K-NEAREST NEIGHBOR DENGAN ADAPTIVE BOOSTING DAN SYNTHETIC MINORITY OVERSAMPLING TECHNIQUE UNTUK KLASIFIKASI DATA TIDAK SEIMBANG

Breast cancer is non-skin that caused by several factors, including glandular ducts, cells, and breast support tissue, except for the skin of breast. if not treated immediately will be fatal sufferer, so early detection important patient's safety. The success depends on right diagnosis. Measurement accuracy a diagnosis can assisted statistical methods, namely classification. K-Nearest Neighbor classification algorithm based nearest neighbor easy to implement. In process, there are problems when faced with imbalanced data. Imbalanced data cause algorithms tend focus majority class. Data imbalance overcome using Synthetic Minority Oversampling Technique (SMOTE). Ensemble methods applied improve performance classification, one which Adaptive Boosting. This study applies combined Boosting SMOTE handling results this are, handle problem application produce an 80%, sensitivity 83,33%, specificity 66,67%, G-Mean value 74,54%. So it concluded