【作者】 刘健；

【导师】 张海涛；

【作者基本信息】 吉林大学 ， 情报学， 2015， 博士

【摘要】 随着基于信息生态视角的知识组织研究不断深入，与图书馆知识生态化与组织方向相关的研究，也逐渐呈现出了与生态理论交叉演进式发展的局面，这种发展的实质是实现图书馆知识服务的生态化与可持续性发展。本文立足于图书馆的知识生态化实践，以协同进化的哲学生态观为指导，利用计算机科学与生命科学的相关方法，并从多角度对数字环境下的图书馆知识生态化原理与过程展开研究，从而，在理论与实践两个层面推进图书馆知识的全面生态化进程。首先，从核心概念、理论基础以及研究进展三个方面对协同进化理论进行了述评，并给出了协同进化、知识生态系统以及自组织神经网络等关键词的定义，并对本研究所涉及的核心概念进行界定。接着，对协同进化理论的相关理论基础进行了详细的论述，并在分析各个理论基础内在联系的同时，研究协同进化理论得以产生与发展的根源。之后，在梳理协同进化理论与协同进化算法研究进展的基础上，分析其理论与算法发展的轨迹，通过总结与分析相关学者的关注重点，捋顺协同进化理论的研究脉络，一方面，可以为论文的后续研究提供坚实的理论基础和方法论依据，另一方面，可以在实际构建系统过程提供理论指导与支撑。以图书馆的知识生态化过程为研究对象，围绕着图书馆的读者知识服务需求，在借鉴协同进化相关理论研究成果的基础上，结合生态科学与计算机科学的相关方法分析了图书馆知识生态系统的可行性以及现实性，构建起了一套具备生命体特征的图书馆知识生态系统模型，在构成的基础要件、逻辑依据、运行机制和实现手段等方面，进行全面且有针对性的探讨，并进一步的展现了其运行机制与机理，阐释了基于协同进化理论的知识生态系统在知识生态化过程中扮演的角色以及所起到的作用。其次，研究了如何构建基于协同进化的图书馆生态系统模型，对生态系统理论及协同进化理论进行了梳理，通过对多种图书馆生态系统构建基本因素的分析、评价与优化，完善了系统模块构成。在构建生态系统逻辑结构模型的基础上，研究了其各要素形态、组配方法以及生态化运行机制，深入分析了其类生命体的协同进化的特征，在为服务对象提供更高效知识的同时，进行知识再造、创新与增值，通过生态环境下各子系统协作处理、相互作用、相互促进的过程，为实现图书馆知识生态化奠定系统模型基础，同时，借鉴了生态学中协同进化的理论，在界定了图书馆知识协同进化内涵的同时，重点分析了竞争与协同对图书馆知识进化的影响，在研究其自组织的过程与机理的基础上，阐明了知识协同进化是通过图书馆知识要素间的关联，来促进知识群落共同发展的自组织活动，根据知识进化内在动因，揭示知识协同进化的运作机理，从而，使图书馆中的知识种群可以在生态化过程中，达到全面协调可持续的发展。再次，利用生物学中协同进化的概念，通过对协同进化理论的梳理，提出了图书馆知识协同进化的定义，即知识协同进化是基于图书馆知识群落内各要素之间的关联，研究其催化知识自身与其它群落共同发展的机制，最终达到共同进化的目的。分析了知识协同进化的两种模式，阐述了知识间协同进化的主要方式是协同，而主要形式是自组织。知识群落间普遍存在的既竞争又协作的辩证统一关系，直接促使知识个体通过自我完善的方式，影响自身所处的知识群落，处于同样进化过程中的其它知识群落又在某种程度上影响着这个知识个体发展的路线，最终促进整个图书馆知识系统达到最优化运行。图书馆作为一个生长着的有机类生命体，面对用户个性化的知识需求，经常会调整自身的数据结构或聚类方式，以适应用户需求。进而，在利用深度知识挖掘技术为图书馆生态化知识服务提供了坚实的数据基础的同时，通过对知识挖掘过程的跟踪以及挖掘相关技术的研究，为图书馆知识高效利用提供理论与技术支撑，并将自组织神经网络引入到图书馆知识聚合中来，在分析自组织神经网络的图书馆知识聚合模式基础上，结合自组织神经网络算法，完成了图书馆知识神经网络的训练，以构建评价函数发现知识聚类中心，在对离散分布状态下的知识进行实时标引、紧密关联的基础上，实现了资源与路径的关联知识聚合。最后，对知识生态化过程中所涉及到的技术进行了论述，梳理了元数据抽取技术、数据挖掘技术与自组织人工神经网络技术等生态化技术的特点，在此基础上深入研究了元数据抽取过程中，并利用相关算法帮助图书馆知识实现了元数据化。进而分析了如何对大量的元数据进行深度数据挖掘，以便有效的对高度序化的元数据进行神经元组织，使其具有生命体的典型特征，从而促使知识个体与群落的协同进化，由此从技术层面实现构建了基于协同进化的图书馆知识生态化技术路线图。利用MATLAB软件工具完成对图书馆知识种群生态化的过程的可视化仿真，为图书馆知识协同进化过程的可视化模拟，提供了一种全面直观的解决方案。在上述工作的基础上，基于Linux内核系统平台，利用C++语言搭建图书馆知识生态系统程序模块，并将虚拟化封装技术引入图书馆知识生态系统的构建中。通过构建图书馆知识生态系统虚拟化模型，展现了其封装机制、实施方法以及应用优势，最终，实际构建起了可以完成图书馆知识生态化的软件系统。综上所述，本文以图书馆知识生态化过程为研究对象，以协同进化理论为依托，围绕着图书馆读者知识的创新性服务需求，结合生态学与计算机科学的相关方法，详尽论述了图书馆知识生态系统的可行性以及现实性，在构成的基础要件、运行机制与实现手段等方面，进行全面且有针对性的探讨，以期能够完整阐释基于协同进化理论的图书馆知识生态系统在知识生态化过程中扮演的角色以及所起到的作用。更多还原

【Abstract】 With the gradual deepening of the research on knowledge organization from thetheoretical perspective of information ecology, and the related studies onorganizational directions of library knowledge ecologicalization, the paper presentsthe development of a cross-evolutionary ecological theory, whose essence is therealization of the ecologicalization and sustainable development of knowledgeservices in library. It, based on the practices of library knowledge ecologicalizationand under the guidance of co-evolutionary philosophical ecology concept, employsthe correlated methods of computer science and life science, and conducts studies onthe principles and processes of knowledge ecologicalization within the environmentof digitalized library from multiple perspectives, thus promoting a comprehensivelibrary knowledge ecologicalization both in theory and in practice.Firstly, co-evolutionary theory is reviewed from three aspects: core concepts,basic theories and research advancement, definitions of key words are given, such asco-evolution, knowledge ecosystem, self-organized neural networks, and so on, andthe key concepts involved are to be redefined. Then, the bases of related concepts ofco-evolutionary theory would be discussed in detail, and its orgin and evolvement areto be researched while the internal relations of each theoretical basis analyzed. Afterthis, on the basis of the combing of co-evolutionary theory and the progression ofinquiries on co-evolutionary algorithm, its theoretical and algorithmic advancementsare evaluated. Summarizing and analyzing the focuses of relevant scholars andordering the exploration of co-evolutionary theory can, on the one hand, provide asolid theoretical foundation and methodology basis for follow-up studies, and on theother hand, provide a theoretical guidance and support in the actual process ofbuilding up the system. Subsequetly, the utilization of the reviews of co-evolutionarytheory makes the objectives of the reserach and the ensuing studies more explicit. Thepaper, using library knowledge ecologicalization as the research target, centers onreaders’ needs of knowledge services, and, by referring the related works ofco-evolutinary theory, it combines the ecological science and computer science toprobe into the feasibility and reality of library knowledge ecosystem and construct amodel of the ecosystem with a set of specific features of life. Several respects, such asbasic elements of composition, rationales, mechanisms, and means of achievementand so on, are deliberated in a comprehensive and targeted way, and its operationalmechanisms are further demonstrated, and the roles of knowledge ecosystem based onthe co-evolutionary theory played in the process of knowledge ecologicalization are expounded.Secondly, how to build up a model of library ecosystem centered on co-evolutionis inquired, and the theories of ecological system and co-evolution are combed.Through the analysis, evaluation and optimization of a variety of basic elements oflibrary ecosystem, the paper would improve the constitution of the system module. Onthe basis of constructing a logical structure model of ecosystem, it studies themorphology of each element, grouping methods, and ecologicalized operationalmechanisms, and dissects the features of co-evolution of quasi-living creatures indepth. As a more efficient knowledge provided for service objects, knowledge isrecreated, innovated, and appreciated. In the process of the coordination, interaction,and mutual promotion of each subsystem in ecological environment, it lays thefoundation for the accomplishment of library knowledge ecologicalization to presentlogicality for subsequent chapters. Bringing up the co-evolutionary theory in ecologyand defining the connotation of knowledge coevolution in library, it would scrutinizethe impact of competition and coordination on library knowledge evolution.Exploring its self-organized processes and mechanisms, it would clarify theknowledge co-evolution is through the association between knowledge elements inlibrary to promote common development of self-organized activities in knowledgecommunity. According to the intrinsic motivation of knowledge evolution, it revealsoperational mechanisms of knowledge co-evolution; therefore, knowledge populationcan achieve a comprehensive, coordinated and sustainable development during theprocess of ecologicalization.Thirdly, using the concept of co-evolution of biology, evolutionary theorycombing through the proposed co-evolution of the definition of library knowledge,that knowledge is the study of co-evolution of the elements associated with thecatalytic knowledge in the library of knowledge between communities based on theirown a common mechanism with other community development, and ultimatelyachieve the purpose of co-evolution. Analysis of the co-evolution of knowledge intwo modes, namely, self-organization and competition and complementarity. Themain way knowledge coevolution between competition and complementarity, inwhich the main form of knowledge ecosystem co-evolution of knowledge isself-organization, and thus, the use of correlation between knowledge communities,the ecological environment in the library, and aKnowledge can be affected by anindividual’s own evolution in which the knowledge communities and othercommunities of knowledge, whereas in other communities in the knowledge of the evolutionary process again in some way affect the course of individual developmentof knowledge, and ultimately promote the entire book Museum knowledge systems tooptimize development. The growth of the library as an organic life form, in the face ofthe knowledge needs of individual users, often adjust its data structures or clusteringmethods to accommodate user needs, and therefore, the depth of knowledge miningecological knowledge for library services provide a solid foundation of data, throughthe study of library knowledge mining process and related technology to providetheoretical and technical support for the self-organizing neural network is introducedto the library of knowledge aggregation to the efficient use of knowledge for thelibrary, in the analysis of self-organizing neural Library Knowledge aggregationnetwork model based on the combination of self-organizing neural network algorithmto complete the training of the neural network library knowledge, build evaluationfunction found knowledge cluster centers in the state of knowledge in a discretedistribution in real time indexing, closely related based on the realization of thelibrary resources and associated knowledge independent route aggregation.Finally, the ecological characteristics of knowledge involved in the process oftechnology were discussed, combing the meta data extraction, data mining technologyand the characteristics of self-organizing artificial neural network technology andother ecological technology, on the basis of in-depth study of the yuan data extractionprocess, how to make use of the relevant algorithms to help achieve the metadatalibrary of knowledge. Further analysis of how a large number of in-depth data miningmetadata to highly effective sequence of metadata neuronal tissue, it has the typicalcharacteristics of a living body, in order to achieve its individual or community ofco-evolution, by This from a technical level to achieve the co-evolution of the librarybuilding knowledge-based eco-technology roadmap. Using MATLAB software toolfor complete visual simulation library population ecology knowledge of the process,the process of co-evolution of knowledge for visual simulation library provides acomprehensive visual analog solution. Based on the above work, based on the Linuxcore system platform, using C++language to build a library of knowledge ecosystemprogram modules and virtualization technology into the library encapsulatesknowledge ecosystem building by building virtual library knowledge ecosystemmodel,demonstrated its encapsulation mechanism, implementation methodology andapplication advantages, ultimately, you can actually build a complete library ofsoftware systems ecological knowledge.In summary, this paper library knowledge of ecological processes for the study, in order to co-evolution theory-based, innovative services around the needs of libraryusers knowledge, combined with the related methods of ecology and computerscience, and a detailed discussion of the books feasibility and realistic knowledgeecosystem museum, in basic elements of composition, operation mechanism andmeans of achieving so comprehensive and targeted to explore in order to be able toexplain the complete theory of coevolution based library of knowledge in theknowledge ecosystem ecological processes and the role of the role.更多还原