Multilayer perceptron with local constraint as an emerging method in spatial data analysis

The use of Geographic Information Systems has revolutionalized the handling and the visualization of geo-referenced data and has underlined the critic role of spatial analysis. The usual tools for such a purpose are geostatistics which are widely used in Earth science. Geostatistics are based upon several hypothesis which are not always verified in practice. On the other hand, Artificial Neural Network (ANN) a priori can be used without special assumptions and are known to be flexible. This paper proposes to discuss the application of ANN in the case of the interpolation of a geo-referenced variable.

Using self organizing maps on compositional data

Self-organizing maps (Kohonen 1997) is a type of artificial neural network developedto explore patterns in high-dimensional multivariate data. The conventional versionof the algorithm involves the use of Euclidean metric in the process of adaptation ofthe model vectors, thus rendering in theory a whole methodology incompatible withnon-Euclidean geometries.In this contribution we explore the two main aspects of the problem:1. Whether the conventional approach using Euclidean metric can shed valid resultswith compositional data.2. If a modification of the conventional approach replacing vectorial sum and scalarmultiplication by the canonical operators in the simplex (i.e. perturbation andpowering) can converge to an adequate solution.Preliminary tests showed that both methodologies can be used on compositional data.However, the modified version of the algorithm performs poorer than the conventionalversion, in particular, when the data is pathological. Moreover, the conventional ap-proach converges faster to a solution, when data is \well-behaved".Key words: Self Organizing Map; Artificial Neural networks; Compositional data

Time Series Input Selection using Multiple Kernel Learning

In this paper we study the relevance of multiple kernel learning (MKL) for the automatic selection of time series inputs. Recently, MKL has gained great attention in the machine learning community due to its flexibility in modelling complex patterns and performing feature selection. In general, MKL constructs the kernel as a weighted linear combination of basis kernels, exploiting different sources of information. An efficient algorithm wrapping a Support Vector Regression model for optimizing the MKL weights, named SimpleMKL, is used for the analysis. In this sense, MKL performs feature selection by discarding inputs/kernels with low or null weights. The approach proposed is tested with simulated linear and nonlinear time series (AutoRegressive, Henon and Lorenz series).

Task 4: Testing Iowa Portland Cement Concrete Mixtures for the AASHTO Mechanistic-Empirical Pavement Design Procedure, May 2008

The present research project was designed to identify the typical Iowa material input values that are required by the Mechanistic-Empirical Pavement Design Guide (MEPDG) for the Level 3 concrete pavement design. It was also designed to investigate the existing equations that might be used to predict Iowa pavement concrete for the Level 2 pavement design. In this project, over 20,000 data were collected from the Iowa Department of Transportation (DOT) and other sources. These data, most of which were concrete compressive strength, slump, air content, and unit weight data, were synthesized and their statistical parameters (such as the mean values and standard variations) were analyzed. Based on the analyses, the typical input values of Iowa pavement concrete, such as 28-day compressive strength (f’c), splitting tensile strength (fsp), elastic modulus (Ec), and modulus of rupture (MOR), were evaluated. The study indicates that the 28-day MOR of Iowa concrete is 646 + 51 psi, very close to the MEPDG default value (650 psi). The 28-day Ec of Iowa concrete (based only on two available data of the Iowa Curling and Warping project) is 4.82 + 0.28x106 psi, which is quite different from the MEPDG default value (3.93 x106 psi); therefore, the researchers recommend re-evaluating after more Iowa test data become available. The drying shrinkage (εc) of a typical Iowa concrete (C-3WR-C20 mix) was tested at Concrete Technology Laboratory (CTL). The test results show that the ultimate shrinkage of the concrete is about 454 microstrain and the time for the concrete to reach 50% of ultimate shrinkage is at 32 days; both of these values are very close to the MEPDG default values. The comparison of the Iowa test data and the MEPDG default values, as well as the recommendations on the input values to be used in MEPDG for Iowa PCC pavement design, are summarized in Table 20 of this report. The available equations for predicting the above-mentioned concrete properties were also assembled. The validity of these equations for Iowa concrete materials was examined. Multiple-parameters nonlinear regression analyses, along with the artificial neural network (ANN) method, were employed to investigate the relationships among Iowa concrete material properties and to modify the existing equations so as to be suitable for Iowa concrete materials. However, due to lack of necessary data sets, the relationships between Iowa concrete properties were established based on the limited data from CP Tech Center’s projects and ISU classes only. The researchers suggest that the resulting relationships be used by Iowa pavement design engineers as references only. The present study furthermore indicates that appropriately documenting concrete properties, including flexural strength, elastic modulus, and information on concrete mix design, is essential for updating the typical Iowa material input values and providing rational prediction equations for concrete pavement design in the future.

Multiple Kernel Learning of Environmental Data. Case study: Analysis and Mapping of Wind Fields

The paper presents the Multiple Kernel Learning (MKL) approach as a modelling and data exploratory tool and applies it to the problem of wind speed mapping. Support Vector Regression (SVR) is used to predict spatial variations of the mean wind speed from terrain features (slopes, terrain curvature, directional derivatives) generated at different spatial scales. Multiple Kernel Learning is applied to learn kernels for individual features and thematic feature subsets, both in the context of feature selection and optimal parameters determination. An empirical study on real-life data confirms the usefulness of MKL as a tool that enhances the interpretability of data-driven models.

Statistical methods for river runoff prediction

Methods used to analyze one type of nonstationary stochastic processes?the periodically correlated process?are considered. Two methods of one-step-forward prediction of periodically correlated time series are examined. One-step-forward predictions made in accordance with an autoregression model and a model of an artificial neural network with one latent neuron layer and with an adaptation mechanism of network parameters in a moving time window were compared in terms of efficiency. The comparison showed that, in the case of prediction for one time step for time series of mean monthly water discharge, the simpler autoregression model is more efficient.

Kysynnän ennustaminen leipomotuotteille ja ennusteen tarkkuuden arviointi

Tässä diplomityössä tutkittiin kysynnän ennustamista Vaasan & Vaasan Oy:n tuotteille. Ensin työssä perehdyttiin ennustamiseen ja sen tarjoamiin mahdollisuuksiin yrityksessä. Erityisesti kysynnän ennustamisesta saatavat hyödyt käytiin läpi. Kysynnän ennustamisesta haettiin ratkaisua erityisesti ongelmiin työvuorosuunnittelussa.Työssä perehdyttiin ennustemenetelmiin liittyvään kirjallisuuteen, jonka oppien perusteella tehtiin koe-ennustuksia yrityksen kysynnän historiadatan avulla. Koe-ennustuksia tehtiin kuudelle eri Turun leipomon koe-tuotteelle. Ennustettavana aikavälinä oli kahden viikon päiväkohtainen kysyntä. Tämän aikavälin erityisesti peruskysynnälle etsittiin ennustetarkkuudeltaan parasta kvantitatiivista ennustemenetelmää. Koe-ennustuksia tehtiin liukuvilla keskiarvoilla, klassisella aikasarja-analyysillä, eksponentiaalisen tasoituksen menetelmällä, Holtin lineaarisella eksponenttitasoituksen menetelmällä, Wintersin kausittaisella eksponentiaalisella tasoituksella, autoregressiivisillä malleilla, Box-Jenkinsin menetelmällä ja regressioanalyysillä. Myös neuroverkon opettamista historiadatalla ja käyttämistä ongelman ratkaisun apuna kokeiltiin.Koe-ennustuksien tulosten perusteella ennustemenetelmien toimintaa analysoitiin jatkokehitystä varten. Ennustetarkkuuden lisäksi arvioitiin mallin yksinkertaisuutta, helppokäyttöisyyttä ja sopivuutta yrityksen monien tuotteiden ennustamiseen. Myös kausivaihteluihin, trendeihin ja erikoispäiviin kiinnitettiin huomiota. Ennustetarkkuuden huomattiin parantuvan selvästi peruskysyntää ennustettaessa, jos ensin historiadata esikäsittelemällä puhdistettiin erikoispäivistä ja –viikoista.

Aplicação de redes neurais artificiais na identificação de gasolinas adulteradas comercializadas na região de Londrina - Paraná

The multilayer perceptron network was used to classify the gasoline. The main parameters used in the classification were established by the Ordinance nº 309 of the Agência Nacional do Petróleo, but without informing the network the legal limits of these parameters. The network used had 10 neurons in a single hidden layer, learning rate of 0.04 and 250 training epochs. The application of artificial neural network served classify 100% of the commercialized gas in the region of Londrina-PR and to identify the tampered gasoline even those suspected of tampering.

Quantitative structure-retention relationships analysis of retention index of essential oils

Genetic algorithm and multiple linear regression (GA-MLR), partial least square (GA-PLS), kernel PLS (GA-KPLS) and Levenberg-Marquardt artificial neural network (L-M ANN) techniques were used to investigate the correlation between retention index (RI) and descriptors for 116 diverse compounds in essential oils of six Stachys species. The correlation coefficient LGO-CV (Q²) between experimental and predicted RI for test set by GA-MLR, GA-PLS, GA-KPLS and L-M ANN was 0.886, 0.912, 0.937 and 0.964, respectively. This is the first research on the QSRR of the essential oil compounds against the RI using the GA-KPLS and L-M ANN.

Determinação de constituintes químicos em madeira de eucalipto por Pi-CG/EM e calibração multivariada: comparação entre redes neurais artificiais e máquinas de vetor suporte

Multivariate models were developed using Artificial Neural Network (ANN) and Least Square - Support Vector Machines (LS-SVM) for estimating lignin siringyl/guaiacyl ratio and the contents of cellulose, hemicelluloses and lignin in eucalyptus wood by pyrolysis associated to gaseous chromatography and mass spectrometry (Py-GC/MS). The results obtained by two calibration methods were in agreement with those of reference methods. However a comparison indicated that the LS-SVM model presented better predictive capacity for the cellulose and lignin contents, while the ANN model presented was more adequate for estimating the hemicelluloses content and lignin siringyl/guaiacyl ratio.

PREDICTING THE BOILING POINT OF PCDD/Fs BY THE QSPR METHOD BASED ON THE MOLECULAR DISTANCE-EDGE VECTOR INDEX

The quantitative structure property relationship (QSPR) for the boiling point (Tb) of polychlorinated dibenzo-p-dioxins and polychlorinated dibenzofurans (PCDD/Fs) was investigated. The molecular distance-edge vector (MDEV) index was used as the structural descriptor. The quantitative relationship between the MDEV index and Tb was modeled by using multivariate linear regression (MLR) and artificial neural network (ANN), respectively. Leave-one-out cross validation and external validation were carried out to assess the prediction performance of the models developed. For the MLR method, the prediction root mean square relative error (RMSRE) of leave-one-out cross validation and external validation was 1.77 and 1.23, respectively. For the ANN method, the prediction RMSRE of leave-one-out cross validation and external validation was 1.65 and 1.16, respectively. A quantitative relationship between the MDEV index and Tb of PCDD/Fs was demonstrated. Both MLR and ANN are practicable for modeling this relationship. The MLR model and ANN model developed can be used to predict the Tb of PCDD/Fs. Thus, the Tb of each PCDD/F was predicted by the developed models.

SIMULTANEOUS DETERMINATION OF PROTOCATECHUIC ALDEHYDE AND PROTOCATECHUIC ACID USING THE LOCALIZED SURFACE PLASMON RESONANCE PEAK OF SILVER NANOPARTICLES AND CHEMOMETRIC METHODS

A simple and sensitive spectrophotometric method is proposed for the simultaneous determination of protocatechuic acid and protocatechuic aldehyde. The method is based on the difference in the kinetic rates of the reactions of analytes with [Ag(NH3)2]+ in the presence of polyvinylpyrrolidone to produce silver nanoparticles. The data obtained were processed by chemometric methods using principal component analysis artificial neural network and partial least squares. Excellent linearity was obtained in the concentration ranges of 1.23-58.56 µg mL-1 and 0.08-30.39 µg mL-1 for PAC and PAH, respectively. The limits of detection for PAC and PAH were 0.039 and 0.025 µg mL-1, respectively.

REDUÇÃO DE VARIÁVEIS DE ENTRADA DE REDES NEURAIS ARTIFICIAIS A PARTIR DE DADOS DE ANÁLISE DE COMPONENTES PRINCIPAIS NA MODELAGEM DE OXIGÊNIO DISSOLVIDO

The objective of this work is to demonstrate the efficient utilization of the Principal Components Analysis (PCA) as a method to pre-process the original multivariate data, that is rewrite in a new matrix with principal components sorted by it's accumulated variance. The Artificial Neural Network (ANN) with backpropagation algorithm is trained, using this pre-processed data set derived from the PCA method, representing 90.02% of accumulated variance of the original data, as input. The training goal is modeling Dissolved Oxygen using information of other physical and chemical parameters. The water samples used in the experiments are gathered from the Paraíba do Sul River in São Paulo State, Brazil. The smallest Mean Square Errors (MSE) is used to compare the results of the different architectures and choose the best. The utilization of this method allowed the reduction of more than 20% of the input data, which contributed directly for the shorting time and computational effort in the ANN training.

Spatial modeling techniques for mapping and characterization of acid sulfate soils

Acid sulfate (a.s.) soils constitute a major environmental issue. Severe ecological damage results from the considerable amounts of acidity and metals leached by these soils in the recipient watercourses. As even small hot spots may affect large areas of coastal waters, mapping represents a fundamental step in the management and mitigation of a.s. soil environmental risks (i.e. to target strategic areas). Traditional mapping in the field is time-consuming and therefore expensive. Additional more cost-effective techniques have, thus, to be developed in order to narrow down and define in detail the areas of interest. The primary aim of this thesis was to assess different spatial modeling techniques for a.s. soil mapping, and the characterization of soil properties relevant for a.s. soil environmental risk management, using all available data: soil and water samples, as well as datalayers (e.g. geological and geophysical). Different spatial modeling techniques were applied at catchment or regional scale. Two artificial neural networks were assessed on the Sirppujoki River catchment (c. 440 km2) located in southwestern Finland, while fuzzy logic was assessed on several areas along the Finnish coast. Quaternary geology, aerogeophysics and slope data (derived from a digital elevation model) were utilized as evidential datalayers. The methods also required the use of point datasets (i.e. soil profiles corresponding to known a.s. or non-a.s. soil occurrences) for training and/or validation within the modeling processes. Applying these methods, various maps were generated: probability maps for a.s. soil occurrence, as well as predictive maps for different soil properties (sulfur content, organic matter content and critical sulfide depth). The two assessed artificial neural networks (ANNs) demonstrated good classification abilities for a.s. soil probability mapping at catchment scale. Slightly better results were achieved using a Radial Basis Function (RBF) -based ANN than a Radial Basis Functional Link Net (RBFLN) method, narrowing down more accurately the most probable areas for a.s. soil occurrence and defining more properly the least probable areas. The RBF-based ANN also demonstrated promising results for the characterization of different soil properties in the most probable a.s. soil areas at catchment scale. Since a.s. soil areas constitute highly productive lands for agricultural purpose, the combination of a probability map with more specific soil property predictive maps offers a valuable toolset to more precisely target strategic areas for subsequent environmental risk management. Notably, the use of laser scanning (i.e. Light Detection And Ranging, LiDAR) data enabled a more precise definition of a.s. soil probability areas, as well as the soil property modeling classes for sulfur content and the critical sulfide depth. Given suitable training/validation points, ANNs can be trained to yield a more precise modeling of the occurrence of a.s. soils and their properties. By contrast, fuzzy logic represents a simple, fast and objective alternative to carry out preliminary surveys, at catchment or regional scale, in areas offering a limited amount of data. This method enables delimiting and prioritizing the most probable areas for a.s soil occurrence, which can be particularly useful in the field. Being easily transferable from area to area, fuzzy logic modeling can be carried out at regional scale. Mapping at this scale would be extremely time-consuming through manual assessment. The use of spatial modeling techniques enables the creation of valid and comparable maps, which represents an important development within the a.s. soil mapping process. The a.s. soil mapping was also assessed using water chemistry data for 24 different catchments along the Finnish coast (in all, covering c. 21,300 km2) which were mapped with different methods (i.e. conventional mapping, fuzzy logic and an artificial neural network). Two a.s. soil related indicators measured in the river water (sulfate content and sulfate/chloride ratio) were compared to the extent of the most probable areas for a.s. soils in the surveyed catchments. High sulfate contents and sulfate/chloride ratios measured in most of the rivers demonstrated the presence of a.s. soils in the corresponding catchments. The calculated extent of the most probable a.s. soil areas is supported by independent data on water chemistry, suggesting that the a.s. soil probability maps created with different methods are reliable and comparable.

Configuration et exploitation d'une machine émotionnelle

Dans ce travail, nous explorons la faisabilité de doter les machines de la capacité de prédire, dans un contexte d'interaction homme-machine (IHM), l'émotion d'un utilisateur, ainsi que son intensité, de manière instantanée pour une grande variété de situations. Plus spécifiquement, une application a été développée, appelée machine émotionnelle, capable de «comprendre» la signification d'une situation en se basant sur le modèle théorique d'évaluation de l'émotion Ortony, Clore et Collins (OCC). Cette machine est apte, également, à prédire les réactions émotionnelles des utilisateurs, en combinant des versions améliorées des k plus proches voisins et des réseaux de neurones. Une procédure empirique a été réalisée pour l'acquisition des données. Ces dernières ont fourni une connaissance consistante aux algorithmes d'apprentissage choisis et ont permis de tester la performance de la machine. Les résultats obtenus montrent que la machine émotionnelle proposée est capable de produire de bonnes prédictions. Une telle réalisation pourrait encourager son utilisation future dans des domaines exploitant la reconnaissance automatique de l'émotion.