Optimal Linear Regression on Classifier Outputs
نویسندگان
چکیده
We consider the combination of the outputs of several classifiers trained independently for the same discrimination task. We introduce new results which provide optimal solutions in the case of linear combinations. We compare our solutions to existing ensemble methods and characterize situations where our approach should be preferred.
منابع مشابه
Predicting corporate financial distress based on integration of support vector machine and logistic regression
The support vector machine (SVM) has been applied to the problem of bankruptcy prediction, and proved to be superior to competing methods such as the neural network, the linear multiple discriminant approaches and logistic regression. However, the conventional SVM employs the structural risk minimization principle, thus empirical risk of misclassification may be high, especially when a point to...
متن کاملApplication of Linear Regression and Artificial NeuralNetwork for Broiler Chicken Growth Performance Prediction
This study was conducted to investigate the prediction of growth performance using linear regression and artificial neural network (ANN) in broiler chicken. Artificial neural networks (ANNs) are powerful tools for modeling systems in a wide range of applications. The ANN model with a back propagation algorithm successfully learned the relationship between the inputs of metabolizable energy (kca...
متن کاملBoosting SVM Classifiers with Logistic Regression
The support vector machine classifier is a linear maximum margin classifier. It performs very well in many classification applications. Although, it could be extended to nonlinear cases by exploiting the idea of kernel, it might still suffer from the heterogeneity in the training examples. Since there are very few theories in the literature to guide us on how to choose kernel functions, the sel...
متن کاملAdjusting the Outputs of a Classifier to New a Priori Probabilities May Significantly Improve Classification Accuracy: Evidence from a multi-class problem in remote sensing
In the present study, we introduce a simple iterative procedure that allows to correct the outputs of a classifier with respect to the new a priori probabilities of a new data set to be scored, even when these new a priori probabilities are unknown in advance. We also show that a significant increase in classification accuracy can be observed when using this procedure properly. More specificall...
متن کاملCombining texture features from the MLO and CC views for mammographic CADx
The purpose of this study was to investigate approaches for combining information from the MLO and CC mammographic views for Computer-aided Diagnosis (CADx) algorithms. Feature level and classifier output level combinations were explored. Linear discriminant analysis (LDA) with step-wise feature selection from a set of Haralick’s texture features was used to develop classifiers for distinguishi...
متن کامل