以標準化Metadata 為核心發展金融機構Ontology 之探討

Abstract

索引典（Thesaurus）與知識本體論（Ontology）兩者皆可提供結構化的詞彙用以描述特定領域知識（Domain Knowledge）。然而想要進一步對領域知識創造出更具有豐富與彈性的知識內涵描述，知識本體論相較於索引典，它包含了比索引典更為廣泛的語意關係（Semantic Relationship），且其概念定義較為明確。因此，本研究將以某金融機構為例，利用既有之索引典發展出與金融業務相關之Ontology。並利用都柏林核心集（Dublin Core）為建立文件管理的詮釋資料（Metadata）；此外使用立法院國會索引典中與財務金融相關詞彙索引典作為輸入之標準，以節省重新開發特定領域 Ontology所需龐大的時間與成本，並建立個案金融機構的Ontology。 本文將以（Object Management Group，簡稱為OMG）所提出之MOF（Meta Object Facility）的四層架構為本研究之核心，並參考其他企業模型理論（Enterprise Framework Modeling）如：Zen of Meta-Modeling，並進一步將其擴充為五層之架構。本研究之五層架構由上而下分別為，第一層為Thing，第二層為資源描述架構RDF/RDF'S，第三層為網絡層Network，第四層為Metadata Ontology，第五層為Domain Knowledge，並以此架構做為企業知識管理系統之基礎並發展雛型系統。

Thesaurus can provide structured vocabularies for describing domain knowledge, however, if we want to create a knowledgech description of an object, such as required by the semantic web, thesaurus turn out to provide only pant of the knowledge needed. In this paper we try to solve the problems related to capture background knowledge for banking and finance resources. We scribe a case study in which we attempt to construct an Ontology for a subset of banking and finance, and use the thesaurus of Taiwan congress as input. At the same time, we also follow Dublin core as well as metadata standard. In this paper, we improve the 4 layers of framework, it is created by meta object facility (MOF) of Open anagement Group (OMG), we are extending the framework from 4 layers to 5 layers, including Thing, RDFS, networks, metadata Ontology and domain knowledge, we can apply this framework to develop knowledge management system in the enterprise.