Adaboost를 이용한 교사학습과 KNN-Adaboost를 이용한 준교사학습방법의 성능 비교

Performance comparison of supervised learning(semi-supervised learning(KNN-AdaboAodsta)b oost) and

In the field of pattern recognition, classification problem and clustering problem are linked to teacher learning in machine learning supervised learning and unsupervised learning problem. semisupervised learning study blended supervised learning and unsupervised learning is active. In this paper, propose a new semisupervised learning method model that combine K-NN classification with AdaBoost algorithm. K-NN-AdaBoost combination model(semisupervised learning) has more performance than adaboost model(supervised learning) using data that has only output data throw experiment that compare error rate of semisupervised learning with of supervised learning using data that has output data and data that has not output data