透過您的圖書館登入
IP:3.134.77.195
  • 學位論文

以知識本體為基礎的個人化營養配方諮詢系統

An Ontology-Based Consultation System for Personal Nutrition Formula

指導教授 : 戚玉樑

摘要


由於食品技術與網路資訊的發達,提供現代人多元化的食物選擇空間,然而醫病與營養知識精緻化演進、及食品攝取多面向因素考量需求,使得一般大眾在面臨多樣化、差異化的飲食選擇時,常因為食物特性的不易辨識、各類食品間的相互關係不明確等因素,反而易於造成營養攝取選擇的困擾。為處理此類問題,本研究提出一個以知識本體為基礎的個人化營養配方諮詢系統,利用本體技術塑模食品營養領域的知識概念、分析概念間屬性關係,用以建置本體知識庫,透過推論技術將關聯複雜度高的隱性(Implicit)知識明確化,再與現行資料庫相連結,利用已有的資料為實例。藉由本體結構的語意層(Semantic layer)與資料庫的資料層(Data Layer)結合,以解決現有資訊系統在語意表達、使用者認知處理、知識分享與再利用等條件能力不足之問題。最後,利用Protégé作為網路本體語言的編輯工具,建立包含概念元素、屬性及描述邏輯的概念模型,並且實作個人化需求諮詢系統,系統使用推論引擎Pellet與語意網開發工具集Jena,依據本體架構與規則限制進行推論。在實作階段依序展示資訊擷得、本體建立與知識推論等過程,研究結果藉由情境問題的評估模式顯示,諮詢系統確能透過使用者營養補給、口感喜好等需求條件,有效滿足多樣性、彈性化及個人化的諮詢需求。

並列摘要


Development of food technology and network information offers modern pluralistic choices for food. However, advancement of knowledge in nutrition and medicine, adds various new factors to consider for food selection. For the general public it is difficult to distinguish food characteristics and to choose the right food for comprehensive nutrition. To address this problem, this study propose “An Ontology-Based Consultation System for Personal Nutrition Formula”. The study utilizes ontology technology to form a new knowledge concept in the field of food and nutrition, and analyses the benefit relationships. The Ontology-based knowledge system makes implied knowledge clear through inference technology. We combine the Semantic layer of ontology structure and the Data Layers by database. This enables solving of the problem through sharing and reuse of knowledge, adding to the user's cognitive treatment using the existing information system. Secondly, we utilize an editor of web-ontology language using Protégé tools to set up concept elements, attributes, and to describe a logic conceptual model. Further, we build an individualized demand consulting system that uses inference engine “Pellet” and a Java framework for building the Semantic Web application “Jena”, in order to implement inference according to the frame of ontology and restricted rules. At the construction stage, this study shows the processes of setting up approach, knowledge inference and information retrieval, etc. Finally, the system was evaluated by situation questions. The result of this study shows that the consulting system can really meet a variety of flexibility and individualized consultation demands effectively.

參考文獻


2. 洪建德 (1992),電腦輔助營養諮詢,中華營誌,17 (1&2),55-69
4. 張德明、吳芳禎、劉淑芬 (2004),以模糊推論建構營養膳食諮詢系統中配菜功能,中華營誌,29(1),59-71。
7. 戚玉樑 (2006),知識擷取與知識表達協同程序於建構本體的概念架構,資訊管理學報,13 (2),193-216.
5. 盧立卿、何錦雲、張芳馨、許依玲 (2001),整合臺灣食品營養相關資料:師大食品營養素資料庫管理系統之研發,中華營誌,26 (4),286-276
2. Berners-Lee T., Hendler J. and Lassila O. (2001). The Semantic Web, Scientific American, 284 (5), 34-44

延伸閱讀