Automatic learning for the classification of chemical reactions and in statistical thermodynamics

This Thesis describes the application of automatic learning methods for a) the classification of organic and metabolic reactions, and b) the mapping of Potential Energy Surfaces(PES). The classification of reactions was approached with two distinct methodologies: a representation of chemical reactions based on NMR data, and a representation of chemical reactions from the reaction equation based on the physico-chemical and topological features of chemical bonds. NMR-based classification of photochemical and enzymatic reactions. Photochemical and metabolic reactions were classified by Kohonen Self-Organizing Maps (Kohonen SOMs) and Random Forests (RFs) taking as input the difference between the 1H NMR spectra of the products and the reactants. The development of such a representation can be applied in automatic analysis of changes in the 1H NMR spectrum of a mixture and their interpretation in terms of the chemical reactions taking place. Examples of possible applications are the monitoring of reaction processes, evaluation of the stability of chemicals, or even the interpretation of metabonomic data. A Kohonen SOM trained with a data set of metabolic reactions catalysed by transferases was able to correctly classify 75% of an independent test set in terms of the EC number subclass. Random Forests improved the correct predictions to 79%. With photochemical reactions classified into 7 groups, an independent test set was classified with 86-93% accuracy. The data set of photochemical reactions was also used to simulate mixtures with two reactions occurring simultaneously. Kohonen SOMs and Feed-Forward Neural Networks (FFNNs) were trained to classify the reactions occurring in a mixture based on the 1H NMR spectra of the products and reactants. Kohonen SOMs allowed the correct assignment of 53-63% of the mixtures (in a test set). Counter-Propagation Neural Networks (CPNNs) gave origin to similar results. The use of supervised learning techniques allowed an improvement in the results. They were improved to 77% of correct assignments when an ensemble of ten FFNNs were used and to 80% when Random Forests were used. This study was performed with NMR data simulated from the molecular structure by the SPINUS program. In the design of one test set, simulated data was combined with experimental data. The results support the proposal of linking databases of chemical reactions to experimental or simulated NMR data for automatic classification of reactions and mixtures of reactions. Genome-scale classification of enzymatic reactions from their reaction equation. The MOLMAP descriptor relies on a Kohonen SOM that defines types of bonds on the basis of their physico-chemical and topological properties. The MOLMAP descriptor of a molecule represents the types of bonds available in that molecule. The MOLMAP descriptor of a reaction is defined as the difference between the MOLMAPs of the products and the reactants, and numerically encodes the pattern of bonds that are broken, changed, and made during a chemical reaction. The automatic perception of chemical similarities between metabolic reactions is required for a variety of applications ranging from the computer validation of classification systems, genome-scale reconstruction (or comparison) of metabolic pathways, to the classification of enzymatic mechanisms. Catalytic functions of proteins are generally described by the EC numbers that are simultaneously employed as identifiers of reactions, enzymes, and enzyme genes, thus linking metabolic and genomic information. Different methods should be available to automatically compare metabolic reactions and for the automatic assignment of EC numbers to reactions still not officially classified. In this study, the genome-scale data set of enzymatic reactions available in the KEGG database was encoded by the MOLMAP descriptors, and was submitted to Kohonen SOMs to compare the resulting map with the official EC number classification, to explore the possibility of predicting EC numbers from the reaction equation, and to assess the internal consistency of the EC classification at the class level. A general agreement with the EC classification was observed, i.e. a relationship between the similarity of MOLMAPs and the similarity of EC numbers. At the same time, MOLMAPs were able to discriminate between EC sub-subclasses. EC numbers could be assigned at the class, subclass, and sub-subclass levels with accuracies up to 92%, 80%, and 70% for independent test sets. The correspondence between chemical similarity of metabolic reactions and their MOLMAP descriptors was applied to the identification of a number of reactions mapped into the same neuron but belonging to different EC classes, which demonstrated the ability of the MOLMAP/SOM approach to verify the internal consistency of classifications in databases of metabolic reactions. RFs were also used to assign the four levels of the EC hierarchy from the reaction equation. EC numbers were correctly assigned in 95%, 90%, 85% and 86% of the cases (for independent test sets) at the class, subclass, sub-subclass and full EC number level,respectively. Experiments for the classification of reactions from the main reactants and products were performed with RFs - EC numbers were assigned at the class, subclass and sub-subclass level with accuracies of 78%, 74% and 63%, respectively. In the course of the experiments with metabolic reactions we suggested that the MOLMAP / SOM concept could be extended to the representation of other levels of metabolic information such as metabolic pathways. Following the MOLMAP idea, the pattern of neurons activated by the reactions of a metabolic pathway is a representation of the reactions involved in that pathway - a descriptor of the metabolic pathway. This reasoning enabled the comparison of different pathways, the automatic classification of pathways, and a classification of organisms based on their biochemical machinery. The three levels of classification (from bonds to metabolic pathways) allowed to map and perceive chemical similarities between metabolic pathways even for pathways of different types of metabolism and pathways that do not share similarities in terms of EC numbers. Mapping of PES by neural networks (NNs). In a first series of experiments, ensembles of Feed-Forward NNs (EnsFFNNs) and Associative Neural Networks (ASNNs) were trained to reproduce PES represented by the Lennard-Jones (LJ) analytical potential function. The accuracy of the method was assessed by comparing the results of molecular dynamics simulations (thermal, structural, and dynamic properties) obtained from the NNs-PES and from the LJ function. The results indicated that for LJ-type potentials, NNs can be trained to generate accurate PES to be used in molecular simulations. EnsFFNNs and ASNNs gave better results than single FFNNs. A remarkable ability of the NNs models to interpolate between distant curves and accurately reproduce potentials to be used in molecular simulations is shown. The purpose of the first study was to systematically analyse the accuracy of different NNs. Our main motivation, however, is reflected in the next study: the mapping of multidimensional PES by NNs to simulate, by Molecular Dynamics or Monte Carlo, the adsorption and self-assembly of solvated organic molecules on noble-metal electrodes. Indeed, for such complex and heterogeneous systems the development of suitable analytical functions that fit quantum mechanical interaction energies is a non-trivial or even impossible task. The data consisted of energy values, from Density Functional Theory (DFT) calculations, at different distances, for several molecular orientations and three electrode adsorption sites. The results indicate that NNs require a data set large enough to cover well the diversity of possible interaction sites, distances, and orientations. NNs trained with such data sets can perform equally well or even better than analytical functions. Therefore, they can be used in molecular simulations, particularly for the ethanol/Au (111) interface which is the case studied in the present Thesis. Once properly trained, the networks are able to produce, as output, any required number of energy points for accurate interpolations.

Development of a global IT charging model for BASF Group

With ‘GS Strategy 2025’ BASF Business Services GmbH was formed to centrally steer all IT related topics of BASF group. Thus, a global charging system has to be designed, which complies to international transfer price regulations and the strategy of BASF SE. This work project develops a charging system with a following evaluation. The direct charging system benefits from its cost transparency upsides but comes with a higher administrative effort due to volume-based charging. In contrast, the indirect charging system convinces because of easy handling, which is the result of the application of suitable allocation keys. Regarding the complex group structure of BASF SE with more than 300 legal entities in 80 countries, the lower administrative effort of the indirect charging system outweighs the benefits of the direct charging model and should be used by BASF group.

Economias de escala em centros hospitalares

RESUMO - Perante o actual contexto de contenção de gastos no sector da saúde e consequente preocupação com a eficiência do sistema, tem‐se assistido a mudanças várias no modelo de gestão e organizacional do sistema de saúde. Destaca‐se a alteração da estrutura hospitalar, com vista à racionalização dos seus recursos internos, onde as fusões hospitalares têm assumido um papel determinante. Em Portugal, nos últimos 10 anos, assistiu‐se a uma significativa redução do número de hospitais (de sensivelmente 90 para 50 unidades), exclusivamente através das fusões e sem quaisquer alterações no número de estruturas físicas existentes. Não obstante os argumentos justificativos desta reforma, a avaliação dos objectivos implícitos é insuficiente. Neste âmbito, pretendeu‐se com este estudo contribuir para a análise do impacte da criação de centros hospitalares na redução de gastos, isto é, verificar se a consolidação e consequente reengenharia dos processos produtivos teve consequencias ao nível da obtenção de economias de escala. Para esta análise usou‐se uma base de dados em painel, onde se consideraram 75 hospitais durante 7 anos (2003‐2009), número que foi reduzindo ao longo do período em análise devido às inúmeras fusões já referidas. Para avaliar os ganhos relativos às fusões hospitalares, ao nível da eficiência técnica e das economias de escala, recorreu‐se à fronteira estocástica especificada função custo translog. Estimada a fronteira, foi possível analisar três centros hospitalares específicos, onde se comparou o período pré‐fusão (2005‐2006) com o período após a fusão (2008‐2009). Como variáveis explicativas, relativas à produção hospitalar, considerou‐se o número de casos tratados e os dias de internamento (Vita, 1990; Schuffham et al., 1996), o número de consultas e o número de urgências, sendo estas variáveis as mais comuns na literatura (Vita, 1990; Fournier e Mitchell, 1992; Carreira, 1999). Quanto à variável dependente usou‐se o custo variável total, que compreende o total de custos anuais dos hospitais excepto de imobilizado. Como principais conclusões da investigação, em consequência da criação dos centros hospitalares, são de referir os ganhos de escala na fusão de hospitais de reduzida dimensão e com mais serviços complementares. --------ABSTRACT - Driven by the current pressure on resources induced by budgetary cuts, the Portuguese Ministry of Health is imposing changes in the management model and organization of NHS hospitals. The most recent change is based on the creation of Hospital Centres that are a result of administrative mergers of existing hospitals. In less than 10 years the number of hospitals passed from around 90 to around 50, only due to the mergers and without any change in the existing number of physical institutions. According to the political discourse, one of the main goals expected from this measure is the creation of synergies and more efficiency in the use of available resources. However, the merger of the hospitals has been a political decision without support or evaluation of the first experiments. The aim of this study is to measure the results of this policy by looking at economies of scale namely through reductions in the expenditures, as expected and sought by the MoH. Data used covers 7 years (2003‐2009) and 75 hospitals, number that has been reduced my the enoumerous mergers during the last decade. This work uses a stochastic frontier analysis through the translog cost function to examine the gains from mergers, which were decomposed into technical efficiency and economies of scale. It was analised these effects by the creation of three specific hospital centers, using a longitudinal approach to compare the period pre‐merger (2003‐2006) with the post‐merger period (2007‐09). To measure changes in inpatient hospital production volume and length of stay are going to be considered as done by Vita (1990) and Schuffham et al. (1996). For outpatient services the number of consultations and emergencies are going to be considered (Vita, 1990; Fournier e Mitchell, 1992; Carreira, 1999). Total variable cost is considered as the dependent variable explained the aforementioned ones. After a review of the literature results expected point to benefits from the mergers, namely a reduction in total expenditures and in the number of duplicated services. Results extracted from our data point in the same direction, and thus for the existence of some economies of scale only for small hospitals.

Geospatial technologies for public participation: better decisions for smarter cities?

Dissertation submitted in partial fulfillment of the requirements for the Degree of Master of Science in Geospatial Technologies.

Credit cycle identification: A Markov-switching application

A Work Project, presented as part of the requirements for the Award of a Masters Degree in Finance from the NOVA – School of Business and Economics

Apoio à implementação do Sistema de Informação Territorial da Administração Interna e à análise de padrões espaciais

Este documento visa apresentar o trabalho de Estágio realizado na Direção Geral de Administração Interna (DGAI) como parte integrante e conclusiva do Mestrado em Gestão do Território na Área de Especialização em Deteção Remota e Sistemas de Informação Geográfica, ministrado pela Faculdade de Ciências Sociais e Humanas da Universidade Nova de Lisboa. O Estágio desenvolveu-se em duas fases. A primeira detém um caráter mais genérico e teve como objetivo prestar apoio à consolidação do Sistema de Informação Territorial da Administração Interna (SITAI), onde, entre diversas atividades desenvolvidas, a principal, e conforme exposto no presente relatório, reside na adaptação de dados do Instituto Nacional de Estatística (Censos 2011) ao território nacional após a Reorganização Administrativa do Território das Freguesias (RATF). A segunda enfoca o fenómeno da Violência Doméstica (VD) e consiste no tratamento e análise de vários tipos de dados provenientes da base de dados do Ministério da Administração Interna (MAI) relacionados com o mesmo. Não obstante esta segmentação de trabalhos, é objetivo deste Estágio a aplicação dos Sistemas de Informação Geográfica (SIG) envolvendo dados de realidades específicas e a produção de informação capaz de se constituir como instrumento de suporte à decisão e desenvolvimento de políticas de segurança interna pelas quais o MAI é responsável.

O sigilo bancário e a fraude fiscal

Tax evasion and fraud threaten the economic and social objectives of modern tax systems, precluding the state funding for the satisfaction of collective needs and the fair distribution of wealth, being a violation of basic principles and values of our society. In tax law, to give tax administration the necessary powers to supervise and control the information provided by taxpayers and combat tax evasion and fraud, over the last years the grounds for a derogation of bank secrecy without judicial authorization have been extended, which raises some constitutional compatibility issues. Similarly, this tendency of making this legal regime more flexible and increasing automatic exchange of information has been followed by the European Union and the international community. Banking secrecy, as a professional secrecy, is an instrument to protect the right to privacy but also appears as an anti-abuse and repressive mechanism of evasive and fraudulent behaviors. Because of the conflict of interests will always be necessary to make a practical agreement between them, ensuring the legality and the due guarantees of the taxpayers but also an effective way to combat tax evasion and fraud. Bank secrecy cannot be one method to, behind the right to privacy, taxpayers practice illegal activities. But the practice of these irregular conducts also does not justify a total annihilation of the right to banking secrecy, uncovering all documents and bank information’s. Although considering the legislative changes, the administrative derogation of bank secrecy will always be what the tax administration does of it.

Produção hospitalar : a importância da complexidade e da gravidade

RESUMO - A definição e medição da produção são questões centrais para a administração hospitalar. A produção hospitalar, quando se consideram os casos tratados, baseia-se em dois aspectos: a definição de sistemas de classificação de doentes como metodologia para identificar produtos e a criação de índices de casemix para se compararem esses mesmos produtos. Para a sua definição e implementação podem ser consideradas características relacionadas com a complexidade dos casos (atributo da oferta) ou com a sua gravidade (atributo da procura), ou ainda características mistas. Por sua vez, a análise do perfil e da política de admissões dos hospitais adquire um maior relevo no contexto de novas experiências previstas e em curso no SNS e da renovada necessidade de avaliação e regulação que daí decorrem. Neste estudo pretendeu-se discutir a metodologia para apuramento do índice de casemix dos hospitais, introduzindo- se a gravidade dos casos tratados como atributo relevante para a sua concretização. Assim, foi analisada uma amostra de 950 443 casos presentes na base de dados dos resumos de alta em 2002, tendo- -se dado particular atenção aos 31 hospitais posteriormente constituídos como SA. Foram considerados três índices de casemix: índice de complexidade (a partir do peso relativo dos DRGs), índice de gravidade (a partir da escala de mortalidade esperada do disease staging recalibrada para Portugal) e índice conjunto (média dos dois anteriores). Verificou-se que a análise do índice de complexidade, de gravidade e conjunto dá informações distintas sobre o perfil de admissões dos hospitais considerados. Os índices de complexidade e de gravidade mostram associações distintas às características dos hospitais e dos doentes tratados. Para além disso, existe uma diferença clara entre os casos com tratamento médico e cirúrgico. No entanto, para a globalidade dos hospitais analisados observou-se que os hospitais que tratam os casos mais graves tratam igualmente os mais complexos, tendo-se ainda identificado alguns hospitais em que tal não se verifica e, quando possível, apontado eventuais razões para esse comportamento.