摘要
提出一种基于中图分类法的用户兴趣模型,形式化地描述了用户兴趣模型的建立和学习过程。在建立用户兴趣模型时,需要对代表用户兴趣的中图分类号进行挖掘,由于传统的Apriori数据挖掘算法更适合于处理无序的集合,而中图分类号中的号码是有序的。提出了一种改进的算法来自动构建用户兴趣模型,并据此开发了一个科技文献过滤系统作为典型应用。
A user profile model based on the Chinese library classification is proposed and the construction and update process of this model are developed. As the traditional Apriori algorithm handles the unordered item sets efficiently, while the item sets in the chinese library classification are ordered, an improved Apriori algorithm is proposed to create the initial user profile model composed of the item sets automatically,and then the model and the algorithm are integrated into a scientific literature filter system as a representative application.
出处
《计算机应用与软件》
CSCD
北大核心
2007年第8期85-86,108,共3页
Computer Applications and Software
关键词
信息过滤
数据挖掘
用户兴趣模型
中图分类法
Information filter Data mining User profile model Chinese library classification
