國內外的學者針對中醫藥典籍文章的整理,投入了大量的研究工作,將中醫藥典籍的文章數位化,資訊化,且隨著電腦及網路的普級,許多對中醫藥相關研究的學者,開始建立典籍知識庫的網站及軟體,但是目前現有的系統都只有提供中醫藥典籍的基本關鍵字查詢,而使得使用者在面對大量典籍資料時無法明確的判斷那些是相對關鍵字是重要的,也無法做更詳盡的關聯性分析。 本研究以知識本體論(Ontology)的概念,利用資訊擷取(Information Extraction) 及資料探勘(Data mining)的觀念,將1186萬字中醫藥典籍的數位文字資料做為資料庫,用關聯式資料庫的方法建置一個中醫藥典籍的知識庫查詢系統,本研究的知識庫所提供查詢功能,能以類別、年代、作者等做為主要欄位的關鍵字之關聯查詢,及布林查詢,在查詢關鍵字時能計算關鍵字在中醫藥典籍的詞頻來代表此關鍵字的重要性。 此知識庫的建立是為了能夠提供國內外學者能對於中醫藥典籍的查詢更有效率的找到使用者所想要的關鍵字資訊,藉此加速中醫藥典籍研究的發展。
The domestic scholar has done a massive research for ancient Chinese medicine encyclopedia in order to make those books more effective use. Furthermore, they digitalize ancient Chinese medicine encyclopedia into knowledge database in order to provide a internet inquiry service system. Those works are remarkable, but still lack in providing relational analysis. The purpose of our research is to build a relational knowledge database inquiry system for Chinese medicine books. It is going to base on structural Ontology, concept of Information Extraction and Data mining, and from these 1.18 million digitalize words, which published by Committee on Chinese Medicine and Pharmacy. In our system, we provide the inquiry function by keywords of category, dynasty, and author, also our system can find out the important term by calculated the word frequency. The last function is to exchange the medicine unit in different dynasty. In our research, we believe that it can provide a better way of efficiently searching through our system and also provide researchers with an aid to help with their research in finding the information they need from the our system.