Linguistic models for decision support

Linguistic modelling is a rather new branch of mathematics that is still undergoing rapid development. It is closely related to fuzzy set theory and fuzzy logic, but knowledge and experience from other fields of mathematics, as well as other fields of science including linguistics and behavioral sciences, is also necessary to build appropriate mathematical models. This topic has received considerable attention as it provides tools for mathematical representation of the most common means of human communication - natural language. Adding a natural language level to mathematical models can provide an interface between the mathematical representation of the modelled system and the user of the model - one that is sufficiently easy to use and understand, but yet conveys all the information necessary to avoid misinterpretations. It is, however, not a trivial task and the link between the linguistic and computational level of such models has to be established and maintained properly during the whole modelling process. In this thesis, we focus on the relationship between the linguistic and the mathematical level of decision support models. We discuss several important issues concerning the mathematical representation of meaning of linguistic expressions, their transformation into the language of mathematics and the retranslation of mathematical outputs back into natural language. In the first part of the thesis, our view of the linguistic modelling for decision support is presented and the main guidelines for building linguistic models for real-life decision support that are the basis of our modeling methodology are outlined. From the theoretical point of view, the issues of representation of meaning of linguistic terms, computations with these representations and the retranslation process back into the linguistic level (linguistic approximation) are studied in this part of the thesis. We focus on the reasonability of operations with the meanings of linguistic terms, the correspondence of the linguistic and mathematical level of the models and on proper presentation of appropriate outputs. We also discuss several issues concerning the ethical aspects of decision support - particularly the loss of meaning due to the transformation of mathematical outputs into natural language and the issue or responsibility for the final decisions. In the second part several case studies of real-life problems are presented. These provide background and necessary context and motivation for the mathematical results and models presented in this part. A linguistic decision support model for disaster management is presented here – formulated as a fuzzy linear programming problem and a heuristic solution to it is proposed. Uncertainty of outputs, expert knowledge concerning disaster response practice and the necessity of obtaining outputs that are easy to interpret (and available in very short time) are reflected in the design of the model. Saaty’s analytic hierarchy process (AHP) is considered in two case studies - first in the context of the evaluation of works of art, where a weak consistency condition is introduced and an adaptation of AHP for large matrices of preference intensities is presented. The second AHP case-study deals with the fuzzified version of AHP and its use for evaluation purposes – particularly the integration of peer-review into the evaluation of R&D outputs is considered. In the context of HR management, we present a fuzzy rule based evaluation model (academic faculty evaluation is considered) constructed to provide outputs that do not require linguistic approximation and are easily transformed into graphical information. This is achieved by designing a specific form of fuzzy inference. Finally the last case study is from the area of humanities - psychological diagnostics is considered and a linguistic fuzzy model for the interpretation of outputs of multidimensional questionnaires is suggested. The issue of the quality of data in mathematical classification models is also studied here. A modification of the receiver operating characteristics (ROC) method is presented to reflect variable quality of data instances in the validation set during classifier performance assessment. Twelve publications on which the author participated are appended as a third part of this thesis. These summarize the mathematical results and provide a closer insight into the issues of the practicalapplications that are considered in the second part of the thesis.

Adaptive Markov chain Monte Carlo and Bayesian filtering for state space models

This thesis is concerned with the state and parameter estimation in state space models. The estimation of states and parameters is an important task when mathematical modeling is applied to many different application areas such as the global positioning systems, target tracking, navigation, brain imaging, spread of infectious diseases, biological processes, telecommunications, audio signal processing, stochastic optimal control, machine learning, and physical systems. In Bayesian settings, the estimation of states or parameters amounts to computation of the posterior probability density function. Except for a very restricted number of models, it is impossible to compute this density function in a closed form. Hence, we need approximation methods. A state estimation problem involves estimating the states (latent variables) that are not directly observed in the output of the system. In this thesis, we use the Kalman filter, extended Kalman filter, Gauss–Hermite filters, and particle filters to estimate the states based on available measurements. Among these filters, particle filters are numerical methods for approximating the filtering distributions of non-linear non-Gaussian state space models via Monte Carlo. The performance of a particle filter heavily depends on the chosen importance distribution. For instance, inappropriate choice of the importance distribution can lead to the failure of convergence of the particle filter algorithm. In this thesis, we analyze the theoretical Lᵖ particle filter convergence with general importance distributions, where p ≥2 is an integer. A parameter estimation problem is considered with inferring the model parameters from measurements. For high-dimensional complex models, estimation of parameters can be done by Markov chain Monte Carlo (MCMC) methods. In its operation, the MCMC method requires the unnormalized posterior distribution of the parameters and a proposal distribution. In this thesis, we show how the posterior density function of the parameters of a state space model can be computed by filtering based methods, where the states are integrated out. This type of computation is then applied to estimate parameters of stochastic differential equations. Furthermore, we compute the partial derivatives of the log-posterior density function and use the hybrid Monte Carlo and scaled conjugate gradient methods to infer the parameters of stochastic differential equations. The computational efficiency of MCMC methods is highly depend on the chosen proposal distribution. A commonly used proposal distribution is Gaussian. In this kind of proposal, the covariance matrix must be well tuned. To tune it, adaptive MCMC methods can be used. In this thesis, we propose a new way of updating the covariance matrix using the variational Bayesian adaptive Kalman filter algorithm.

An improved HPLC method for the quantitation of 6-mercaptopurine and its metabolites in red blood cells

A procedure is described for the rapid determination of the intra-erythrocyte concentration of 6-mercaptopurine (6-MP) and its metabolites, 6-thioguanine nucleotides (6-TGN) and 6-methylmercaptopurine (6-MMP). Erythrocytes (8 x 10(8) cells) in 350 µl Hanks solution containing 7.5 mg dithiothreitol were treated with 50 µl 70% perchloric acid. The precipitate was removed by centrifugation (13,000 g) and the supernatant hydrolyzed at 100°C for 45 min. After cooling, 100 µl was analyzed directly by HPLC using a Radialpack Resolve C18 column eluted with methanol-water (7.5:92.5, v/v) containing 100 mM triethylamine. 6-TG, 6-MP and the hydrolysis product of 6-MMP, 4-amino-5-(methylthio)carbonyl imidazole, were monitored at 342, 322 and 303 nm using a Shimadzu SPD-M10A diode array UV detector. The analytes eluted at 5.3, 6.0 and 10.2 min, respectively. The calibration curves were linear (r² > 0.998), and the analytical recoveries were 73.2% for 6-TG, 119.1% for 6-MP and 97.4% for 6-MMP. The intra- and inter-assay variations were highest for 6-MP (9.6 and 14.3%, respectively). The lowest detectable concentrations were 3, 3 and 25 pmol/8 x 10(8) erythrocytes for 6-TG, 6-MP and 6-MMP, respectively. The quantification limits (coefficients of variation <15%) were 8, 10 and 70 pmol/8 x 10(8) erythrocytes for 6-TG, 6-MP and 6-MMP, respectively. The method was applied to the analysis of 183 samples from 36 children under chemotherapy for acute lymphoblastic leukemia. The concentrations of the metabolites in the red cells of the patients ranged from 0 to 1934 pmol/8 x 10(8) erythrocytes for 6-TGN, and from 0 to 105.8 and 0 to 45.9 nmol/8 x 10(8) erythrocytes for 6-MP and 6-MMP, respectively. The procedure gave results that were in agreement with those obtained with other methods designed to detect cases of non-compliance with treatment, including patient interviews and medical evaluation, among others, demonstrating its applicability to monitoring the treatment of leukemic children.

Modeling a Network of Heat Exchangers

This research work addresses the problem of building a mathematical model for the given system of heat exchangers and to determine the temperatures, pressures and velocities at the intermediate positions. Such model could be used in nding an optimal design for such a superstructure. To limit the size and computing time a reduced network model was used. The method can be generalized to larger network structures. A mathematical model which includes a system of non-linear equations has been built and solved according to the Newton-Raphson algorithm. The results obtained by the proposed mathematical model were compared with the results obtained by the Paterson approximation and Chen's Approximation. Results of this research work in collaboration with a current ongoing research at the department will optimize the valve positions and hence, minimize the pumping cost and maximize the heat transfer of the system of heat exchangers.

An HPLC-UV method for the measurement of permeability of marker drugs in the Caco-2 cell assay

The Caco-2 cell line has been used as a model to predict the in vitro permeability of the human intestinal barrier. The predictive potential of the assay relies on an appropriate in-house validation of the method. The objective of the present study was to develop a single HPLC-UV method for the identification and quantitation of marker drugs and to determine the suitability of the Caco-2 cell permeability assay. A simple chromatographic method was developed for the simultaneous determination of both passively (propranolol, carbamazepine, acyclovir, and hydrochlorothiazide) and actively transported drugs (vinblastine and verapamil). Separation was achieved on a C18 column with step-gradient elution (acetonitrile and aqueous solution of ammonium acetate, pH 3.0) at a flow rate of 1.0 mL/min and UV detection at 275 nm during the total run time of 35 min. The method was validated and found to be specific, linear, precise, and accurate. This chromatographic system can be readily used on a routine basis and its utilization can be extended to other permeability models. The results obtained in the Caco-2 bi-directional transport experiments confirmed the validity of the assay, given that high and low permeability profiles were identified, and P-glycoprotein functionality was established.

Body composition measures of obese adolescents by the deuterium oxide dilution method and by bioelectrical impedance

The objectives of the present study were to describe and compare the body composition variables determined by bioelectrical impedance (BIA) and the deuterium dilution method (DDM), to identify possible correlations and agreement between the two methods, and to construct a linear regression model including anthropometric measures. Obese adolescents were evaluated by anthropometric measures, and body composition was assessed by BIA and DDM. Forty obese adolescents were included in the study. Comparison of the mean values for the following variables: fat body mass (FM; kg), fat-free mass (FFM; kg), and total body water (TBW; %) determined by DDM and by BIA revealed significant differences. BIA overestimated FFM and TBW and underestimated FM. When compared with data provided by DDM, the BIA data presented a significant correlation with FFM (r = 0.89; P < 0.001), FM (r = 0.93; P < 0.001) and TBW (r = 0.62; P < 0.001). The Bland-Altman plot showed no agreement for FFM, FM or TBW between data provided by BIA and DDM. The linear regression models proposed in our study with respect to FFM, FM, and TBW were well adjusted. FFM obtained by DDM = 0.842 x FFM obtained by BIA. FM obtained by DDM = 0.855 x FM obtained by BIA + 0.152 x weight (kg). TBW obtained by DDM = 0.813 x TBW obtained by BIA. The body composition results of obese adolescents determined by DDM can be predicted by using the measures provided by BIA through a regression equation.

In-house method validation and occurrence of alpha-, beta-endosulfan, endosulfan sulphate, lambda-cyhalothrin, procymidone and trifluralin residues in strawberry

A method for determination of organohalogen pesticides in strawberry by gas chromatography with electron capture detection was validated and applied in a monitoring program. Linearity, matrix effects, and day effect were evaluated for the analytes alpha-endosulfan, beta-endosulfan, endosulfan sulphate, lambda-cyhalothrin, procymidone, and trifluralin. The linear range varied according to the chromatographic response of the analyte. Significant matrix effects were observed. The mean recoveries ranged from 74.6 to 115.4%, with repeatability standard deviations between 1.6 and 21.0% and intermediate precision between 5.9 and 21.0%. Detection, quantification and decision limit, and detection capacity ranged from 0.003 to 0.007 mg/kg, 0.005 to 0.013 mg/kg; 0.003 to 3.128 mg/kg; and 0.005 to 3.266 mg/kg, respectively. The method was fit for the purpose of monitoring organohalogen residues in strawberries. Residues of these pesticides were detected in 124 of the 186 samples analyzed between 2009 and 2011 in the state of Minas Gerais. Nine of them did not comply with the current legislation requirements; among them, seven (3.8%) had residues of unauthorized pesticide for the culture of strawberry, one (0.5%) had residues above the maximum residue limit, and another one (0.5%) exhibited both non-conformities.

A modified nominal stress method for fatigue assessment of steel plates with thermally cut edges

Thermal cutting methods, are commonly used in the manufacture of metal parts. Thermal cutting processes separate materials by using heat. The process can be done with or without a stream of cutting oxygen. Common processes are Oxygen, plasma and laser cutting. It depends on the application and material which cutting method is used. Numerically-controlled thermal cutting is a cost-effective way of prefabricating components. One design aim is to minimize the number of work steps in order to increase competitiveness. This has resulted in the holes and openings in plate parts manufactured today being made using thermal cutting methods. This is a problem from the fatigue life perspective because there is local detail in the as-welded state that causes a rise in stress in a local area of the plate. In a case where the static utilization of a net section is full used, the calculated linear local stresses and stress ranges are often over 2 times the material yield strength. The shakedown criteria are exceeded. Fatigue life assessment of flame-cut details is commonly based on the nominal stress method. For welded details, design standards and instructions provide more accurate and flexible methods, e.g. a hot-spot method, but these methods are not universally applied to flame cut edges. Some of the fatigue tests of flame cut edges in the laboratory indicated that fatigue life estimations based on the standard nominal stress method can give quite a conservative fatigue life estimate in cases where a high notch factor was present. This is an undesirable phenomenon and it limits the potential for minimizing structure size and total costs. A new calculation method is introduced to improve the accuracy of the theoretical fatigue life prediction method of a flame cut edge with a high stress concentration factor. Simple equations were derived by using laboratory fatigue test results, which are published in this work. The proposed method is called the modified FAT method (FATmod). The method takes into account the residual stress state, surface quality, material strength class and true stress ratio in the critical place.

Solving large sparse linear systems over the field GF(2)

Nowadays problem of solving sparse linear systems over the field GF(2) remain as a challenge. The popular approach is to improve existing methods such as the block Lanczos method (the Montgomery method) and the Wiedemann-Coppersmith method. Both these methods are considered in the thesis in details: there are their modifications and computational estimation for each process. It demonstrates the most complicated parts of these methods and gives the idea how to improve computations in software point of view. The research provides the implementation of accelerated binary matrix operations computer library which helps to make the progress steps in the Montgomery and in the Wiedemann-Coppersmith methods faster.

Investigation of the composition of linear alkylbenzenes with emphasis on the identification and quantitation of some trace compounds using GS/MS system in both electron impact and chemical ionization modes

Linear alkylbenzenes, LAB, formed by the Alel3 or HF catalyzed alkylation of benzene are common raw materials for surfactant manufacture. Normally they are sulphonated using S03 or oleum to give the corresponding linear alkylbenzene sulphonates In >95 % yield. As concern has grown about the environmental impact of surfactants,' questions have been raised about the trace levels of unreacted raw materials, linear alkylbenzenes and minor impurities present in them. With the advent of modem analytical instruments and techniques, namely GCIMS, the opportunity has arisen to identify the exact nature of these impurities and to determine the actual levels of them present in the commercial linear ,alkylbenzenes. The object of the proposed study was to separate, identify and quantify major and minor components (1-10%) in commercial linear alkylbenzenes. The focus of this study was on the structure elucidation and determination of impurities and on the qualitative determination of them in all analyzed linear alkylbenzene samples. A gas chromatography/mass spectrometry, (GCIMS) study was performed o~ five samples from the same manufacturer (different production dates) and then it was followed by the analyses of ten commercial linear alkylbenzenes from four different suppliers. All the major components, namely linear alkylbenzene isomers, followed the same elution pattern with the 2-phenyl isomer eluting last. The individual isomers were identified by interpretation of their electron impact and chemical ionization mass spectra. The percent isomer distribution was found to be different from sample to sample. Average molecular weights were calculated using two methods, GC and GCIMS, and compared with the results reported on the Certificate of Analyses (C.O.A.) provided by the manufacturers of commercial linear alkylbenzenes. The GC results in most cases agreed with the reported values, whereas GC/MS results were significantly lower, between 0.41 and 3.29 amu. The minor components, impurities such as branched alkylbenzenes and dialkyltetralins eluted according to their molecular weights. Their fragmentation patterns were studied using electron impact ionization mode and their molecular weight ions confirmed by a 'soft ionization technique', chemical ionization. The level of impurities present i~ the analyzed commercial linear alkylbenzenes was expressed as the percent of the total sample weight, as well as, in mg/g. The percent of impurities was observed to vary between 4.5 % and 16.8 % with the highest being in sample "I". Quantitation (mg/g) of impurities such as branched alkylbenzenes and dialkyltetralins was done using cis/trans-l,4,6,7-tetramethyltetralin as an internal standard. Samples were analyzed using .GC/MS system operating under full scan and single ion monitoring data acquisition modes. The latter data acquisition mode, which offers higher sensitivity, was used to analyze all samples under investigation for presence of linear dialkyltetralins. Dialkyltetralins were reported quantitatively, whereas branched alkylbenzenes were reported semi-qualitatively. The GC/MS method that was developed during the course of this study allowed identification of some other trace impurities present in commercial LABs. Compounds such as non-linear dialkyltetralins, dialkylindanes, diphenylalkanes and alkylnaphthalenes were identified but their detailed structure elucidation and the quantitation was beyond the scope of this study. However, further investigation of these compounds will be the subject of a future study.

On the Ewald method and the quartic Helmholtz free energy of an anharmonic metallic crystal

Thesis (M. Sc.) - Brock University, 1978.

The application of the structural correlation method to determine the reaction coordinate for a putative ring closure reaction

The cr ystal structure of the compound 2-benzoylethylidene-3-(2,4- dibromophenyl)-2,3-dihydro-5-phenyl-l,3,4-thiadiazole* C23H16Br2NZOS (BRMEO) has been determined by using three dimensiona l x-ray diffraction data. The crys tal form is monoclinic, space group P21/c, a = 17.492(4), o -.t' 0 R 0 b =: 16.979(1), c = 14.962(1) A, "X. =o= 90 ',= 106.46(1) , z = 8, graphite-monochromatized Mo~ rad iation, Jl= 0.710J3~, D = 1.62g/cc and o D = 1.65g/cc. The data were col lected on ~ Nonius CAD-4 c diffractometer. The following atoms were made anisotropic: Br, S, N, 0, C7, and C14-C16 for each i ndependent molecu le ; the rest were left isotropic. For 3112 independent refl ec tions with F > 6G\F), R == 0.057. The compound has two independent molecules within the asymmetric unit. Two different conformers were observed which pack well together. /l The S---O interaction distances of 2.493(6) and 2 . 478(7) A were observed for molecules A and B respectively. These values are consistent with earlier findings for 2-benzoylmethylene-3-(2,4-dibromophenyl)- ~~ 2,3-dihydro-5-phenyl-l,3,4-thiadiazole C22H14Br2N20S (BRPHO) and 2-benzoylpropylidene-3-(2,4-dibromophenyl)-2,3-dihydroiii ,'r 5-phenyl-l,3,4-thiadiazole C24H18Br2N20S (BRPETO ) where S---O distances are l ess than the van der Waals (3.251\) but greater than those expected for () a single bond (1.50A). From the results and the literature it appears obvious that the energy/reaction coordinate pathway has a minimum between the end structures (the mono- and bicyclic compounds). * See reference (21) for nomenclature.

Pigment orientation changes detected by low temperature linear dichroism spectroscopy: cold-hardening transition in phycobilisome-containing organisms

Low temperature (77K) linear dichroism spectroscopy was used to characterize pigment orientation changes accompanying the light state transition in the cyanobacterium, Synechococcus sp. pee 6301, and cold-hardening in winter rye (Secale cereale L. cv. Puma). Samples were oriented for spectroscopy using the gel squeezing method (Abdourakhmanov et aI., 1979) and brought to 77K in liquid nitrogen. The linear dichroism (LD) spectra of Synechococcus 6301 phycobilisome/thylakoid membrane fragments cross-linked in light state 1 and light state 2 with glutaraldehyde showed differences in both chlorophyll a and phycobilin orientation. A decrease in the relative amplitude of the 681nm chlorophyll a positive LD peak was observed in membrane fragments in state 2. Reorientation of the phycobilisome (PBS) during the transition to state 2 resulted in an increase in core allophycocyanin absorption parallel to the membrane, and a decrease in rod phycocyanin parallel absorption. This result supports the "spillover" and "PBS detachment" models of the light state transition in PBS-containing organisms, but not the "mobile PBS" model. A model was proposed for PBS reorientation upon transition to state 2, consisting of a tilt in the antenna complex with respect to the membrane plane. Linear dichroism spectra of PBS/thylakoid fragments from the red alga, Porphyridium cruentum, grown in green light (containing relatively more PSI) and red light (containing relatively more PSll) were compared to identify chlorophyll a absorption bands associated with each photosystem. Spectra from red light - grown samples had a larger positive LD signal on the short wavelength side of the 686nm chlorophyll a peak than those from green light - grown fragments. These results support the identification of the difference in linear dichroism seen at 681nm in Synechococcus spectra as a reorientation of PSll chromophores. Linear dichroism spectra were taken of thylakoid membranes isolated from winter rye grown at 20°C (non-hardened) and 5°C (cold-hardened). Differences were seen in the orientation of chlorophyll b relative to chlorophyll a. An increase in parallel absorption was identified at the long-wavelength chlorophyll a absorption peak, along with a decrease in parallel absorption from chlorophyll b chromophores. The same changes in relative pigment orientation were seen in the LD of isolated hardened and non-hardened light-harvesting antenna complexes (LHCII). It was concluded that orientational differences in LHCII pigments were responsible for thylakoid LD differences. Changes in pigment orientation, along with differences observed in long-wavelength absorption and in the overall magnitude of LD in hardened and non-hardened complexes, could be explained by the higher LHCII monomer:oligomer ratio in hardened rye (Huner et ai., 1987) if differences in this ratio affect differential light scattering properties, or fluctuation of chromophore orientation in the isolated LHCII sample.

Cognitive Behavior Therapy for Obsessive-Compulsive Behavior in Children with Autism Spectrum Disorder: A Comparison of Hierarchical Linear Modelling and Visual Analysis

Behavioral researchers commonly use single subject designs to evaluate the effects of a given treatment. Several different methods of data analysis are used, each with their own set of methodological strengths and limitations. Visual inspection is commonly used as a method of analyzing data which assesses the variability, level, and trend both within and between conditions (Cooper, Heron, & Heward, 2007). In an attempt to quantify treatment outcomes, researchers developed two methods for analysing data called Percentage of Non-overlapping Data Points (PND) and Percentage of Data Points Exceeding the Median (PEM). The purpose of the present study is to compare and contrast the use of Hierarchical Linear Modelling (HLM), PND and PEM in single subject research. The present study used 39 behaviours, across 17 participants to compare treatment outcomes of a group cognitive behavioural therapy program, using PND, PEM, and HLM on three response classes of Obsessive Compulsive Behaviour in children with Autism Spectrum Disorder. Findings suggest that PEM and HLM complement each other and both add invaluable information to the overall treatment results. Future research should consider using both PEM and HLM when analysing single subject designs, specifically grouped data with variability.

Finite-Sample Diagnostics for Multivariate Regressions with Applications to Linear Asset Pricing Models

In this paper, we propose several finite-sample specification tests for multivariate linear regressions (MLR) with applications to asset pricing models. We focus on departures from the assumption of i.i.d. errors assumption, at univariate and multivariate levels, with Gaussian and non-Gaussian (including Student t) errors. The univariate tests studied extend existing exact procedures by allowing for unspecified parameters in the error distributions (e.g., the degrees of freedom in the case of the Student t distribution). The multivariate tests are based on properly standardized multivariate residuals to ensure invariance to MLR coefficients and error covariances. We consider tests for serial correlation, tests for multivariate GARCH and sign-type tests against general dependencies and asymmetries. The procedures proposed provide exact versions of those applied in Shanken (1990) which consist in combining univariate specification tests. Specifically, we combine tests across equations using the MC test procedure to avoid Bonferroni-type bounds. Since non-Gaussian based tests are not pivotal, we apply the “maximized MC” (MMC) test method [Dufour (2002)], where the MC p-value for the tested hypothesis (which depends on nuisance parameters) is maximized (with respect to these nuisance parameters) to control the test’s significance level. The tests proposed are applied to an asset pricing model with observable risk-free rates, using monthly returns on New York Stock Exchange (NYSE) portfolios over five-year subperiods from 1926-1995. Our empirical results reveal the following. Whereas univariate exact tests indicate significant serial correlation, asymmetries and GARCH in some equations, such effects are much less prevalent once error cross-equation covariances are accounted for. In addition, significant departures from the i.i.d. hypothesis are less evident once we allow for non-Gaussian errors.