Monte Carlo Evaluation of Classification Algorithms Based on Fisher's Linear Function in Classification of Patients With CHD

Classification comprises a variety of problems, which are solved in several ways. The need for automatic classification methods arises in a number of areas, from voice recognition, to modern automobiles, to the recognition of tumors through x-rays to assist doctors, by classifying emails as legitimate or spam. Due to the importance and complexity of such problems, there is a need for methods that provide greater accuracy and interpretability of the results. Among them the Boosting methods, which have emerged in the field of computation, work by sequentially applying a classification algorithm to reweighted versions of the training data set, giving greater weight to erroneous observations. The aim of this study was to study the Fisher Linear Discriminant Analysis (LDA) model and the same one using Boosting algorithm (AdaBoost) in the presence / absence of coronary heart disease (CHD) problem in patients. The criteria used to make the comparisons were sensitivity, specificity, false positive rate and false negative rate. In addition, Monte Carlo simulation was performed to calculate these rates in different partitions of the training set. The Boosting method was successfully applied in LDA and provided a higher sensitivity than the conventional LDA

Sign In


Reset Password

Please enter your username or email address, you will receive a link to create a new password via email.