| The Semantic Web is an extension of the current Web, in which Web resources are given computer-understandabale semantics, better enabling computers and people to work in cooperation. In order to recognize the Web resources and reason about them by computers in an intelligent and automatic way, how to represent knowledge of an application domain and reason about them have become very important research topics of the Semantic Web. As the knowledge representation model and the logical foundation of the Semantic Web, ontologies and Description Logics (DLs) play important roles in the Semantic Web, and have attracted considerable attention and have been used for knowledge representation and reasoning in various application domains.As the key techniques of the knowledge representation and reasoning in the Semantic Web, the success and proliferation of the Semantic Web largely depends on the construction and storage of ontologies and DL knowledge bases (KBs). On one hand, from the view of construction, many application domains contain much valuable information, and in order to achieve reusability and a high level of interoperability of knowledge, how to extract ontologies and DL KBs from the existing sources has become a key problem of the Semantic Web. On the other hand, from the view of storage, with the development of the Semantic Web, lots of DL KBs and ontologies came into being successively, and thus how to store them effectively has become more and more important. Databases are a main way of storing and managing data resources in many application domains. Therefore, the extraction and storage of DL KBs and ontologies supported by the database techniques have become very important research topics of the Semantic Web.However, information imprecision and uncertainty exist in many real-world applications. The occurrence and studies of fuzzy databases provides the theoretical solution for representing and handling imprecise and uncertain data. Moreover, in order to represent and reason on much fuzzy knowledge which is appearent in many appications of the Semantic Web, the significant research efforts in the Semantic Web community are recently directed toward the fuzzy extensions to DLs and ontologies. Currently, with the requirements of fuzzy knowledge representation and reasoning in the Semantic Web, the extraction and storage of fuzzy DL KBs and fuzzy ontologies become important issues to be solved. Fuzzy databases are an important form of modeling the imprecise and uncertain information in many real-world applications, and this resulted in numerous contributions of fuzzy database techniques in the last thirty years, mainly with respect to the representation and handling of fuzzy data, which provide the theoretical basis for the extraction and storage of fuzzy DL and ontology knowledge bases with fuzzy databases. However, it should be pointed out that, less research has been done in the extraction and storage of fuzzy DL KBs and fuzzy ontologies based on the fuzzy databases.To this end, this paper investigates the extraction and storage of fuzzy Description Logic and ontology knowledge bases supported by fuzzy databases, and includes the following two mian aspects:(i) this paper investigates how to extract fuzzy Description Logic and ontology knowledge bases from fuzzy database models, and focuses on five kinds of typical fuzzy database models (i.e., two kinds of fuzzy conceptual data models-fuzzy ER models and fuzzy UML data models, two kinds of fuzzy logical database models-fuzzy relational database models and fuzzy object-oriented database models, and fuzzy XML data models). Moreover, this paper also studies how to reason on these fuzzy database models with the extracted fuzzy knowledge bases;(ii) this paper investiagtes how to store fuzzy Description Logic and ontology knowledge bases in fuzzy relational databases. The innovative contributions of the paper are as follows:(1) The formal definitions and semantics of the above five kinds of typical fuzzy database models are proposed. Firstly, five kinds of fuzzy database models are further investigated, and some basic notions and structures of each model are introduced. Then, the formal definitions and semantics of five kinds of fuzzy database models are proposed, respectively, and the corresponding examples are also provided. The formal representations of five kinds of fuzzy database models are the basis of establishing correspondences between fuzzy database models and fuzzy DLs/ontologies.(2) Based on the different characters of five kinds of fuzzy database models, several fuzzy DLs are proposed, and a complete formal definition of fuzzy ontologies is given. The fuzzy DLs and the formal definition of fuzzy ontologies provide the knowledge representation model for the extracted fuzzy DL KBs and fuzzy ontologies from the fuzzy database models. Firstly, based on the own characters of five models, their corresponding fuzzy DLs are proposed (including a fuzzy DL called FDLR for representing two fuzzy conceptual data models, a fuzzy DL called f-ALCIQ(D) for representing two fuzzy logical database models, and a fuzzy DL f-ALCQwf-reg for representing the fuzzy XML data models). After describing the reasons for representing five kinds of fuzzy database models by the above three fuzzy DLs, the syntax, semantics, knowledge base and reasoning algorithm for each fuzzy DL are given. Then, in order to represent the extracted target fuzzy ontologies, a complete formal definition of fuzzy ontologies are proposed.(3) The approaches and tools for extracting fuzzy DL KBs from five kinds of fuzzy database models are proposed. Also, how to reason on the fuzzy database models through the reasoning mechanism of the fuzzy DLs is investigated in detail. Firstly, based on the formal definitions of five models and their corresponding fuzzy DLs, the approaches for extracting fuzzy DL KBs from five kinds of fuzzy database models are proposed, and the proofs of correctness of approaches and the corresponding extraction examples are given. Furthermore, following the proposed approaches, the corresponding extraction tools are developed. Finally, based on the extracted fuzzy DL KBs, after transforming the reasoning problems of fuzzy database models into the reasoning problems of fuzzy DLs, the reasoning tasks of fuzzy database models can be checked automatically by the reasoning mechanism of the fuzzy DLs.(4) The approaches for extracting fuzzy ontologies from five kinds of fuzzy database models are proposed in order to realize the automatic construction of fuzzy ontologies. Firstly, based on the formal definitions of five models and fuzzy ontologies, the formal approaches for extracting fuzzy ontologies from five kinds of fuzzy database models are proposed, respectively. Then, the proofs of correctness of approaches and the corresponding extraction examples are given. Note that, since the logical foundation of the fuzzy ontology is the fuzzy DL, how to reason on fuzzy database models based on the extracted fuzzy ontologies, which are similar with the fuzzy DLs, are not included here.(5) The storage approach and tool of fuzzy knowledge bases (including fuzzy OWL ontologies and fuzzy DL f-SHCHOIN(D) and its sub-languages KBs) in fuzzy relational databases are developed. Firstly, based on the widespread use and mature techniques of fuzzy relational databases, and by classifying different constructors of a fuzzy knowledge base, a storage schema of fuzzy knowledge bases in fuzzy relational databases is proposed. On this basis, an example is provided throughout the paper to well explain the approach. The correctness and quality of the approach are discussed. Finally, following the proposed approach, a prototype tool that can automatically store a fuzzy knowledge base in a fuzzy relational database is implemented. |
PDF Full Text Request