Ontology Development for Rice Processing Knowledge Management

Abstract

This paper reports the results of the development for the rice processing ontology construction which is a knowledge base for describing and categorizing of rice research keywords. The method of ontology construction by the domain experts is divided into 2 parts, the first part is an ontology development process and the second part is an ontology evaluation process. The first part consists of  5 steps as follows, step 1: collect the domain concepts and terms from the literatures of the rice research, step 2: completely update the concepts and terms in the previous step, step 3: define the relationship between those terms, step 4: verify these results by the experts in the field and step 5: developed the rice processing ontology model. Nine categories of rice research concepts are identified. 1. Thai rice standard, 2. Rice quality, 3. Rice production process, 4. Properties of rice, 5. Types of rice product, 6. Rice compositions, 7. Rice products, 8. Measurement tools 9. Machines. Rice processing ontology has 907 concepts and 1863 terms. For the ontology evaluation process, the evaluation model is divided into 2 parts which are evaluated by 17 domain experts and user evaluation by comparing the effective information retrieval between the Ontology and the keyword searching. The result revealed that the ontology method could improve the efficiency of information retrieval based on Recall to 0.382 when compared with keyword search which gave the value of 0.074. The comparison shows that ontology searching is superior in retrieving the relevant information in the database system. The ontology technique will help in the classification and explanation of the word meaning for organizing the rice research database system. In addition, it can be developed for other specific purposes in the future.