A multi-model approach to analysis of environmental phenomena

Abstract

This paper introduces a novel data-driven methodology named Evolutionary Polynomial Regression (EPR), which permits the multi-purpose modelling of physical phenomena, through the simultaneous solution of a number of models. Multipurpose modelling or "multi-modelling", enables the user to make a more robust choice of those models aimed at (a) the knowledge based on data modelling, (b) on-line and offline forecasting, and (c) data augmentation (i.e. infilling of missing data in time series). This methodology is particularly useful in modelling environmental phenomena, for which it is usually impossible to obtain physical data at a laboratory scale. In particular, the non-linearity of phenomena and non Gaussian nature of background noise make on-line forecasting complex, and where data are available, they often contain discontinuities (i.e. missing data). The use of EPR in modelling and analysis is illustrated by application to a case study containing all these limitations. The application of EPR to thermal behaviour of a stream gives not only a good physical insight of the phenomenon, but also allows infilling of missing data, resulting in good models that forecast the water temperature.

Traducción automática del lenguaje dactilológico de sordos y sordomudos mediante sistemas adaptativos (AutomAtic trAnslAtion of the dActilologic language of hearing impaired by adaptive systems)

“…Para tal efecto, se evaluaron cinco modelos basados en los principios de la inteligencia computacional, entre los que se cuentan las redes neuronales MADALINE y Base Radial [22], los modelos borrosos adaptativos de tipo ANFIS [23] y Takagi Sugeno [24,25], así como el modelo por evolución EPR (Evolutionary Polynomial Regression) [30]. Para establecer esta relación funcional, cada modelo fue configurado para un total de 16 valores de entrada y una sola salida, lo que arrojó como resultado la reducción de dimensionalidad en un factor de 16 a 1, para los 100 instantes de tiempo de los cuales se compone tanto la matriz de entrada como la señal de referencia.…”

Advances in data-driven analyses and modelling using EPR-MOGA

“…0). This is particularly useful in modelling systems where there is a high probability that the negative coefficient values (a j , 0) are selected to balance the particular realization of errors related to the finite training dataset (Giustolisi et al 2007). undertaken as if the models were linear, and this is made possible by the particular structure of the EPR models.…”

Asset deterioration analysis using multi-utility data and multi-objective data mining

“…The key objective here is therefore to find a systematic means to avoid the problem of over-fitting. In the original singleobjective EPR More recently in Giustolisi et al (2007) the idea of using a multi-objective strategy to constrain a j . 0 during parameter estimation for improving model selection (i.e.…”

“…0). This is particularly useful in modelling systems where there is a high probability that the negative coefficient values (a j , 0) are selected to balance the particular realization of errors related to the finite training dataset (Giustolisi et al 2007). …”