Chinese Journal of Computers   Full Text
Title	The Methodology of Mining Cognitive Maps Based on Data Resources
Authors	CHEN Zhuang1) Montazemi Ali R.2)
Address	1)(School of Computer Science and Engineering, Chongqing Institute of Technology, Chongqing 400050) 2)(Business School, McMaster University, Hamilton ON L8S 4M4, Canada)
Year	2007
Issue	No.8(1446—1454)
Abstract &   Background	Abstract Cognitive Map (CM) is a new kind of method of knowledge management, which has many advantages such as: it is relative easy to use for representing structured knowledge, the inference mechanism can be computed by numeric matrix operation, etc. However, in order to exhibit these advantages about CM, the first step is that the corrected CMs must be obtained. Traditional approaches for obtaining the CMs, including questionnaire method, brainstorming method, and sample learning method, always rely on experience of domain experts. Because these methods put much emphasis on the subjective factors and neglect the objective data resources, they always lose some information. Therefore, this paper proposes a new methodology of mining the CMs based on data resources, which mainly includes database preprocessing technology, optimization algorithm for weight coefficients, and simplification strategy for CMs. The experimental results show that: the new method can mine all possible relationships among all nodes to form the CMs, and can also simplify it according to the significant degree of those relationships; the CMs mined by the new method has more information than the CMs obtained by traditional approaches. keywords cognitive map; data mining; neural network; database; learning arithmetic background This paper is partially supported by National Science Foundation of Chongqing under Grand No. 2005BB2083 entitled "Study on Management Technologies of Data Resources". The authors and their groups have done some works in areas of cognitive map (CM) and data resources management, such as CM’ application for information requirements analysis, the relationship between CM and data mining, and CM’ feedback mechanism. Recently, CM have been gained considerable research interest and applied to many areas, but the methods of obtaining the CMs, including questionnaire, brainstorming and sample learning method, have some deficiencies, such as neglecting the objective data resources and losing information. Therefore, the paper focus on the problem of methodology of mining CMs based on data resources, which is significant step to exhibit CM’ advantages and make it use in some practical domains. The authors’ work is to create a new methodology of mining CMs from data resources. The paper firstly proposes the structure of mining CMs, then gives initialization technologies for the database and creates optimization algorithm for weight coefficients. At last, the authors design simplification strategy of CMs. The experimental results show the methodology is effective.