近代中國教會史知識本體建構方法之研究： 以二十世紀初The Chinese Recorder為例

Abstract

歷史文件包含的知識內容豐富且廣泛，因此有許多研究將焦點放在知識分類架構的產生，但以往許多研究往往只產出了內容分類架構，稱之為 taxonomy。但是這種分類架構大部分只是表達出樹狀的基本分類結構，而少了概念之間的多樣性關係；也就是說，這樣的分類結構並沒有包含更複雜的知識表徵。本研究的目的就是要發展出一套有效且自動的建構程序來產生領域知識本體(Ontology)架構，一方面可以減輕人工介入的成分與專家的負擔，另一方面則可以加速知識本體架構的產出速度與規模。本研究以一個特殊又具有歷史研究意義的資料集，稱為The Chinese Recorder，做為實驗對象。藉著定義並建構出The Chinese Recorder此標的領域的知識表徵，表達出這些歷史文件中更豐富的知識內涵。同時本研究也以語意網路中描述知識本體的標準OWL來進行知識本體的表達。The purpose of this research is to develop an effective and automatic construction process to generate the ontology of a knowledge domain. A historical document always contains very rich and extensive knowledge. Therefore, much research has focused on the generation the classification structure of knowledge; however, most of the research only generated content classification framework known as taxonomy. This kind of classification framework only expresses the basic classification tree structure, but lacks diversified conceptual relationships, in other words, this kind of classification structure does not contain sophisticated knowledge features. Based on a very special and historical meaningful dataset called “The Chinese Recorder” as an experimental target, this study develops an automatic ontology construction process to reduce individual manual intervention and burden of experts. Furthermore, it can accelerate the output speed and scale of the ontological knowledge framework. By defining and constructing the knowledge representation of the domain ontology of the Chinese Recorder, this study expresses the rich knowledge content of the historical document. Meanwhile, it also produces a semantically-oriented and OWL-based ontology.