The potential of neural networks for classification problems has been established by numerous successful applications reported in the literature. One of the major assumptions used in almost all studies is the equal cost consequence of misclassification. With this assumption, minimizing the total number of misclassification errors is the sole objective in developing a neural network classifier. Often this is done simply to ease model development and the selection of classification decision points. However, it is not appropriate for many real situations such as quality assurance, direct marketing, bankruptcy prediction, and medical diagnosis where misclassification costs have unequal consequences for different categories. In this paper, we investigate the issue of unequal misclassification costs in neural network classifiers. Through an application in thyroid disease diagnosis, we find that different cost considerations have significant effects on the classification performance and that appropriate use of cost information can aid in optimal decision making. A cross-validation technique is employed to alleviate the problem of bias in the training set and to examine the robustness of neural network classifiers with regard to sampling variations and cost differences.
Bias and variance play an important role in understanding the fundamental issue of learning and generalization in neural network modeling. Several studies on bias and variance effects have been published in classification and regression related research of neural networks. However, little research has been done in this area for time-series modeling and forecasting. We consider modeling issues related to understanding error components given the common practices associated with neural-network time-series forecasting. We point out the key difference between classification and time-series problems in consideration of the bias-plus-variance decomposition. A Monte Carlo study on the role of bias and variance in neural networks time-series forecasting is conducted. We find that both input lag structure and hidden nodes are important in contributing to the overall forecasting performance. The results also suggest that overspecification of input nodes in neural network modeling does not impact the model bias, but has significant effect on the model variance. Methods such as neural ensembles that focus on reducing the model variance, therefore, can be valuable and effective in time-series forecasting modeling.
We investigate the potential of artificial neural networks in diagnosing thyroid diseases. The robustness of neural networks with regard to sampling variations is examined using a cross-validation method. We illustrate the link between neural networks and traditional Bayesian classifiers. Neural networks can provide good estimates of posterior probabilities and hence can have better classification performance than traditional statistical methods such as logistic regression. The neural network models are further shown to be robust to sampling variations. It is demonstrated that for medical diagnosis problems where the data are often highly unbalanced, neural networks can be a promising classification method for practical use.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.