A NOVEL LNS SEMI SUPERVISED LEARNING ALGORITHM FOR DETECTING BREAST CANCER
Soumya Aruna · 2012
Semi supervised learning is a relatively new area in machine learning which represents the blend of supervised and unsupervised learning. It has the potential of reducing the need of expensive labeled data whenever only a small set of labeled examples are available. In this paper semi supervised learning algorithm combining Logical data analysis based on complete binary tree with Naive Bayes and SVM with learning based on both labeled and unlabeled data is proposed for detecting breast cancer. Few labeled data are used as supportive set to build the diagnostic model which is used for classifying the unlabeled data. Wisconsin breast cancer dataset from the UCI machine learning depository is used for the experiment. This algorithm yielded an accuracy of 98.7% for unlabeled samples.