Based on introducing to the background, this article discusses the goals, types and requirements of library Mashups, and analyzes present situations, features and trends of Mashup application in the field of the library. Three sections of Mashup construction are open data, Web services and presentation layer components of its main objects. The author also points out that open APIs, structured data and support tools are requirements for Mashup implementation.
After introducing the definition, type and feature of Widgets, this paper emphatically analyzes object structures and API standards of UWA and W3C Widget, and discusses the architecture of enterprise Mashup and implementations of the Mashup application based on Widgets. Finally, the related work and trends are given.
This paper introduces the research progress of the name authority control for the contributor, including the sources and the representations of the problems, the effect of name authority control, and the main methods to solve the problems at present as well as their shortcomings. At last, it proposes a suggestion which uses the semantic Web technology to realize name authority control.
On the aspect of the availabililty of information service platform, this paper describes the basic concepts and features of virtual machine, compares and analyses the typical cases of them, and expatiates the detailed realization processes of the high available platform for information services based on virtual machine. The high availability of the service platform is realized by network load balancing to distribute workload evenly in the access intensive virtual applications, and by the methods of resource hypervisor, live migration, virtual snapshots, et al. in the light virtual applications.
This paper makes an analysis on the historical literatures of Ontology domain by using bibliometric analysis method，computer statistic analysis technology, social network analysis software, and draws the literature quantity distribution map，the co-occurrence network of core keywords, which can excavate the development trend，general situation and research hot points of Ontology application domains, giving readers an intuitive，clear understanding, and providing a guide for the following research works.
In order to get more information about how people describe their image information needs, the authors detailedly describe the research progress, including visual information descriptions, classification frameworks, its empirical study, its application, and so on. The paper concludes with a summary of the related work and its trends.
Embedded service is widely used in many applications for its convenience. However, it is usually confined to specific application systems. In this paper, the authors present an Embedded Ubiquitous Personal Knowledge Service(EUPKS) model which aims to breaking through the border of application systems, conducting knowledge organizing and constructing embedded services throughout personal knowledge activity workflows. We describe the principle, the formulation expression, the framework, the tech-architecture and key technologies. Finally, a prototype system is given based on the desktop search tool of National Science Library (NSL).
To enhance the retrieval accuracy of information search engine, this paper proposes an information retrieval system based on Ontology and document refinement, which is realized by employing the semantic description and relevance of Ontology to the system. It describes the using of LSI to replace the traditional VSM in the results of sorting process. Using a comparative experiment, the authors show the new approach is more feasible and effective than VSM, which can improve the performance upto 10.55%-17.63%.
This article gives an account of the steps of how to automatically build a large-scale sentence-level English-Chinese parallel corpus based on websites. Specifically speaking, the following questions are addressed: the criterions which are used to grab websites are set and words library is worked out; the websites are automatically grabbed by making use of the tool ‘Wget’; the English-Chinese parallel sentences extracted from websites are subsequently processed and the Chinese sentences are segmented based on Conditional Random Field. Finally, the building of English-Chinese parallel corpus is completed which includes 1 017 963 English-Chinese parallel sentences stored in database which are automatically extracted from 675 308 websites.
This paper raises a method of Web pages extracting which is based on feature orienting boarder forecast for extracting the Web archive effective content in high-speed. Two tools named ROST CM and ROST Text Extractor, is developed to build the training data set and test the algorithm. Theory and experiment show that the algorithm is suitable for Simplified Chinese, Traditional Chinese and English Web pages,and can be well adapted to news and blog Web archive management.
This paper introduces the development of newly-emerging social Q&A sites and their related studies. By analyzing the elements and operation process of Chinese Yahoo Answers, the usage patterns of social Q&A site are presented. The result shows that there are significant differences in users’ interests, usage motivation, knowledge structure and question formulation, while the overall answer quality is not satisfied. This paper lays the foundation for a deeper understanding of usage behavior.
To promote the digital service of library system, it needs to build an alerting system and monitors the quality of network service continuously, which will accelerate the speed of technical support response. The architecture and functions of the monitoring system are designed in the paper，and some coding is made to combine with the open source software. The network of National Science Library is now under supervising 7×24 by the monitoring system, and achieves good effects in the actual application.
This paper analyzes the meaning of library information push technology firstly. Then, combining with the practice in the author’s library, it introduces the design ideas, and key technologies of library information push system based on screen saver.
Through researching the Fedora open source repository system and drawing on experience in the construction of other universities at home and abroad, and combining with the basic strategy of the construction of college library and the actual capabilities，the paper analyzes and designs the logical system architecture and structure. Then it gives the system development framework, and extends the library digital object model. Ultimately，it carries out the solutions to Fedora-based digital library.
In combination with the application and practice in the library of Guangxi University, this paper proposes the necessity of transplanting the platform of MELINETS II during actual working and gives a detailed explanation to the transplantation and optimization of the MELIENTS II system platform.
Due to the issue that today’s library management systems are unable to meet the needs of e-book acquisition operations, the paper puts forward the technical idea of developing an embedded e-book acquisition system based on MELINETS Ⅱ. The library management system adopts ADO and ODAC technologies to integrate distributed and heterogeneous e-book databases and then achieves the functions of unified checking, distribution acceptance and comprehensive statistical analysis of e-book acquisition.
