面向虚拟企业的信息搜索技术研究
文献类型:学位论文
作者 | 田力威 |
学位类别 | 博士 |
答辩日期 | 2004-01-15 |
授予单位 | 中国科学院沈阳自动化研究所 |
授予地点 | 中国科学院沈阳自动化研究所 |
导师 | 马纪虎 ; 尹朝万 |
关键词 | 虚拟企业构建与重构 信息搜索系统 强化学习 多代理协作 信息滤波 |
其他题名 | The Information Searching Technology Research Oriented The Virtual Enterprise |
学位专业 | 机械电子工程 |
中文摘要 | 面向虚拟企业的信息搜索技术是一个集人工智能、多代理协作、信息、存储管理等技术于一体的、新的虚拟企业信息技术研究中的热点问题。由于它一方面能够满足信息资源规模剧增的虚拟企业成员间、虚拟企业内部与外部间的信息交换和信息共享,同时有效地保证各个成员企业对信息隐私的要求;另一方面它既能在Internet准确搜索、自动发现企业合作伙伴,支持虚拟企业的构建与重构;又能运行于每个成员企业内部的网络系统中实现对信息的个性化访问与管理。因此,信息搜索技术已成为面向虚拟企业信息管理工具研究的新方向。然而目前虚拟企业各成员所拥有通用搜索引擎,其仅为内部的信息搜索工具,对于企业间的信息交互和共享,特别是对联盟企业构建和重构决策和过程的支持则显得力不从心。由此本文对面向虚拟企业的信息搜索技术作了系统地、深入地研究,并在以下几个方面作出了贡献: 1、针对虚拟企业的构建和信息管理的需求提出了具有专业搜索引擎的虚拟企业信息管理系统模型、体系结构和技术框架,面向虚拟企业专业搜索引擎则是虚拟企业信息系统中的高效信息管理工具。 2、提出了专业强化因子和贝叶斯文本分类器优化策略以及基于强化学习的专业搜索算法。 3、本文提出了基于延迟强化学习算法、多代理协作技术和派算法的合作伙伴发现优化搜索算法和侯选合作伙伴确定算法。 4、提出了以资源优化重分配和多搜索代理间管理范围高效协商为特点的搜索系统可重构算法。 5、给出了面向虚拟企业专业搜索系统的多层信息滤波系统。 6、本文提出了面向虚拟企业信息搜索的个性化策略。 7、给出并设计了面向虚拟企业搜索引擎V-SEARCH的实现结构、代理机制和实现技术。 |
索取号 | TH166/T57/2003 |
英文摘要 | The information Search technology faceing the virtual enterprise (VE) is a new hot point of the virtual enterprise information technology research, which is made of artificial intelligence, Multi-agent cooperation, communication technology, storage management and so on. One side, it can satisfy the progressively information share and exchange between the virtual enterprise members or inside and outside of a VE, the requirement of protecting intimate information for the VE. On the other hand, it can search and discover exactly collaborator for the VE to support building and rebuilding of the VE on the internet, but also run on the internal network for implementing to access individually and manage the internal information of the virtual enterprise.the information searching technology have been became the new research orientation of the research for the virtual enterprise information management tool. However, search engines belonging to the member enterprise, are barely a intertal information searching tool. It will look ability not equal to its ambition when it faced information exchange and share, specially supporting a decision-making and process of the virtual enterprise’ building and rebuilding. So this paper study profoundly and roundly for the information searching technology of the virtual enterprise, and to devote hereinafter several aspects: 1、According to requirement of building an virtual enterprise and information management in an virtual enterprise, To support an virtual enterprise’information management system model, system framework and thechnology frame with a professional search engine. The profrssional search engine for the virtual enterprise is an efficient informatioan management tool in the virtual enterprise information system. 2、 To import Reinforcement learning and Multi-agent collaboration technology into the special intelligent search system. Furthermore, to present the Optimal professional searching algorithms by special reinforcement reward and Bayes text classifiers basing on the reinforcement learning. 3、This paper presents the Optimal-searching algorithm for explore a collaborator and select a potential collaborator, basing on a daleyed Reinforcement learning, mutli-agent collaboration technology and calculus. 4、To present a seaching system-rebuilding algorithm whose characteristic are resources optimal reallocate and efficient negotiation about management scope between multiple searching agents. 5、To present a mutli-level information flitering system to a virtual enterprise Oriented and collaborative economical flitering algorithm. 6、This paper discuess the individual policy for the information searching in the virtual enterprise. 7、To present and design the implementing structure, agent mechanism and implementing technology of the V-SEARCH system, which is the search engine for the virtual enterprise. |
语种 | 中文 |
公开日期 | 2012-08-29 |
产权排序 | 1 |
分类号 | TH166 |
源URL | [http://ir.sia.ac.cn/handle/173321/9566] ![]() |
专题 | 沈阳自动化研究所_工业信息学研究室_先进制造技术研究室 |
推荐引用方式 GB/T 7714 | 田力威. 面向虚拟企业的信息搜索技术研究[D]. 中国科学院沈阳自动化研究所. 中国科学院沈阳自动化研究所. 2004. |
入库方式: OAI收割
来源:沈阳自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。