This paper proposes a standard system frame of long-term digital preservation technology for China national science and technology plan according to the standardization’s requirements of the plan, and analyses general standards, professional standards, standards selection and the importance of application.
This paper analyses realistic demands of automatic metadata extraction, elaborates related research on automatic metadata extraction and compares three typical automatic extractors of metadata, including DROID, NLNZ Metadata Extractor and Metadata Miner Catalogue PRO. On the basis of discussing present limitations of automatic metadata extraction, the authors give a summary and prediction of this technology.
The audio and video module of consulting system is designed. This paper elaborates the module technology architecture，communication process, key point technology mechanism and the development tendency of the future.
Based on the introduction of several separated phenomena of approaches to CLIR, this paper raises the integrated approach of query translation and retrieve process to CLIR and discusses the methods of this approach.
This paper firstly defines tag, tagging, folksonomy, then analyzes the limitation of collaborative tagging system. In order to achieve reorganization of the user tags and better tag navigation, browsing mechanism, the authors propose a method on using hierarchical agglomerative clustering algorithm to cluster the tags. Finally experiments certify the tag cluster method.
In order to resolve the disadvantages of traditional TFIDF in text filtering, the authors propose a text information filtering algorithm based on feature extraction. This paper briefly analyses the text information filtering principles and processes, and then focuses on the design and realization of information filtering algorithm. Experimental results show that the new approach significantly outperforms the traditional information filtering method.
This paper studies the relation extraction of domain Ontology, using the technology of machine learning and natural language processing. The extraction method on hierarchy relation and domain relation is introduced in detail. Test results show that this mothod is valid.
Through computing the similarity and relevancy based on Ontology, the author uses semantic reasoning to make connotative semantics clear in order to mine fully the information correlative with the search content. This paper implements the relevant information’s semantic fusion and improves the precision and recall.The method of computing semantic similarity and relevancy based on Ontology is studied and is proved feasible by implementing the semantic search method.
In this paper, the academician Ontology is constructed via topdown and kernelexpanding methodology to express academician concept.The paper describes the constructing process in detail,and gives the classes, relations and some templates for collecting instance among the academician Ontology. Moreover, it provides the key strategies and codes for instance storing,which will be helpful for expression of human concept in relative fields.
In view of the shortcomings of traditional methods for analyzing public opinions, this paper proposes a new idea of public opinion analysis under the Web,and then designs a model for it. Experiments show that the proposed model is an effective solution to analyzing public opinion under the Web.
Taking Chinese blog groups as the object, this paper reveals the characteristics of blog groups and their evolution mechanism. Then the authors establish a complex network based on the relathonships of the common-member in blog groups. Finally, the authors propose an evolving model of the blog groups complex network. The simulation results based on this model are in quite good agreement with the empirical statistical results.
This paper gives the definition of the formalization and analyzes the principles. Two key problems are discussed. It provides the effective approach of the formalization. The author proposes an indicator mapping model, which is multi-level and two-dimension. This quantitative model realizes the transform of evaluation goals and metrics. The author also designs the indicators mapping model definition language, which is used for formalized description of indicators.
Aiming at the managing current situation on computer and network center room in libraries, the author adopts SNMP to realize the general intelligent management concerning the hardware equipments of computer and network center room,running situations of digital resources and the environment of computer and network center room.
The paper analyses the shortcomings of the current CD attached books management system and gives out some solutions. According to this, the authors take advantage of software and database technology to design and implement a CD download system, so that readers can make the best use of the CD attached books in libraries.
Based on the security demands of grid, the paper proposes a retracted and extensible security strategy of access control as well as the corresponding security model. It also presents a flexible access control security policy, which supports the inherent dynamics, flexibility and expansibility of grid computing environment. Based on the strategy design, the corresponding access control security model is able to support identity authentication and access control through XML-based inquiring request / assertive response language SAML and extensible access control language XACML.
This paper summarizes the difficulties in data format transform between information retrieval and information analysis, and designs a general simple procedure of data transform, which is combined with text file, Word software, VC++ programme and Excel software. The usage and realization algorithm of the software are also introduced.
The application of SFX at Tsinghua university library is briefly introduced. Then the implementation of SFX and the shortages of INNOPAC as SFX source are analyzed. Finally, a method on extending SFX service based on INNOPAC Z39.50 protocol is explained.
This paper employs transparent proxy technology to control users’ behavior on Internet, and the authors develop a Web management system for the access control of Squid proxy-server using Ajax and PHP, which improvs the managing efficiency of the electronic reading room.
The published papers, the cited papers, and the criteria for evaluation of journals in area of library and information science are analyzed and evaluated using the database of 2004-2006 from CSSCI. The aim of the present paper is to reveal the unique feature and increase the academic influence of New Technology of Library and Information Service.