Knowledge capture for Thai word segmentation by using CommonKADS

Abstract

Knowledge capture is an important key in a business world where huge quantities of data are available via the Internet. Knowledge, as usable information, is a necessary element in the success of any organization. The recent growth of online information available in the form of academic paper related to algorithm and tool of Thai word segmentation distributed in various web sites, however it has not been organized in a systematic way. Thus, this study tries to propose a knowledge capture methods to support knowledge management activities. To perform the objectives of the study, knowledge engineering techniques take a very important role in the knowledge capture process in various ways such as to build knowledge model, to simplify access to the information their contain and better ways to represent the knowledge explicitly. In this study, many knowledge engineering methods have been compared to select a suitable method to be applied to solve the problem of knowledge capture from academic papers; i.e. SPEDE, MOKA and CommonKADS. The CommonKADS methodology is selected because it provides sufficient tools such as a model suite and templates for different knowledge intensive tasks. However, creating and representing knowledge model create difficulties to knowledge engineer caused the ambiguity and unstructured of the source of knowledge. Therefore, the objectives of this paper are to propose the methodology to capture knowledge for academic papers by using the knowledge engineering approach. The academic papers which content related to algorithm and tools of Thai word segmentation are used as a case study to demonstrate the proposed methodology.