustering algorithm and the recommendation algorithm. Real-time performance analysis shows that the system performance is good.