The paper introduces some researches on ETL application in digital libraries,and analyzes classification and application field of ETL in resources construction,user service,resources sharing,system interoperability of digital libraries.
This paper summarizes the definition of usability,and briefly introduces some foreign cases of usability evaluation on open-source institutional repository-building software.Then the usability method is proposed to evaluate open-source software based on the purpose and significance of building institutional repository.
In view of the existing problems such as interoperability,resource building and sharing,server failure,load balancing, knowledge management. The authors propose a distributed framework of digital library based on the technology of OAI and topic map after analyzing and researching the technology of P2P,OAI and topic map.
In order to improve sharing and acquisition of the scientific data, query system architecture based on multi-titers metadata is proposed on study of the scientific resource data features and XML technology. It focuses on multi-tier metadata structure of scientific resource database depicted by XML Schema and its application in the optimization for the scientific resource databases query. Research has shown that this framework is helpful for the data location and search and it promotes data utilization ratio in scientific research, production and living.
The authors propose a resource organizational architecture model in this paper,and then design a resource sharing service architecture model for information sharing management in the Grid. This model consists of the resource and user layer,service domain layer and Virtual Organization(VO) layer. Finally,The unthors analyse the functions and characteristics of the model, and propose the direction of the later study.
In this paper, considering the limitation of keywords-based information retrieval, a framework of Ontology-based semantic information retrieval systems is put forward according to the hybrid Ontology-based integration approaches, and the functions of every module are described. In addition, some key issues are discussed, including query and resource processing, resource searching, and Ontology mapping.
With the problem existing in the current knowledge organization, modular Ontology is introduced to knowledge organization, and a method of open Knowledge Organization(Onto-KO)is put forward, and open knowledge organization structure,its organization principle,semantics relations between Ontology modules,and arithmetic are analyzed. The method can make knowledge divided and assigned to users, which causes users to anticipate in building and maintaining knowledge together, and leads to open knowledge organization realized, so the complexity of knowledge organization is reduced and knowledge organization quality and cooperative ability are improved to some degree.
Knowledge representation and matching-reasoning are two key steps for a semantic knowledge management system. In order to realize the semantic management of literatures supply chain knowledge, the authors put forward the literatures supply chain knowledge Ontology model and formalize the model using RDF (Resource Description Framework) and advanced Voronoi diagram. The authors setup the qualitative reasoning rules based on the RDF formalized model and put forward the quantitative reasoning arithmetic based on the advanced Voronoi diagram formalized model. The experiments show the reasoning rules and arithmetic based on the formalized model could get rational results in literatures supply chain knowledge management.
This paper analyzes and reviews the research on Topic Maps under the investigation of the articles and the Websites about them. Besides that, the authors also make some advices on how to develop Topic Maps in China.
After analyzing the reference monitor in digital rights management, the paper gives multi-layer architecture and third-party architecture in order to solve the shortcomings in the current DRM systems. Finally it analyzes the key elements of the architecture.
Data cleaning problem is surveyed in this paper. Firstly, the background of data cleaning problem and research status is explained. Then, the definition and objects of data cleaning are given. The basic principle and some models of data cleaning are presented. Related algorithms and tools are analyzed and evaluation methods of data cleaning are proposed. Finally, the future research topics and application related to data cleaning problems are discussed.
This paper brings forward two models for person-name entity extraction based on the comparison of math theory between HMM and CRFs, one using word role label based HMM and the other using character role label based CRFs, then validates and compares the effect of both by open-testing and applying in practice, and thereby proves in practice that CRFs is fitter for sequence labeling and object classifying than HMM.
This paper introduces the architecture and features of Mediawiki Wiki engine, and analyses the install ation,configuration,the construction of classification system and the extension setup of Mediawiki, then gives a brief view of how to use the Mediawiki Wiki engine to construct the subject information portal. Meanwhile, the authors take library science,information science and archives science subject information portal as an example to explain the system.
This paper analyzes architecture of the modules in degree paper system and the features and functions of DSpace.After analyzing the advantages of constructing the degree paper system with DSpace, the authors also give the introduction of working practice about the implementation.
According to the problem that the OPAC(Online Public Access Catalogue) search machine lacks of the proprietary operating system environment in the library of university，the author puts forward an OPAC machine solution based on windows platform. Also introduces the design and implementation of OPAC search machine based on windows platform.
Under the non-credible environment, transmitting sensitive information such as plain text passwords is easy to lead to security issues of login of digital library. The paper proposes an improved solution based on a one-way Hash function, which uses information technology such as message digest and JavaScript, and sensitive information such as plain text passwords will be hashed for the fixed-length and single-effective “digital fingerprint” before entering the non-credible environment, so it can enhance present security of login of digital library.
Based on the four-corner coding system of the author,an “automatic generating system of Chinese author number” is designed in this paper. After testing the system,the librarians can get the author number quickly and accurately through this system,which meets our library requirements well.
A self-designed integration system of university library information resource is proposed based on the current status of resource integration of digital libraries. On the basis of ASP and SQL Server database, the system combines the bibliography information of the traditional papery ILASⅡ system with the electronic book system Apabi, using the architecture of entity and virtual method. Application in the Library of Ocean University of China shows that it can provide the readers with better information service in library operation.
