【作者】 牟向伟；

【导师】 陈燕；

【作者基本信息】 大连海事大学 ， 管理科学与工程， 2010， 博士

【摘要】 电子政务作为国家管理部门与社会民众之间有效便捷的沟通手段与服务平台,起到了至关重要的作用。随着网络时代的发展,网络上信息资源“量”的丰富和“质”的稀缺影响了电子政务信息化建设的效果,迫切需要一种能够发现用户内在需求,并主动提供信息服务的功能。近年来在国内外兴起的个性化推荐成为解决这些问题的重要途径之一。个性化信息资源的建设组织过程是一项系统工程,为了满足电子政务个性化信息服务发展的需要,本文首先在分析了个性化电子政务服务的内容、建设模式、目标和原则的基础上,提出了以门户网站为核心的电子政务个性化信息服务体系和构建方法。阐述了在信息集成环境下信息资源的权限管理和访问控制方法,并建立统一的模型来描述集成后的信息资源和用户特征。本文使用多种基于数据挖掘的方法对用户的兴趣进行深层次的发现,并对经典的协同过滤算法进行改进,使算法在保证准确度的同时提高了运行效率。最后,提出了一种模糊语义个性化推荐系统模型,并使用FALC模糊描述逻辑语言实现了该模型。论文以云南省航务海事管理局电子政务门户网站和内容管理系统为研究案例,验证了本文提出的以门户网站为核心的电子政务个性化信息服务体系以及相关模型和算法。实例证明,本文提出的模糊语义个性化推荐系统和实现方法是实用可行的。它可以在语义环境下为用户提供更准确的和扩展性更高的个性化推荐服务,使得电子政务中的信息资源能被系统高效的吸收和利用,从而满足用户的个性化信息需求,发挥资源的最大效益,实现资源的合理配置和共享。更多还原

【Abstract】 E-government has playing a very crucial role as the service platform between national authorities and civil society. With the development of the Internet, the contradiction between the "plentiful quantity" and the "inferior quality" has an harmful effect on e-government development. It is urgent to find a way which can discover the user’s internal needs and provide recommenderation information sevice. Personalized recommendation system has become a vital way to solve these problems.The construction of personalized information resources is a system engineering work. In order to meet the needs of the e-government development, firstly, a personalized information service system construction method is proposed with a core of portal website, including the object, pattern, target and principle of the system.afer information integrated, the authority management and access control are introduced to enhance system security. Secondly, a unified model is established to describe the integration of information resources and user profile. Thirdly, the data mining methods are used to discover the user interest in the integrated information and the classic collaborative filtering algorithm is improved by the stability degree. Finally, a fuzzy semantic model of personalized recommendation system is proposed, and a kind of fuzzy description logic language is used to implement the model.Taking the Portal website of Maritime Administration of Yunnan Province as an example, the proposed personalized e-government system, related models and algorithms were verified. The case showed that the proposed fuzzy semantic personalized recommendation system and methods are practical and feasible, which can provide users with more accurate and personalized recommendation service, and make e-government information resources highly efficient utilized and meet the user’s personalized needs.更多还原