An Extended RELIEF-DISC for Handling of Incomplete Data to Improve the Classifier Performance

作者: E. Chandra Blessie , E. Karthikeyan , V. Thavavel

DOI: 10.1007/978-3-642-28926-2_60

关键词:

摘要: Feature selection is one of the important issues in machine learning. Some RELIEF based algorithms are considered as most successful for assessing quality features. RELIEF-DISC which was shown to be efficient estimating features, cannot handle incomplete data continuous In this paper, we propose an extended algorithm by introducing a new approach deal with noisy and sets. We investigate performance Decision Tree classifier imputing missing values algorithm. The datasets taken from UCI ML Repository.

参考文章(11)
D Kasprzak, G Kalton, The treatment of missing survey data Survey Methodology. ,vol. 12, pp. 1- 16 ,(1986)
Edgar Acuña, Caroline Rodriguez, The Treatment of Missing Values and its Effect on Classifier Accuracy Springer, Berlin, Heidelberg. pp. 639- 647 ,(2004) , 10.1007/978-3-642-17103-1_60
E. Chandra Blessie, E. Karthikeyan, RELIEF-DISC: An Extended RELIEF Algorithm Using Discretization Approach for Continuous Features international conference on emerging applications of information technology. pp. 161- 164 ,(2011) , 10.1109/EAIT.2011.39
P. Ranjit Jeba Thangaiah, B. Mehala, G R Govindarajulu, K. Vivekanandan, Selecting Scalable Algorithms to Deal With Missing Values ,(2009)
Jerzy W. Grzymala-Busse, Ming Hu, A Comparison of Several Approaches to Missing Attribute Values in Data Mining Lecture Notes in Computer Science. pp. 378- 385 ,(2000) , 10.1007/3-540-45554-X_46
J. Ross Quinlan, C4.5: Programs for Machine Learning ,(1992)
Peter Clark, Tim Niblett, The CN2 Induction Algorithm Machine Learning. ,vol. 3, pp. 261- 283 ,(1989) , 10.1023/A:1022641700528
Ralph Neuneier, Volker Tresp, Subutai Ahmad, Efficient Methods for Dealing with Missing Data in Supervised Learning neural information processing systems. ,vol. 7, pp. 689- 696 ,(1994)
Wolfgang Gaul, Frederick R. McMorris, David Banks, Phipps Arabie, "Classification, Clustering, and Data Mining Applications" ,(2004)