Current Proceedings on Technology
Yazarlar: Onur Inan, Nihat Yilmaz, Mustafa Serter Uzer
Konular:-
Anahtar Kelimeler:Diabetes diseases,Support vector machine,K-means algorithm
Özet: The most important factors that prevent pattern recognition from functioning rapidly and effectively are the noisy and inconsistent data in databases. This article presents a new data elimination method based on clustering algorithms for diagnosis of diabetes diseases. In this method, K-means Algorithm is used for clustering based data elimination system for the elimination of noisy and inconsistent data and Support Vector Machines is used for classification. This newly developed approach was tested in the diagnosis of diabetes. The data set used in the diagnosis of these diseases is the Pima Indians Diabetes data sets obtained from the UCI database. The proposed system achieved 93.65% classification success rates from this data set. Classification accuracies for these data sets were obtained through using 10-fold cross-validation method. According to the result, the proposed method of performance is successful compared to other results attained, and seems very promising for pattern recognition applications.