PAMC: Partitioning around Medoids for Classification

被引：0
作者：
Department of Computer Applications, Samrat Ashok Technological Institute, Vidisha -464001, India [1 ]
机构：
[1] Department of Computer Applications, Samrat Ashok Technological Institute
来源：
Inf. Technol. J. | 2006年 / 6卷 / 1102-1105期
关键词：
Classification; Clustering; Data mining;
D O I：
10.3923/itj.2006.1102.1105
中图分类号：
学科分类号：
摘要：
Integration of Association Rule Mining and Classification has produced Associative Classification techniques which in many cases have shown better classification accuracy than conventional classifiers. Motivated by this study we explore integration of clustering and classification techniques and propose a clustering based classification approach in this study. Our algorithm is based on Partitioning Around Medoids (PAM) clustering algorithm. Clustering process is unsupervised in general. We develop a classifier model by making use of available class label knowledge of training examples during clustering process. In present study we find accuracy of results is near to other popular classification methods. At present our algorithm can be applied to objects whose features are represented by continuous attributes. © 2006 Asian Network for Scientific Information.
引用
收藏
页码：1102 / 1105
页数：3
相关论文
共 9 条
[1]  
Cover T., Hart P., Nearest neighbor pattern classification, IEEE Transactions on Information Theory, 13, pp. 21-27, (1967)
[2]  
Han J., Kamber M., Data Mining: Concepts and Techniques, (2000)
[3]  
Hettich S., Bay S.D., The UCI KDD Archive, (1999)
[4]  
Kaufman L., Rousseau P., Finding Groups in Data, (1990)
[5]  
Lim T.S., Loh W.Y., Shih Y.S., A comparison of prediction accuracy, complexity and training time of thirty-three old and new classification algorithms, Machine Learning, 40, pp. 203-229, (2000)
[6]  
Swami D.K., Jain R.C., A survey of associative classification algorithms, ADIT. J. Eng., 12, pp. 51-55, (2005)
[7]  
Vilalta R., Achari M., Eick C., Class decomposition via clustering: A new framework for low-variance classifiers, Proceedings of the 3rd IEEE International Conference on Data Mining (ICDM03), pp. 673-676, (2003)
[8]  
Yin X., Han J., CPAR: Classification based on Predictive Association Rules, Proceeding of the International Conference on Data Mining, pp. 331-335, (2003)
[9]  
Zeng H.J., Wang X.H., Chen Z., Lu H., Ma W.Y., CBC: Clustering Based Text classification requiring minimal labeled data, Proceedings of ICDM, 2003, pp. 443-450, (2003)
1