基于上下文偏好的XML个性化查询方法

A Personalized Query Approach for XML Based on Contextual Preference

为了解决用户个性化查询的问题,提出XML上下文偏好查询方法.用户的偏好查询结果依赖于查询提交时的上下文环境.该方法将上下文建模为多维属性的集合,并且上下文可以使用一种层级结构表示.然后,提出XML上下文偏好模型,将上下文偏好存储在特定的数据结构—兴趣树中,并提出构建兴趣树的预处理方法,保持兴趣树的规模最小.进一步地,提出XML上下文偏好松弛和查询方法,满足用户的个性化查询要求.实验结果表明,本文提出的上下文偏好查询方法能够获得较高的查全率和查准率,较好地满足用户的需求和偏好.

In order to resolve the problem of user＇s personalized query, a method of XML contextual preference query is proposed. The query results of user＇s preference depend on the context condition at the time of their submission. Context is modeled as a set of multidimensional attributes, and context is expressed by hierarchy structure. Moreover, an XML contextual preference model is pro- posed. Contextual preferences are stored in a data structure, called the profile tree. A preprocessed method of generated profile tree is proposed, and it maintains minimum size of profile tree. Furthermore, a method of XML contextual preference relaxation and query is proposed, and it can satisfy request of user＇s personalized query. The experimental results demonstrate that contextual preference query approach, proposed in this paper, has the high recall and precision, and can meet the user＇s needs and preferences efficiently.