Basic Theories and Key Technologies of Semantic Web

After the introduction of the semantic web concept, researchers in Chinese institutions and universities have paid great attention to it. They have carried out in-depth research on the Semantic Web research content, some gratifying achievements have been made in the Basic Semantic Web theory and key platform technologies.

The Semantic Expression Theory of semantic web is based on AI's research on knowledge expression in the past 40 years. Meanwhile, it must satisfy the universality, dynamic nature and heterogeneity of Internet-oriented expressions. For example, RDF is a framework system, and OWL is based on description logic ). From the perspective of Knowledge Expression development: In the early 1960s S, the traditional first-order-logic was applied to knowledge expression. In the 1960s s, New Knowledge Representation Methods emerged, for example, Semantic Network (Quine), Framework System (Minsky), generative system, etc. In the late 1970s s-the beginning of 1990s s, there is a knowledge expression system represented by KL-ONE, based on the characteristics of Semantic Network and framework system, woods and brachman proposed a KL-ONE knowledge expression system. A large number of KL-ONE-based knowledge expression languages and practical systems emerged, such as: nikl, loom, classic, back, Kandor, kryton and so on; from the beginning of the 1990s s to the present: the expression of description logic as the basic knowledge has become the mainstream, the KL-ONE family gradually evolved into the current description logic (Germany dfki ). From a historical perspective, the semantic theory of Semantic Web expressions is based on description logic.

Semantic Web requires not only static information, but also dynamic services (Semantic Web Services). Therefore, semantic web logical reasoning includes not only static knowledge (Ontology) it also includes the reasoning of dynamic knowledge (service. The traditional descriptive logic is mainly used to represent and reason static knowledge, and cannot effectively represent and reason dynamic knowledge. Therefore, the traditional descriptive logic (such as shoiq (d) and SHOIN + (D) cannot provide a reasonable logical foundation for semantic web. That is to say, it is inadequate to use traditional descriptive logic as the logical foundation of semantic web. Based on the above reasons, a new description logic, namely, dynamic description logic (DDL), is proposed based on the characteristics and requirements of semantic web ), the DDL organically integrates Static and Dynamic Knowledge Representation and reasoning to form a unified formal framework. It has clear semantic features, provides deterministic reasoning services, and can effectively represent and reason dynamic processes and operating mechanisms. The dynamic description logic DDL is a promotion of the traditional description logic (adding the expression and reasoning for actions). It can provide a reasonable logical foundation for semantic web, it makes up for the shortcomings of traditional descriptive logic as the foundation of Semantic Web Logic. The Chinese Emy of Sciences (CAS) conducts a systematic and in-depth research on the dynamic description logic, and establishes a set of system theories including the syntax, semantics, and reasoning of the dynamic description logic.

The network provides a global platform for resource sharing and problem solving. To enable the network to realize its full potential, the network must be converted into a huge space for knowledge and service. In terms of web-based knowledge management, the computing department of the Chinese Emy of Sciences proposes a knowledge-based Data Model and Its query algebra, which can effectively manage and share heterogeneous resources and services in the network, the modified Knowledge Model consists of the Knowledge Model, resource instance library, and integrity constraints. The knowledge model is used to obtain recognized domain knowledge: Concepts and attributes, and integrity constraints are used to enrich the resource description, it specifies a series of rules to standardize Resource Modeling and obtain background knowledge related to the field. On this basis, they also establish a query language based on this knowledge model, it laid a good theoretical foundation for Web Knowledge Management.

Southeast University has carried out a series of work in terms of Web Knowledge Representation, Semantic Web Logic framework, Web Information Security and Semantic Web Services. Published semantic web articles at international conferences and journals such as www2003, future generation computer systems, and Journal of Web semantics. In addition, through active participation 3C discussions on draft technical specifications work, and comments on Semantic Web technical specifications have received the attention of the RDF core and webonto working groups.

Zhejiang University has studied the knowledge representation of Web Information. For the different nature of Web Information, the semantic expression model based on cognitive logic is used, the process Semantic Expression Model Based on temporal logic and the semantic expression or description methods of media stream data (non-text data) are studied.

In ontology-based application research, Zhejiang University established the first semantic-based database grid in China, supports dynamic semantic registration, distributed semantic query, and knowledge-level semantic browsing of database resources in a grid environment. Initially, the Chinese medicine library was established to provide the foundation for the integrated language system. Based on the semantic web, a semantic browser with ontology reasoning service was developed successfully.

In the key technology research of the ontology support platform, Tsinghua University researched and implemented a web-based Distributed Ontology System (wodos ), the topic of ontology distribution, language support, ontology granularity, and inference integration is studied in depth.

During the development of the ISO sc32 International Standard "MMF (Meta Model Framework) for ontology, wuhan University puts forward the theory and method of the ontology meta-model commitment, and is recognized and supported by international counterparts. 1.32.22.01.03.00 International Standardization Project; this project has studied the Model Driven Architecture Compilation Method Based on the meta-model, providing a theoretical and new method to solve the semantic interoperability problem in semantic web.

The University of Science and Technology of China has conducted in-depth research on Ontology-Based Semantic Modeling and ontology evolution.

 

 

 

 

2. Basic Web Service Theory Research

Semantic Web Services establish ontology-based service descriptions, discovery, and automatic assembly to implement automated and Intelligent Web Services. Research on Semantic Web Services includes:

Wuhan University proposed the theory and method of ontology meta-modeling, and submitted the UML commitment method of ontology and the text of the UML Profile for Owl standard solution to ISO International Standards Organization; in terms of Semantic Web Service Information Registration, Information resource library Organization and Management Platform Research, they developed based on ebXML/oasis, ISO 11179 (MCM) prototype of the web service and management platform of R & R, an international standard software component resource library.

Zhejiang University has made long-term Research on Distributed Network middleware and database sharing middleware, and has made some useful achievements in Web-Oriented Distributed Workflow Management. From 2002, I started to study Service-Oriented Workflow Technology and explored the architecture of distributed workflows in a grid environment. I proposed and implemented a semantic-based scalable information grid platform, defines and implements a group of Internet-Oriented Knowledge Services.

Fudan University has studied the establishment of a web service-based logistics management system.

In the study of key web services technologies, Tsinghua University has implemented web services to varying degrees by combining the J2EE + XML-based tourism e-commerce platform with the Web Service (www.luopan.com, service sharing and development can be achieved between partners in the industry. At present, they are studying Automatic Discovery Based on Ontology Web Services and the automatic implementation mechanism of Semantic Web Services in the first field.

The Institute of computing of the Chinese Emy of Sciences analyzed the shortcomings of the current service description language CDL, SDL, and larks, and combined with the characteristics of Web Services, Semantic Web Services, and grid services, sdlsin is a service description language with semantics, inheritance, and support for negotiation. Based on sdlsin, this paper studies the dynamic service matching problem and proposes four types of service matching algorithms, that is, syntax-based approximate service matching algorithms, semantic-based approximate service matching algorithms, precise service matching algorithms, and insert service matching algorithms. In addition, we use the descriptive logic's effective reasoning function, especially its effective judgment on the concept inclusion relationship, to combine it with service management. Fully utilizes descriptive logic to provide clear model-meaning of the Analects of Confucius and effective conceptual layered reasoning services, and proposes a service matching algorithm based on descriptive logic, this paper studies in detail how to use the theory and inference engine of descriptive logic to achieve automatic service layering and service matching, and proposes five service matching algorithms, that is, the main service matching algorithm, strong service matching algorithm, K-Nearest Neighbor service matching algorithm, K-approximate service matching algorithm, and m-Nearest Neighbor N-approximate service matching algorithm. These methods are both semantic-based service matching. The service layering mechanism is used to achieve effective and efficient service matching and overcome the shortcomings of Service matching based on semantic distance. Meanwhile, in order to correctly match the most satisfactory subject, a quality of service (QoS) model is proposed. This model objectively reflects the characteristics of non-functional attributes of services, various calculation and Evaluation Methods on the QoS model are studied. Based on the QoS model, a QoS-driven service matching algorithm is studied, this algorithm overcomes the shortcomings in the service matching algorithms proposed by K sycara, K arisha, g j wickler and Z Zhang.