On the basis of concept lattice theory, this article proposes a new method named Granularity Concept Analysis (GCA) based on formal concept analysis for the keywords analysis, and different granularity concept lattices are built with GCA focusing on the keywords of Ontology literatures in the open access resources. By mining of different granularity concept lattices, the analysis on high-frequency keywords and relative low-frequency keywords are executed in Ontology field, and the structure of knowledge and inherent association are discovered in Ontology field, in order to explore a new approach for literature keywords analysis.

The paper introduces the definition, characteristics and categories of linked data applications.After analyzing the key issues in linked data application architecture, it brings forward that we should pay attention to the data linkage reuse and creation, query results representation, query processing performance, data quality and validation for a linked data application. Finally, it presents some suggestions for linked data applications in library.

The foundation and framework of the community in P2P network is studied, and a community information retrieval system is proposed based on the feature of community. Then, the workflow of community-based information search in P2P is analyzed, and the major issues such as the selection of objective communities, the decision-making of neighbor peers,the realization of cross-community retrieval, and the maintenance of peers, are also discussed in detail. Finally, the future work is considered.

The paper mainly reviews the semantic relation extraction studies used key verbs(i.e.semantic verb),and introduces the methods of biomedical semantic relations extraction. The key verbs used in the present researches are manually selected, so the verb selection method involves more subjective factors, and influences the objectivity, accuracy and comprehensiveness of key verbs selection.Therefore, how to find these verbs automatically may be a key task of relations mining.

Based on the problems in typical term context similarity algorithm, the paper puts forward a new term similarity algorithm which constructs context patterns automatically by sentences dependencies analysis and then computes term similarity by mapping context patterns. The algorithm provides a better way to construct term context patterns. Meanwhile, term context characters are kept well in patterns. The paper also presents the specific implementation steps of new algorithm, and evaluates the algorithm on basis of gene engineering field experiment data set. Experiment result demonstrates that the algorithm has an obvious improvement in computing performance.

Based on the text clustering requirement from the embedded NSTL Recommending System, this paper researches on the BWP algorithm, and analyzes the shortage of the BWP. Then an improved algorithm is proposed to optimize the calculation of the distance within the single sample cluster. The improved algorithm enlarges the range of clusters number based on the BWP. Moreover, it changes the partial optimum into the whole optimum. At last, the test result shows it is effective and efficient.

This paper introduces a method for trademark retrieval system using combination features. In this method, a color histogram is first generated for color description, 2D Fourier transform and Zernike moment are then combined to apply to shape features extraction. This method is evaluated on MPEG-7 CE2B and 1 000 trademark images database and is compared with the existing methods. The experimental results show that the performance of this method is acceptable.

In this paper, relevance criteria oriented academic information retrieval system success model is constructed. The main components of the model include relevance criteria,academic retrieval system character, Information System Success Model (ISSM) and TEDS model. And the relationships in the model are analyzed and hypotheses are proposed.

Due to lack of the analytical function for spatial data in emergency information systems, based on the GIS functions, the paper designs a method for calculating spatial relationship between emergency spots and surrounding geographical factors, and then integrates these factors in the process of emergency information analysis. Finally, a filtering method is proposed to figure out the meaningful spatial information from huge amount of spatial predicates, and realize the spatial enabled function of emergency knowledge discovery.

This article studies dynamic evolution of the small world characteristic and degree correlation characteristic of a large OSN,then analyzes the changes in the behavior characteristics of Internet users and the impacts that the changes of these structural variables have on word of mouth in this OSN. In addition, it also indicates the issues that companies should pay attention to when implementing word of mouth by OSNs, and gives some suggestion for word of mouth of companies in different stages of OSNs which can help carry out word of mouth marketing.

This paper analyzes data standardization of multiple attribute evaluation in science and technology evaluation. The results show the evaluation score of side standardization is smaller than tunable standardization. Tunable standardization may increase evaluation score of lower level journals, linear standardization may affect evaluation score but cant affect evaluation ranking, and negative indicator standardization may change the evaluation ranking. The authors give a new linear negative indicator standardization method, and point out that the negative classical indicator standardization should be abandoned, and the evaluation institute should make known to the public of data standardization method so as to keep evaluation justice.

In order to provide more convenient use of library commercial and self-built databases for user, Beijing Normal University Library designs and realizes a new database navigation system based on MetaLib X-Server API. This database navigation system can provide more comprehensive database use information, including databases using PPT, use skills and consulting telephone. It also provides more useful functions such as “adding database link to user’s IE browser favorites websites”, “different databases of the same creator combining, grouping, hiding and displaying”.

For the current problem that FAQ has a single function, this article introduces how to construct a FAQ system based on KBPublisher, and describes how to resolve the issues that open-source software does not support searching in Chinese and how to integrate third-party applications through secondary development of the KBPublisher. Finally, it realizes a feature-rich local FAQ system which integrates real-time reference tools and Web access statistics.

This paper aims to construct FAQ-based question answering system for the book publishing,and both question indexing and retrieval problems are involved. In the question indexing phase, the part-of-speech tagging and semantic class based shallow technologies are used. In the question retrieval phase, vector space model and semantic analyzing based similarity methods are performed. Finally,the question answering system is implemented.

To solve the inconvenience of Chinese character input and the inefficiency of the OPAC search, this paper studies the pinyin search. Collecting, using, mining and sorting out the bibliographic data and the users search behavior, it builds a dynamic pinyin database to achieve pinyin input assist based on AJAX, searching prompts based on ASP and directing pinyin searching function. It can give search suggest to the users, simplify input and enhance the search experience.