用户模型驱动的遥感信息智能服务方法

摘要

传统的浏览下载的遥感信息服务模式存在被动性、同一性等缺点，如何主动、准确地满足用户的个性化需求已成为遥感信息服务的焦点问题。本文针对遥感信息在空间和波谱上的覆盖特性，引入区间数学的方法建立了用户模型，描述用户兴趣在遥感信息核心元数据上的分布特征。提出关联度和兴趣度概念用来评价遥感信息对用户兴趣的满足程度，设计了基于拓扑关系的关联函数定量计算关联度。通过将待分发遥感信息作为备选方案构建了决策矩阵，从而将遥感信息的智能服务问题转化为多属性决策问题，实现了面向用户兴趣的遥感信息主动推荐。%To overcome l imits of traditional passive remote sensing data distribution methods,personal i-zation is an inevitable trend of spatial information service.User profi le is the basic foundation for personal-ized,active and accurate spatial information dissemination service,which can be used in the definition of users’preferences or interests of remote sensing data.In view of the coverage characteristics of remote sensing data in spatial and spectrum,this paper adopted the interval mathematics method into the repre-sentation of user profi les.Based on user profi les,the concepts of the correlation degree and interest degree are introduced.In addition,by the construction of decision matrixes,the solutions of the generations of intel l igent services can be converted into the solutions of multiple attribute decision problems.Final ly,an instance is given to show the usabi l ity of the presented method.