This article proposes a method to gather merchandise information based on focused crawler, which integrates the Web topic link analysis and topic content semantic analysis. Through the statistical learning to Ontology during the crawling, the reference of domain-specific Ontology is optimized continuously. The experiment results show that comparing with other conventional crawling algorithms，this method is more effective, as it can prevent the occurrence of topic drift and bring a higher topic harvest rate.
This paper analyzes the basic model of Cross-Language Information Retrieval(CLIR) and the techniques of translation disambiguation. Then it optimizes translation disambiguation based on co-occurrences between pairs of terms and computing weights of terms,and constructs a CLIR system for book searching. The experiments show that the translation quality and the retrieval effectiveness are improved.
In this paper， a new model is proposed for distributed commodity information integration, which is built upon RESTful Web services and Mashup technology. And based on the model, a prototype system is developed for one-stop online Chinese book information search, and the applicability and efficiency of the model are tested in some level.
In this paper，a commodity information retrieval model is presented, which integrates semantic retrieval and multi-attribute decision method. Firstly, semantic similarity is computed by constructing semantic vector-space in order to realize the semantic consistency between retrieved result and customer’s query. Besides, TOPSIS method is also utilized to construct the comparison mechanism of commodity by calculating the utility value of each retrieved commodity. Finally, the experiment is conducted in terms of accuracy and customer acceptance rate, and the results verify the effectiveness of the model，which can improve the precision of the commodity information retrieval.
Making the structure model of service combination as logical starting point, the authors analyze and study the service discovery, matching and combination of digital libraries. They also explore the service combination application of digital library service combination based on OWL-S API.
Based on the introduction of DNET’s characteristics and function， the article analyzes in-depth the information space, digital object model of DRIVER, and the service-oriented architecture of DNET. And it finds out how to deal with the harvest information from heterogeneous repository and how to orchestrate different services for DNET.It also discusses the suitability of DNET, the role of standardization in digital repository federation and the services of digital repository infrastructure.
After revealing the current research status and features of Mashup tools, this paper introduces some relevant technologies and classification models of Mashup tools. And based on this, the paper designs a new classification model composed of Data Mashup tool, Presentation Mashup tool and Enterprise Mashup tool. In the same time, it analyzes the main functions and crucial technologies of typical Mashup tools.
This paper first introduces the concept of Semantic Mashup and the role of Semantic Web in the Mashup process. Then it summarizes the crucial technologies of Semantic Mashup, such as semantic data describing technology, semantic protocol specifications, Ontology-based Mashup reasoning technology. In addition, the paper analyzes some oversea on-going research projects, specifically for KC3 Browser, Bio2RDF, SBWS and Semantic REST. And finally, it gives some advice on how to promote the development of Semantic Mashup.
Based on the general definition and dissertation for semantic similarity measuring of Ontology, this paper makes a review of research on the Ontology-based semantic similarity measures，introduces edge counting measures, information content measures, feature-based measures and hybrid measures respectively. At last, it points out the direction of future work from macroscopic perspective.
This paper first introduces standard XML reference instants and XML data instants based on the weighted XML data model. Then it displays the expression ways of constraints in DTD. Furthermore, the paper also shows the approaches on how to implement similarity algorithm，with an emphasis on how to find out a matching node with standard XML reference instants and to get the similarity algorithm of standard XML reference instants and that of XML data instants.
This paper makes a retrospect for the evolution of semantic description of Web service, especially the evolution after OWL-S and WSDL-S. It also reviews the features and application status of all kinds of semantic description mechanism of Web service such as SAWSDL and WSMO-Lite，the description mechanisms of RESTful Web service as well as SA-REST and hREST, MicroWSMO. Finally, the author gives a future vision on development of semantic technology for semantic Web service.
This article makes an analysis of 2 461 papers published in the journal Scientometrics during 1978-2008. By means of scientometric methods and tools (TDA and Excel), the article presents the development trend of scientometrics, and maps the distributions of (the research-related) countries with charts for comparison and collaborations matrix. Then it analyzes 31 top institutions whose number of papers is not less than 20 and reveals the subject areas based on keywords.
The knowledge transfer network initiated by the knowledge transfer can be regarded as a knowledge network which is made up of individuals' knowledge citation relations. This paper proposes a knowledge network analysis framework for the knowledge transfer network based on a two-dimensional social network analysis, which can illustrate the evolutionary status of the knowledge transfer process. Then the paper discusses the difference between the social network analysis and the knowledge network analysis in the research of knowledge transfer process.
The system of arms information extraction based on Ontology consists of two parts: knowledge base and processing subsystem. It realizes the arms category determination based on text categorization, and the arms object determination based on named entity recognition. According to information extraction rules based on syntax and semantic constraint，it implements the information integration based on the Ontology in semantic level to some extent.
This paper imports data mining and information extraction techniques, introduces Web course resources quantitative evaluation index, and builds a model for quantitative evaluation system for real-time tracking and monitoring key indicators, which exports data as the supplementary of expert subjective evaluation. Preliminary experiment shows that this quantitative evaluation has a strong objectivity and feasibility.
In view of data migration in digital library, this paper analyzes the existed data environment of Library of Beihang University, and with the storage structure of DSpace, it proposes an approach for data migration based on .NET framework. This solution could integrate the data from heterogeneous data sources and transfer them to the new DSpace system by means of data extraction, data cleaning, DC generation and data transform. The implementation of this system could improve the information usage and data sharing effectively.
