Abstract

Abstract

Abstract
Accounting for misclassification costs is important in many practical applications of machine learning, and cost sensitive techniques for classification have been studied extensively. Utility-based learning provides a generalization of purely cost-based approaches that considers both costs and benefits, enabling application to domains with complex cost-benefit settings. However, there is little work on utility- or cost-based learning for regression. In this paper, we formally define the problem of utility-based regression and propose a strategy for maximizing the utility of regression models. We verify our findings in a large set of experiments that show the advantage of our proposal in a diverse set of domains, learning algorithms and cost/benefit settings.

Abstract

Abstract

Abstract
We propose a new linear regression model for interval-valued variables. The model uses quantile functions to represent the intervals, thereby considering the distributions within them. In this paper we study the special case where the Uniform distribution is assumed in each observed interval, and we analyze the extension to the Symmetric Triangular distribution. The parameters of the model are obtained solving a constrained quadratic optimization problem that uses the Mallows distance between quantile functions. As in the classical case, a goodness-of-fit measure is deduced. Two applications on up-to-date fields are presented: one predicting duration of unemployment and the other allowing forecasting burned area by forest fires.