吳定謙
Ting-chien Wu
碩士論文(2009)
本體論為基之經驗知識表達與推理方法研究
Development of a Method for Ontology-Based Empirical Knowledge Representation and Reasoning
關鍵字 Keywords
本體論、知識表達、知識推理、經驗知識、OWL、本體推論
OWL、Ontology、Knowledge Reasoning、Knowledge Representation、Empirical Knowledge、Ontology reasoning
摘要
二十一世紀為「知識經濟」時代，企業競爭優勢已從過去有形的設備、資本與勞力轉換成現今無形的知識。隱性知識往往象徵企業價值所在，係為難以文件化的經驗知識，且常隱藏於個人心智模式中，若無法有效轉化為組織知識，則將隨著人員的離開而消失，造成公司重要智慧資產的流失。因此，將企業內個人的經驗知識，透過系統化方式轉化成組織層面的外顯知識，以有效管理與分享這些寶貴的個人經驗知識，實為企業創造更高知識價值的重要課題之一。
有鑑於此，本研究目的為發展一「本體論為基之經驗知識表達與推理方法」，針對經驗知識予以結構化的表達模式，協助知識需求者明確地了解經驗知識，以幫助未來問題解決及決策制定，且利用本體推論的技術，使經驗知識能有效的分享與再利用。
本研究之主要研究項目包括：（1）設計一本體論為基之多層次經驗知識表達模式，（2）建立一OWL為基之經驗知識本體，（3）發展一本體論為基之單層次經驗知識推理方法，及（4）發展一本體論為基之跨層次經驗知識推理方法。
Abstract
With the advent of knowledge economy, knowledge has become the most important asset of enterprises in the 21st century. There are also two types of knowledge: tacit and explicit. Tacit knowledge is personal knowledge that is often difficult to articulate and communicate. It has been characterized as being unarticulated and context specific. As compared to explicit knowledge, tacit knowledge is difficult to codify. But, articulating tacit knowledge is important for any organization so it does not lose that knowledge if the individual who owns it leaves the organization.
A empirical knowledge is a kind of tacit knowledge, which is the knowledge or understanding by experience. However, empirical knowledge is ill-structured, and therefore difficult to formalize. Since empirical knowledge is not easily represented or stored, empirical knowledge sharing is ineffective. Some examples of empirical knowledge include heuristics, skills, and special know-how.
The objective of this research is to develop a method to convert empirical knowledge into formatted representation, assist knowledge demanders to efficiently and accurately understand the useful meaning from empirical knowledge, to facilitate empirical knowledge sharing and reusing by ontology reasoning rule, and help problem-solving and decision-making in the future. The objective can be achieved through (1) design of a model for Ontology-Based Multi-Layers empirical knowledge representation, (2) develop of OWL-Based empirical knowledge ontology, (3) develop of a method for Ontology-Based Single-Layers empirical knowledge reasoning, and (4) develop of a method for Ontology-Based Cross-Layers empirical knowledge reasoning.