| Ontology is emerged from philosophy, imported from knowledge engineering and then applied more and more in various computer-supported computation and application structure. In the network, the foundations of next-gen semantic web are also natural language understanding, semantic reasoning and semantic tagging under ontology theory guidance. Therefore, the establishment, perfect and evolution along with environment changes of ontology become the key point of current research. This thesis discusses how to make ontology evolve along with environment change under environment composed by services and finally make ontology fits in the environment. Mainly this thesis researches the capture of environment change, analyze the change, the emergence of evolution requirements, evolution strategy and execution. The key problems are analysis of environment change and the emergence of evolution requirements. The solution this thesis proposes is: 1.describe environment and its changes on the basis of OWL format XML file ontology. 2. Analyze the changes. Firstly transform the change related service into OWL-S described semantic service definition and then analyze OWL-S profile using Wordnet and importance algorithm to find out key concept of current definition. Finally apply semantic matching and ontology search algorithm with domain ontology concept from OWL to find out the differences. 3. Make out atomic and combined ontology evolution requirements to OWL domain ontology from these differences. 4. Analyze and filter these evolution requirements according to evolution strategy and then apply Kaon2 to do the evolution process to OWL domain ontology and keeps consistency. 5. Design and implementations of service environment driven ontology requirements generation system in detail. 6. Verifies the feasibility and correctness of this system by constructing service environment and simulate environment changes, and finally verifies the research content of this thesis.Through the research of this thesis, ontology can now automatically capture environment changes and generate evolution requirements so that it fits in the environment more. In the mean time, we integrate current ontology repository management, manual evolution tool, semantic search and semantic reasoning to form a complete ontology evolution management tool. |
