Clustering incomplete mixed numerical and categorical datasets is one of the challenging task. Traditional algorithms like k-prototype algorithm is used for mixed dataset, but is limited to only complete datasets. To handle such incomplete datasets we use modified squeezer algorithm, which includes the new dissimilarity measure for incomplete dataset with mixed numerical and categorical attribute values. In this modified squeezer algorithm it not only cluster the incomplete dataset, it also need not to input the missing values and need not to initialize any clusters at the beginning. This algorithm is compared with traditional k-prototype algorithm on benchmark datasets. The experimental results shows that the modified squeezer algorithm gives better accuracy than the traditional algorithm and also it overcomes the limitation of initial clusters.

mixed dataset, k-prototype, modified squeezer algorithm, dissimilarity measure

[1] M.V.Jagannatha Reddy and Dr. B. Kavitha, “clustering mixed numerical and categorical dataset using similarity weight and filter method”, International journal of Database Theory and Applications, vol-5, no-1 March- (2012), pp-121-134
[2] H. Zhexue, “Extension to the K-means algorithm for clustering large data sets with categorical values”, Data Mining and Knowledge Discovery, (1998), pp. 283-304.
[3] T. Covões and E. Hruschka, “A study of K-Means-based algorithms for constrained clustering”, Intelligent Data Analysis, vol. 17, no. 3, (2013), pp. 485-505.
[4] H. Zhexue, “Clustering large data sets with mixed numeric and categorical values”, Proceedings of the 1th pacific-Asia Conference on Knowledge Discovery & Data Mining. Singapore: World Scientific, (1997), pp. 21-34.
[5] W. Qian, W. Cheng and F. Zhenyuan, “Summary of k-means clustering algorithm”, Electronic Design Engineering, vol. 20, no. 7, (2012), pp. 21-24.
[6] C. Dan and W. Zhenhua, “A K-prototypes Algorithm Based on Improved Initial Center Points”, Computer Knowledge and Technology, (2010) November.
[7] C. Sotirios, “A fuzzy c-means-type algorithm for clustering of deal with mixed numeric and categorical attributes employing a probabilistic dissimilarity functional”, Expert Systems with Applications, vol. 38, no. 7, (2011), pp. 8684-8689.
[8] W. Fengmei and H. Lixia, “A Missing Data Imputation Method Based on Neighbor Rules”, Computer Engineering, vol. 38, no. 21, (2012).
[9] X. Fang and Z. Guizhu, “Clustering algorithm based on Modified Shuffled Frog Leaping Algorithm and K-means”, Computer Engineering and Applications, vol. 49, no. 1, (2013), pp. 176-180.
[10] Takashi Furukawa, Shin-ichi Ohnishi, and Takahiro Yamanoi “On a Fuzzy c-means Algorithm for Mixed Incomplete Data Using Partial Distance and Imputation” Proceedings of the International MultiConference of Engineers and Computer Scientists 2014 Vol I, IMECS 2014, March 12 - 14, 2014, Hong Kong.
[11] Vaishali H. Umathe, Prof. Gauri Chaudhary. “A Review on Incomplete Data And Clustering” (IJCSIT) International Journal of Computer Science and Information Technologies, Vol. 6 (2) , 2015, pp 1225-1227
[12] J. Twisk, M. de Boer, W. de Vente and M. Heymans, “Multiple imputation of missing values was not necessary before performing a longitudinal mixed-model analysis”, Journal of Clinical Epidemiology, vol. 66, no. 9, (2013), pp. 1022-1028.
[13] Wu Sen, Chen Hong and Feng Xiaodong “Clustering algorithm for incomplete data sets with mixed numeric and categorical Attributes” IJDTA, vol. 6 No. 5 2013, pp 95-104.
[14] W. Guoyin, “Expansion in the theory of rough set in incomplete information system”, Journal of computer research and development, vol. 33, no. 10, (2002), pp. 1239-1240.