Comparison of Classification Algorithm Performances in Knowledge Mapping for Organization Experts

The knowledge map describes who has what knowledge (tacit), where the knowledge collected, and helps to learn the jobs and expertise in organization. In many organizations there is a lack of directions to manage knowledge and knowledge map. Data classification is one technique using in knowledge mapping. This paper proposes an approach for knowledge mapping of experts in organization by comparing the performances of four classification algorithms. The classification proposed in this paper in the domain of energy expert. We measured prediction performances by comparing algorithms with four classification algorithms: two types of decision trees (ID3, C4.5) and two rule-based (OneR and Prism). These four algorithms are measured their effectiveness with K-fold cross-validation method on their classification correctness. The results show that C4.5 algorithm is the best one in decision tree, and Prism is the best one in rule-based. Among the four algorithms, C4.5 is the best performance in classification for knowledge mapping. Future research and implication are also suggested.

[1]  Yiyu Yao,et al.  Classification Based on Logical Concept Analysis , 2006, Canadian Conference on AI.

[2]  Geoffrey Holmes,et al.  A diagnostic tool for tree based supervised classification learning algorithms , 1999, ICONIP'99. ANZIIS'99 & ANNES'99 & ACNN'99. 6th International Conference on Neural Information Processing. Proceedings (Cat. No.99EX378).

[3]  Martin J. Eppler Making knowledge visible through intranet knowledge maps: concepts, elements, cases , 2001, Proceedings of the 34th Annual Hawaii International Conference on System Sciences.

[4]  P. Bobbie,et al.  Classification of Arrhythmia Using Machine Learning Techniques , 2005 .

[5]  Zhe Huang,et al.  Applying AI technology and rough set theory to mine association rules for supporting knowledge management , 2003, Proceedings of the 2003 International Conference on Machine Learning and Cybernetics (IEEE Cat. No.03EX693).

[6]  Fu-Ren Lin,et al.  Knowledge map creation and maintenance for virtual communities of practice , 2006, Inf. Process. Manag..

[7]  Remo Aslak Burkhard,et al.  Beyond Excel and Powerpoint: knowledge maps for the transfer and creation of knowledge in organizations , 2005, Ninth International Conference on Information Visualisation (IV'05).

[8]  P. Radeva,et al.  Learning Models and Algorithms for Rule Extraction and Prediction , 2003 .

[9]  Gerhard Knolmayer,et al.  Document Classification Methods for Organizing Explicit Knowledge , 2002 .

[10]  Martin J. Eppler Toward a Pragmatic Taxonomy of Knowledge Maps: Classification Principles, Sample Typologies, and Application Examples , 2006, Tenth International Conference on Information Visualisation (IV'06).