Comparison of Evolving Fuzzy Systems with an Ensemble Approach to Predict from a Data Stream

An approach to apply ensembles of regression models, built over the chunks of a data stream, to aid in residential premises valuation was proposed. The approach consists in incremental expanding an ensemble by systematically generated models in the course of time. The output of aged component models produced for current data is updated according to a trend function reflecting the changes of premises prices since the moment of individual model generation. The method employing general linear model, multiple layer perceptron, and radial basis function networks was empirically compared with evolving fuzzy systems designed for incremental learning from data streams. The results showed that evolving fuzzy systems and general linear models outperformed the ensembles built using artificial neural networks.