Tratamento estatístico dos parâmetros da qualidade das águas da bacia do alto curso do Rio das Velhas

Water quality was monitored at the upper course of the Rio das Velhas, a major tributary of the São Francisco basin located in the state of Minas Gerais, over an extension of 108 km from its source up to the limits with the Sabara district. Monitoring was done at 37 different sites over a period of 2 years (2003-2004) for 39 parameters. Multivariate statistical techniques were applied to interpret the large water-quality data set and to establish an optimal long-term monitoring network. Cluster analysis separated the sampling sites into groups of similarity, and also indicated the stations investigated for correlation and recommended to be removed from the monitoring network. Principal component analysis identified four components, which are responsible for the data structure explaining 80% of the total variance of the data. The principal parameters are characterized as due to mining activities and domestic sewage. Significant data reduction was achieved.

Productivity, Technical eficiency and internationalisation of the Spanish chemical sector

relationship between productivity and international position of Spanish chemical firms in the period 2005-2011. The goal is to determine whether companies that follow and international strategy, either with exports or by investment in foreign countries obtain greater productivity growth than these that do not operate in global market. For this purpose a panel data set with microdata has been created. A preliminary analysis of the evolution of productivity growth in the sector is carried out. The measurement of Total Factor Productivity is performed. With the estimated TFP we analyze the differentials in productivity growth, comparing the effects of export and investment behavior with non-international firms.

Química Nova Interativa - QNInt - o portal do conhecimento da SBQ: conectando ciência e educação

The Química Nova Interativa (QNInt) portal was launched in 2009 by the Brazilian Chemical Society (SBQ) to offer free quality content for broad audiences. QNInt provides peer-reviewed articles from SBQ journals on science & society, chemical concepts, classroom activities and educational research. With 3,000,000 visits, QNInt also offers a unique library of interactive molecules. In the International Year of Chemistry QNInt served for distributing pH kits and registering data from IUPAC's Global Water Experiment, yielding Brazil the largest share of the global pH data set. The portal performance makes QNInt a valuable resource for connecting science to education.

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.

Social capital and health : variations, associations and challenges

Although social capital and health have been extensively studied during the last decade, there are still open issues in current empirical research. These concern for instance the measurement of the concept in different contexts, as well as the association between different types of social capital and different dimensions of health. The present thesis addressed these questions. The general aim was to promote the understanding of social capital and health by investigating the oldest old and the two major language groups in Finland, Swedish- and Finnish-speakers. Another aim was to contribute to the discussion on methodological issues in social capital and health research. The present thesis investigated two empirical data sets, Umeå 85+ and Health 2000. The Umeå 85+ study was a cross-sectional study of 163 individuals aged 85, 90, and 95 or older, living in the municipality of Umeå, Sweden, in the year of 2000. The Health 2000 survey was a national study of 8,028 persons aged 30 or above carried out in Finland in 2000-2001. Different indicators of structural (e.g. social contacts) and cognitive (e.g. trust) social capital, as well as health indicators were used as variables in the analyses. The Umeå 85+ data set was analyzed with factor analysis, as well as univariate and multivariate analysis of variance. The Health 2000 data was analyzed with logistic regression techniques. The results showed that the Swedish-speakers in the Finnish data set Health 2000 had consistently higher prevalence of social capital compared to the Finnish-speakers even after controlling for central sociodemographic variables. The results further showed that even if the language group differences in health were small, the Swedishspeakers experienced in general better self-reported health compared with the Finnish-speakers. Common sociodemographic variables could not explain these observed differences in health. The results imply that social capital is often, but not always, associated with health. This was clearly seen in the Umeå 85+ data set where only one health indicator (depressive symptoms) was associated with structural social capital among the oldest old. The results based on the analysis of the Health 2000 survey demonstrated that the cognitive component of social capital was associated with self-rated health and psychological health rather than with participation in social activities and social contacts. In addition, social capital statistically reduced the health advantage especially for Swedish-speaking men, indicating that high prevalence of social capital may promote health. Finally, the present thesis also discussed the issue of methodological challenges faced with when analyzing social capital and health. It was suggested that certain components of social capital such as bonding and bridging social capital may be more relevant than structural and cognitive components when investigating social capital among the two language groups in Finland. The results concerning the oldest old indicated that the structural aspects of social capital probably reflect current living conditions, whereas cognitive social capital reflects attitudes and traits often acquired decades earlier. This is interpreted as an indication of the fact that structural and cognitive social capital are closely related yet empirically two distinctive concepts. Taken together, some components of social capital may be more relevant to study than others depending on which population group and age group is under study. The results also implied that the choice of cut-off point of dichotomization of selfrated health has an impact on the estimated effects of the explanatory variables. When the whole age interval, 35-64 years, was analyzed with logistic regression techniques the choice of cut-off point did not matter for the estimated effects of marital status and educational level. The results changed, however, when the age interval was divided into three shorter intervals. If self-rated health is explored using wide age intervals that do not account for age-dependent covariates there is a risk of drawing misleading conclusions. In conclusion, the results presented in the thesis suggest that the uneven distribution of social capital observed between the two language groups in Finland are of importance when trying to further understand health inequalities that exist between Swedish- and Finnish-speakers in Finland. Although social capital seemed to be relevant to the understanding of health among the oldest old, the meaning of social capital is probably different compared to a less vulnerable age group. This should be noticed in future empirical research. In the present thesis, it was shown that the relationship between social capital and health is complex and multidimensional. Different aspects of social capital seem to be important for different aspects of health. This reduces the possibility to generalize the results and to recommend general policy implementations in this area. An increased methodological awareness regarding social capital as well as health are called for in order to further understand the cfomplex association between them. However, based on the present data and findings social capital is associated with health. To understand individual health one must also consider social aspects of the individuals’ environment such as social capital.

Modelling head loss along emitting pipes using dimensional analysis

Local head losses must be considered in estimating properly the maximum length of drip irrigation laterals. The aim of this work was to develop a model based on dimensional analysis for calculating head loss along laterals accounting for in-line drippers. Several measurements were performed with 12 models of emitters to obtain the experimental data required for developing and assessing the model. Based on the Camargo & Sentelhas coefficient, the model presented an excellent result in terms of precision and accuracy on estimating head loss. The deviation between estimated and observed values of head loss increased according to the head loss and the maximum deviation reached 0.17 m. The maximum relative error was 33.75% and only 15% of the data set presented relative errors higher than 20%. Neglecting local head losses incurred a higher than estimated maximum lateral length of 19.48% for pressure-compensating drippers and 16.48% for non pressure-compensating drippers.

Bankruptcy prediction using fuzzy methods

In this thesis, a classi cation problem in predicting credit worthiness of a customer is tackled. This is done by proposing a reliable classi cation procedure on a given data set. The aim of this thesis is to design a model that gives the best classi cation accuracy to e ectively predict bankruptcy. FRPCA techniques proposed by Yang and Wang have been preferred since they are tolerant to certain type of noise in the data. These include FRPCA1, FRPCA2 and FRPCA3 from which the best method is chosen. Two di erent approaches are used at the classi cation stage: Similarity classi er and FKNN classi er. Algorithms are tested with Australian credit card screening data set. Results obtained indicate a mean classi cation accuracy of 83.22% using FRPCA1 with similarity classi- er. The FKNN approach yields a mean classi cation accuracy of 85.93% when used with FRPCA2, making it a better method for the suitable choices of the number of nearest neighbors and fuzziness parameters. Details on the calibration of the fuzziness parameter and other parameters associated with the similarity classi er are discussed.

Feature selection using Fuzzy Entropy measures with Yu's Similarity measure

In this study, feature selection in classification based problems is highlighted. The role of feature selection methods is to select important features by discarding redundant and irrelevant features in the data set, we investigated this case by using fuzzy entropy measures. We developed fuzzy entropy based feature selection method using Yu's similarity and test this using similarity classifier. As the similarity classifier we used Yu's similarity, we tested our similarity on the real world data set which is dermatological data set. By performing feature selection based on fuzzy entropy measures before classification on our data set the empirical results were very promising, the highest classification accuracy of 98.83% was achieved when testing our similarity measure to the data set. The achieved results were then compared with some other results previously obtained using different similarity classifiers, the obtained results show better accuracy than the one achieved before. The used methods helped to reduce the dimensionality of the used data set, to speed up the computation time of a learning algorithm and therefore have simplified the classification task

Kansainvälisen osakeportfolion hajauttamishyödyt eri suhdannesykleissä

Tämän tutkimuksen tavoitteena on selvittää onko kansainvälisen hajauttamisen hyöty heikentynyt globaalisti ajan kuluessa. Tutkimusongelmaan pyritään löytämään vastaus korrelaatioanalyysillä sekä Box M-testillä. Teoreettisena viitekehyksenä käytetään Markowitzin luomaa modernia portfolioteoriaa, kansainväliseen hajautukseen liittyvää kirjallisuutta sekä aiheesta aiemmin tehtyjä tutkimuksia. Empiirisenä tutkimusaineistona käytetään Thomson Datastreamin tuottamia kokonaistuottoindeksejä. Indeksit ovat yhdeksältä eri markkina-alueelta ja 30 eri maasta. Maat on jaoteltu 18 kehittyneeseen ja 12 kehittyvään maahan.kaikki tutkielmassa käytetyt tuottoaikasarjat ovat dollarimääräisiä. Tutkimusaineisto kattaa vuodet 1995-2009 sisältäen 783 viikottaista havaintoa, kullekin tuottoindeksille. Tutkimustulosten mukaan kansainvälisen hajauttamisen edut ovat suuremmat sijoitettaessa kehittyville markkinoille, kuin pelkästään kehittyneille markkinoille sijoitettaessa. Tutkimusperiodin alkupuolella kehittyvillä markkinoilla on ollut saatavissa huomattavasti enemmän hajautushyötyä, mutta suhteellinenhyöty suhteessa kehittyneisiin markkinoihin vähenee tultaessa lähemmäs nykyhetkeä. Korrelaatiot ovat nousseet koko ajanjaksolla, mutta on myös ollut osaperiodeja, jolloin korrelaatiokertoimet ovat laskeneet.

The Impact of European Macroeconomic News Announcements on CIVETS Stock Markets

The purpose of this research is to investigate how CIVETS (Colombia, Indonesia, Vietnam, Egypt, Turkey and South Africa) stock markets are integrated with Europe as measured by the impact of euro area (EA) scheduled macroeconomic news announcements, which are related to macroeconomic indicators that are commonly used to indicate the direction of the economy. Macroeconomic announcements used in this study can be divided into four categories; (1) prices, (2) real economy, (3) money supply and (4) business climate and consumer confidence. The data set consists of daily market data from CIVETS and scheduled macroeconomic announcements from the EA for the years 2007-2012. The econometric model used in this research is Exponential Generalized Autoregressive Conditional Heteroscedasticity (EGARCH). Empirical results show diverse impacts of macroeconomic news releases and surprises for different categories of news supporting the perception of heterogeneity among CIVETS. The analyses revealed that in general EA macroeconomic news releases and surprises affect stock market volatility in CIVETS and only in some cases asset pricing. In conclusion, all CIVETS stock markets reacted to the incoming EA macroeconomic news suggesting market integration to some extent. Thus, EA should be considered as a possible risk factor when investing in CIVETS.

Spatial multicriteria decision analysis for the siting of on-shore wind power in Kemiönsaari

Wind power is a low-carbon energy production form that reduces the dependence of society on fossil fuels. Finland has adopted wind energy production into its climate change mitigation policy, and that has lead to changes in legislation, guidelines, regional wind power areas allocation and establishing a feed-in tariff. Wind power production has indeed boosted in Finland after two decades of relatively slow growth, for instance from 2010 to 2011 wind energy production increased with 64 %, but there is still a long way to the national goal of 6 TWh by 2020. This thesis introduces a GIS-based decision-support methodology for the preliminary identification of suitable areas for wind energy production including estimation of their level of risk. The goal of this study was to define the least risky places for wind energy development within Kemiönsaari municipality in Southwest Finland. Spatial multicriteria decision analysis (SMCDA) has been used for searching suitable wind power areas along with many other location-allocation problems. SMCDA scrutinizes complex ill-structured decision problems in GIS environment using constraints and evaluation criteria, which are aggregated using weighted linear combination (WLC). Weights for the evaluation criteria were acquired using analytic hierarchy process (AHP) with nine expert interviews. Subsequently, feasible alternatives were ranked in order to provide a recommendation and finally, a sensitivity analysis was conducted for the determination of recommendation robustness. The first study aim was to scrutinize the suitability and necessity of existing data for this SMCDA study. Most of the available data sets were of sufficient resolution and quality. Input data necessity was evaluated qualitatively for each data set based on e.g. constraint coverage and attribute weights. Attribute quality was estimated mainly qualitatively by attribute comprehensiveness, operationality, measurability, completeness, decomposability, minimality and redundancy. The most significant quality issue was redundancy as interdependencies are not tolerated by WLC and AHP does not include measures to detect them. The third aim was to define the least risky areas for wind power development within the study area. The two highest ranking areas were Nordanå-Lövböle and Påvalsby followed by Helgeboda, Degerdal, Pungböle, Björkboda, and Östanå-Labböle. The fourth aim was to assess the recommendation reliability, and the top-ranking two areas proved robust whereas the other ones were more sensitive.

Benzodiazepines and Cognitive Functioning in Older Adults. With Emphasis on Long-Term Use and Withdrawal

Benzodiazepines (BZD) and benzodiazepine related drugs (RD) are the most commonly used psychotropics among the aged. The use of other psychotropics taken concomitantly with BZD/ RD or their cognitive effects with BZD/RD have not been studied frequently. The aim of this academic thesis was to describe and analyse relationships between the use of BZD/RD alone or concomitantly with antipsychotics, antidepressants, opioids, antiepileptics, opioids and anticholinergics in the aged and their health. Especially, the relationships between long-term use of BZD/RD and cognitive decline were studied. Additionally, the effect of melatonin on BZD/RD withdrawal and the cognitive effects of BZD/RD withdrawal were studied. This study used multiple data sets: the first study (I) was based on clinical data containing aged patients (≥65 years; N=164) admitted to Pori City Hospital due to acute disease. The second data set (Studies II and III) was based on population-based data from the Lieto Study, a clinico-epidemiological longitudinal study carried out among the aged (≥65 years) in the municipality of Lieto. Follow-up data was formed by combining the cohort data collected in 1990-1991 (N=1283) and in 1998-1999 (N=1596) from those who participated in both cohorts (N=617). The third data set (Studies IV and V) was based on the Satauni Study’s data. This study was performed in the City of Pori in 2009-2010. In the RCT part of the Satauni Study, ninety-two long-term users of BZD/RD were withdrawn from their drugs using melatonin against placebo. The change of their cognitive abilities was measured during and after BZD/ RD withdrawal. BZD/RD use was related to worse cognitive and functional abilities, and their use may predict worse cognitive outcomes compared with BZD/RD non-users. Hypnotic use of BZD/RD could be withdrawn with psychosocial support in motivated participants, but melatonin did not improve the withdrawal results compared to those with placebo. Cognitive abilities in psychomotor tests did not show, or showed only modest, improvements for up to six months after BZD/RD withdrawal. This suggests that the cognitive effects of BZD/RD may be longlasting or permanent.

Johtajan jäljillä – Johtaminen varhaiskasvatuksen hajautetuissa organisaatioissa laadun ja pedagogisen tuen näkökulmasta

Tutkimus sijoittuu varhaiskasvatuksen hajautetun organisaation kontekstiin, mutta tulokset ovat siirrettävissä muihinkin suomalaisiin kasvatus- ja opetustoimen organisaatioihin. Hajautettujen organisaatioiden tutkimus on ollut varhaiskasvatuksen kentällä vielä vähäistä, vaikka organisaatiomallin vaikutukset johtajuuden toteuttamiselle ovat merkittävät. Hajautetulla organisaatiolla varhaiskasvatuksessa tarkoitetaan sitä, että yhden johtajan alaisuudessa on monta eri päiväkotia tai erilaisia päivähoitomuotoja. Tämä organisaatiomalli on yhä enenevässä määrin kasvava suomalaisessa varhaiskasvatuksessa. Varhaiskasvatuksen hajautettujen organisaatioiden tutkimuksessa on aiemmin tarkasteltu johtajan ja työntekijöiden ja työntekijöiden keskinäisiä ammatillisia suhteita. Tässä tutkimuksessa näkökulma painottuu johtamiseen ja työskentelyyn hajautetuissa organisaatiossa sinänsä sekä myös laadunarviointiin sekä pedagogiikkaan. Viitekehyksenä tutkimuksessa on LMX-teoria (leader-member-exchange, johtajuuden vaihtoteoria), jossa tarkastellaan esimies-alaissuhdetta ja siihen kiinteästi liittyvää luottamuksen käsitettä. Luottamuksen merkitys hajautetuissa organisaatioissa korostuu, koska esimies ei ole fyysisesti päivittäin läsnä työntekijöiden arjessa. Tutkimuksessa tarkastellaan hajautetuissa varhaiskasvatuksen organisaatioissa työskentelyä seuraavien tutkimuskysymysten avulla: 1) Mitkä ovat varhaiskasvatuksen hajautettujen organisaatioiden johtamisen erityispiirteet? 2) Miten eri työntekijäryhmät kokevat hajautetussa organisaatiossa työskentelyn? 3) Millaisia kokemuksia esimiehillä ja työntekijöillä on heidän yksiköissään toteutetusta laadunarvioinnista? 4) Millaiseksi työntekijät ja esimiehet kokevat esimieheltään saadun tuen? Tutkimuksessa oli kolme eri aineistoa. Ensimmäinen aineisto koostui 11 hajautetun organisaation johtajan haastattelusta. Toinen aineisto (n = 223) sisälsi haastateltujen esimiesten lomakevastausten lisäksi heidän alaisuudessaan toimivien työntekijöiden, 10 esimieskoulutukseen osallistuneen johtajan sekä kolmen erillisyksikön työntekijöiden vastaukset. Kolmas aineisto oli kerätty pääkaupunkiseudulta varhaiskasvatuksen johtajilta lomakekyselynä (n = 112). Aineistoa on analysoitu teorialähtöisen ja aineistolähtöisen sisällönanalyysin ja tilastollisten analyysien avulla Tulokset osoittavat, että johtajat kokivat hallinnollisten töiden vievän paljon aikaa. Esimiehen kanssa eri työpaikassa työskentelevät työntekijät hahmottivat koko organisaation selkeämmin kuin esimiehen kanssa fyysisesti samassa paikassa työskentelevät. Esimiesten käsitysten mukaan laadunarviointia suoritettiin enemmän kuin mitä työntekijöiden mukaan. Työntekijät kaipasivat esimiehiltään tukea yhteistyöhön ja vuorovaikutukseen, pedagogiseen ohjaukseen, kehittämiseen ja toiminnan resursseihin liittyen. Erillisyksikössä työskentelevät kokivat saavansa enemmän tukea kuin esimiehen kanssa fyysisesti samassa yksikössä työskentelevät työntekijät. Sekä esimieheltä saadun pedagogisen tuen että luottamuksen kokemukset kiinnittävät tämän tutkimuksen tulosten mukaan huomion rakenteiden merkitykseen hajautetuissa organisaatioissa. Arviointiin, pedagogiseen tukeen ja tiedonkulkuun liittyvien rakenteiden huomioiminen helpottaa hajautetussa organisaatiossa johtamista. Edellisten lisäksi johtajan selkeä visio omasta johtamistyöstään ja jaetun johtajuuden hyödyntäminen edesauttavat työn hallinnan kokemuksia.

Differential evolution based classification with pool of distances and aggregation operators

The objective of this thesis is to develop and generalize further the differential evolution based data classification method. For many years, evolutionary algorithms have been successfully applied to many classification tasks. Evolution algorithms are population based, stochastic search algorithms that mimic natural selection and genetics. Differential evolution is an evolutionary algorithm that has gained popularity because of its simplicity and good observed performance. In this thesis a differential evolution classifier with pool of distances is proposed, demonstrated and initially evaluated. The differential evolution classifier is a nearest prototype vector based classifier that applies a global optimization algorithm, differential evolution, to determine the optimal values for all free parameters of the classifier model during the training phase of the classifier. The differential evolution classifier applies the individually optimized distance measure for each new data set to be classified is generalized to cover a pool of distances. Instead of optimizing a single distance measure for the given data set, the selection of the optimal distance measure from a predefined pool of alternative measures is attempted systematically and automatically. Furthermore, instead of only selecting the optimal distance measure from a set of alternatives, an attempt is made to optimize the values of the possible control parameters related with the selected distance measure. Specifically, a pool of alternative distance measures is first created and then the differential evolution algorithm is applied to select the optimal distance measure that yields the highest classification accuracy with the current data. After determining the optimal distance measures for the given data set together with their optimal parameters, all determined distance measures are aggregated to form a single total distance measure. The total distance measure is applied to the final classification decisions. The actual classification process is still based on the nearest prototype vector principle; a sample belongs to the class represented by the nearest prototype vector when measured with the optimized total distance measure. During the training process the differential evolution algorithm determines the optimal class vectors, selects optimal distance metrics, and determines the optimal values for the free parameters of each selected distance measure. The results obtained with the above method confirm that the choice of distance measure is one of the most crucial factors for obtaining higher classification accuracy. The results also demonstrate that it is possible to build a classifier that is able to select the optimal distance measure for the given data set automatically and systematically. After finding optimal distance measures together with optimal parameters from the particular distance measure results are then aggregated to form a total distance, which will be used to form the deviation between the class vectors and samples and thus classify the samples. This thesis also discusses two types of aggregation operators, namely, ordered weighted averaging (OWA) based multi-distances and generalized ordered weighted averaging (GOWA). These aggregation operators were applied in this work to the aggregation of the normalized distance values. The results demonstrate that a proper combination of aggregation operator and weight generation scheme play an important role in obtaining good classification accuracy. The main outcomes of the work are the six new generalized versions of previous method called differential evolution classifier. All these DE classifier demonstrated good results in the classification tasks.

Visual analytics for behavioral and niche market segmentation

Companies require information in order to gain an improved understanding of their customers. Data concerning customers, their interests and behavior are collected through different loyalty programs. The amount of data stored in company data bases has increased exponentially over the years and become difficult to handle. This research area is the subject of much current interest, not only in academia but also in practice, as is shown by several magazines and blogs that are covering topics on how to get to know your customers, Big Data, information visualization, and data warehousing. In this Ph.D. thesis, the Self-Organizing Map and two extensions of it – the Weighted Self-Organizing Map (WSOM) and the Self-Organizing Time Map (SOTM) – are used as data mining methods for extracting information from large amounts of customer data. The thesis focuses on how data mining methods can be used to model and analyze customer data in order to gain an overview of the customer base, as well as, for analyzing niche-markets. The thesis uses real world customer data to create models for customer profiling. Evaluation of the built models is performed by CRM experts from the retailing industry. The experts considered the information gained with help of the models to be valuable and useful for decision making and for making strategic planning for the future.