以知識本體及語意規則建構健康膳食諮詢系統

Applying Ontology with Semantic Rules to Develop a Healthy Diet Consultation System

謝珮君(PEI-CHUN HSIEH)

指導教授 ： 戚玉樑

中原大學/商學院/資訊管理研究所/碩士(2008年)

https://doi.org/10.6840/cycu200900641

摘要

健康與日常膳食之間，常藉由營養素的供需關係形成關聯性，單一的疾病或可藉由對照方式獲得膳食之建議，但當影響健康的疾病類型增加時，膳食之間可能因營養素的衝突與排斥，造成諮詢建議的困難；其次，前述各領域的知識常因研究發現或新事證而發生更動，將使得既有的知識網絡產生連鎖性變動，間接造成知識資料在維護上的障礙。本研究針對上述問題，提出以知識本體(Ontology)為核心的知識庫：首先藉由知識工程方法，將此議題建立為一般化的知識模型，包括知識框架及推論規則；其次，將知識模型轉換至系統可處理的表達形式；再其次，事實知識將以實例方式，擴充知識庫的內涵；最後，利用推論機制協助推導內容的隱含關係及知識庫的異動更新。藉由健康膳食諮詢系統先行給定已知的事實資訊，系統進而執行推論程序比對知識庫中的實例特徵，並將所獲得的隱含知識回應予諮詢者；領域專家亦可藉由健康膳食諮詢系統的維護管理功能，進行知識庫的實例特徵更新，以及增加實際狀況之事實以擴充知識庫的完整性。本研究提供部份膳食、疾病、及營養素的實例測試及說明。由實驗結果顯示，由知識本體結合推論規則的設計，確可獲得正確的健康膳食知識，而此項設計也簡化後續知識庫的維護與異動，達到可延伸性與可信賴的知識庫諮詢系統。

關鍵字

語意規則 ； 知識本體 ； 諮詢系統 ； 健康膳食

並列摘要

Healthy diet plays a critical role to support human life. Many health problems can be prevented or alleviated with balanced nutrition. Poor or overloaded nutrition can have an injurious impact to cause serious diseases. A single food contains various kinds of nutrients that is healthy for some occasions but may be dangerous for some circumstances. The healthy diet knowledge is common senses but slight variations at times. Even when the relevance of healthy diet is obtained, translating it to practical dietary advice can be difficult and complicated. This study utilizes knowledge-intensive approaches to represent healthy diet models into ontologies. Additionally, a set of logic rules are developed to help identify semantic relationships between individuals. By following a knowledge-base creation process, the facts of dieses, nutrition, and diet can be inserted as factual knowledge. With the work of knowledge inference, reliable knowledge bases are established to support healthy diet in runtime. Give definitely known fact information in advance with the healthy diet consultation system, system carry out inference procedure than individuals characteristic in knowledge base, and reflect the knowledge of inferring obtained to consultants; The domain expert can also be with the functions of management of maintenance of the healthy diet consultation system, the individuals characteristic of carrying on the knowledge base is upgraded, and increase the fact of the real state in order to the integrity of expanding the knowledge base. One case study is provided to demonstrate the advantages of the proposed design. Experimental lessons show that rules in conjunction with ontological knowledge bases not only serve healthy-eating consolation needs, but also achieve a durable knowledge base and a reliable system.

並列關鍵字

Ontology ； Consultation system ； Healthy diet ； Semantic rules

參考文獻

[73] 戚玉樑，「知識擷取與知識表達協同程序於建構知識本體的概念架構」，資訊管理學報，13(2)，2006年，頁193-215。

[74] 戚玉樑、蔡明宏，「以文件為對象的概念萃取程序協助建立本體定義的雛型架構」，資訊管理學報，14(3)，2007年，頁47-66。

[2] A. G. Anagnostakis, M. Tzima, G. C. Sakellaris, D. I. Fotiadis, and A. C. Likas, “Semantics-based information modeling for the Health-Care administration sector: The citation platform,” IEEE Transactions on Information Technology in Biomedicine, vol. 9, no. 2, pp. 239-247, 2005.

[5] A. Baneyx, J. Charlet, and M. -C. Jaulent, “Building an Ontology of pulmonary diseases with natural language processing tools using textual corpora,” International Journal Of Medical Informatics, vol. 76, pp. 208-215, 2007.

[7] T. J. M. Bench-Capon and P. R. S. Visser, “Ontologies in legal information system; The need for explicit specifications of domain conceptualisations,” International Conference on Artificial Intelligence and Law, pp. 132-141, 1997.