Improved Accuracy In Data Mining Decision Tree Classification Using Adaptive Boosting (Adaboost)

The Decision Tree algorithm is a data mining method that often applied as solution to problem for classification. C5.0 has several weaknesses, including: the and other decision tree methods are biased towards modeling whose features have many levels, some problems model can occur such over-fit or under-fit challenges, big changes logic result in small training, experience inconvenience, imbalance causes low accuracy algorithm. boosting an iterative gives different weights distribution of training each iteration. Each iteration adds weight examples misclassification decreases correct classification, thereby effectively changing data. One example adaboost. purpose this research improve performance classification using adaptive (adaboost) predict hepatitis disease Confusion matrix. Tests been carried out with Matrix use Hepatitis dataset which rate 80.58% error 19.15%. Whereas Adaboost higher 82.98%, 17.02%. This difference caused by adaboost algorithm, because able change weak classifier into strong increasing observations, also reduce rate.