Soft Computing Applied to Distributed Regression with Context-Heterogeneity
Héctor Allende-Cid, Raúl Monge and Héctor Allende
In this paper we present a distributed regression framework to model distributed data with different contexts. Different context is defined as the change of the underlying laws of probability in the distributed sources. Most state of the art methods do not take into account the different context and assume that the data comes from the same statistical distribution. We propose an aggregation scheme for models that are in the same neighborhood in terms of similarity by means of a Soft Computing approach, by using several methodologies like fuzzy membership functions, clustering algorithms, feed-foward neural network, stacked generalization and ensemble approaches. We conduct experiments with synthetic and real data sets to validate our proposal. Our proposed algorithm outperforms other models that follow a traditional approach.
Keywords: Distributed learning, soft computing approaches, regression from distributed sources.