Abstract

Abstract
This paper describes a new method for dealing with multiple regression problems. This method integrates a clustering technique with regression trees, leading to what we have named as clustered regression trees. We use the clustering method to form sub-samples of the given data that are similar in terms of the predictor variables. By proceeding this way we aim at facilitating the subsequent regression modeling process based on the assumption of a certain smoothness of the regression surface. For each of the found clusters we obtain a different regression tree. These clustered regression trees can be used to predict the response value for a query case by an averaging process based on the cluster membership probabilities of the case. We have carried out a series of experimental comparisons of our proposal that have shown a significant predictive accuracy advantage over the use of a single regression tree.

Abstract
The paper begins with the discussion on why we should be concerned with machine learning in the context of distributed AI. The rest of the paper is dedicated to various problems of multi-agent learning. First, a common framework for comparing different existing systems is presented. It is pointed out that it is useful to distinguish when the individual agents communicate. Some systems communicate during the learning phase, others during the problem solving phase, for example. It is also important to consider how, that is in what language, the communication is established. The paper analyses several systems in this framework. Particular attention is paid to previous work done by the authors in this area. The paper covers use of redundant knowledge, knowledge integration, evaluation of hypothesis by a community of agents and resolution of language differences between agents. © Springer-Verlag Berlin Heidelberg 1991.

Abstract
This article presents an alternative approach to the problem of regression. The methodology we describe allows the use of classification algorithms in regression tasks. From a practical point of view this enables the use of a wide range of existing machine learning (ML) systems in regression problems. In effect, most of the widely available systems deal with classification. Our method works as a pre-processing step in which the continuous goal variable values are discretised into a set of intervals. We use misclassification costs as a means to reflect the implicit ordering among these intervals. We describe a set of alternative discretisation methods and, based on our experimental results, justify the need for a search-based approach to choose the best method. The discretisation process is isolated from the classification algorithm, thus being applicable to virtually any existing system. The implemented system (RECLA) can thus be seen as a generic pre-processing tool. We have tested RECLA with three different classification systems and evaluated it in several regression data sets. Our experimental results confirm the validity of our search-based approach to class discretisation, and reveal the accuracy benefits of adding misclassification costs. © 1997 Elsevier Science B.Y.

Abstract
We present a methodology that enables the use of classification algorithms on regression tasks. We implement this method in system RECLA that transforms a regression problem into a classification one and then uses an existent classification system to solve this new problem. The transformation consists of mapping a continuous variable into an ordinal variable by grouping its values into an appropriate set of intervals. We use misclassification costs as a means to reflect the implicit ordering among the ordinal values of the new variable. We describe a set of alternative discretization methods and, based on our experimental results, justify the need for a search-based approach to choose the best method. Our experimental results confirm the validity of our search-based approach to class discretization, and reveal the accuracy benefits of adding misclassification costs.

Abstract