Previsão de vento baseado em técnicas de Data Mining

Mestrado em Engenharia Electrotécnica – Sistemas Eléctricos de Energia

Modelo de data mining para detecção de tumores em exames de rastreio

Dissertação para obtenção do grau de Mestre em Engenharia Informática

Hook removal from a loggerhead sea turtle (Caretta caretta) : Report of a tricky surgery under difficult conditions.

A sub adult Caretta caretta was found on the 23rd August, 2014 ca. 16 nautical miles south off S. Miguel Island, Azores (Northeast Atlantic), with a large pelagic trawl hook inside its mouth. The individual was kept in a basin of sea water and sent by boat to Terceira Island following instructions by the Azores Regional Government via the Environmental Authority in order to be examined by the author and, if possible, undergo the necessary hook removal procedures. In this note, we describe the surgical procedures and how the turtle was evaluated both pre- and post-surgery.

Application of wetland systems in the treatment of acid rock mine drainage

The presented work was conducted within the Dissertation / Internship, branch of Environmental Protection Technology, associated to the Master thesis in Chemical Engineering by the Instituto Superior de Engenharia do Porto and it was developed in the Aquatest a.s, headquartered in Prague, in Czech Republic. The ore mining exploitation in the Czech Republic began in the thirteenth century, and has been extended until the twentieth century, being now evident the consequences of the intensive extraction which includes contamination of soil and sub-soil by high concentrations of heavy metals. The mountain region of Zlaté Hory was chosen for the implementation of the remediation project, which consisted in the construction of three cells (tanks), the first to raise the pH, the second for the sedimentation of the formed precipitates and a third to increase the process efficiency in order to reduce high concentrations of metals, with special emphasis on iron, manganese and sulfates. This project was initiated in 2005, being pioneer in this country and is still ongoing due to the complex chemical and biological phenomenon’s inherent to the system. At the site where the project was implemented, there is a natural lagoon, thereby enabling a comparative study of the two systems (natural and artificial) regarding the efficiency of both in the reduction/ removal of the referred pollutants. The study aimed to assist and cooperate in the ongoing investigation at the company Aquatest, in terms of field work conducted in Zlaté Hory and in terms of research methodologies used in it. Thereby, it was carried out a survey and analysis of available data from 2005 to 2008, being complemented by the treatment of new data from 2009 to 2010. Moreover, a theoretical study of the chemical and biological processes that occurs in both systems was performed. Regarding the field work, an active participation in the collection and in situ sample analyzing of water and soil from the natural pond has been attained, with the supervision of Engineer, Irena Šupiková. Laboratory analysis of water and soil were carried out by laboratory technicians. It was found that the natural lagoon is more efficient in reducing iron and manganese, being obtained removal percentages of 100%. The artificial lagoon had a removal percentage of 90% and 33% for iron and manganese respectively. Despite the minor efficiency of the constructed wetland, it must be pointed out that this system was designed for the treatment and consequent reduction of iron. In this context, it can conclude that the main goal has been achieved. In the case of sulphates, the removal optimization is yet a goal to be achieved not only in the Czech Republic but also in other places where this type of contamination persists. In fact, in the natural lagoon and in the constructed wetland, removal efficiencies of 45% and 7% were obtained respectively. It has been speculated that the water at the entrance of both systems has different sources. The analysis of the collected data shows at the entrance of the natural pond, a concentration of 4.6 mg/L of total iron, 14.6 mg/L of manganese and 951 mg/L of sulphates. In the artificial pond, the concentrations are 27.7 mg/L, 8.1 mg/L and 382 mg/L respectively for iron, manganese and sulphates. During 2010 the investigation has been expanded. The study of soil samples has started in order to observe and evaluate the contribution of bacteria in the removal of heavy metals being in its early phase. Summarizing, this technology has revealed to be an interesting solution, since in addition to substantially reduce the mentioned contaminants, mostly iron, it combines the low cost of implementation with an reduced maintenance, and it can also be installed in recreation parks, providing habitats for plants and birds.

Modelo de data mining para deteção de embolias pulmonares

Trabalho de Projeto para obtenção do grau de Mestre em Engenharia Informática e de Computadores

Estudo da tratabilidade laboratorial e optimização experimental do tratamento de águas residuais de indústrias lácteas

O presente trabalho experimental teve como objectivos estudar a tratabilidade dum efluente lácteo utilizando a técnica de coagulação/floculação e avaliar a possibilidade de utilização do carvão activado granulado (CAG) Aquasorb 2000 como adsorvente para a remoção de compostos orgânicos presentes nos efluentes lácteos pré–tratados por coagulação/floculação, funcionando como um tratamento de polimento. No estudo da tratabilidade do efluente por coagulação/floculação investigou-se a influência de determinadas variáveis como o tipo e dose de coagulante e pH, a fim de encontrar as melhores condições operatórias. A utilização da referida técnica visou a redução do valor de concentração de alguns parâmetros: carência química de oxigénio (CQO); fósforo total e turvação, tendo sido utilizados efluentes desta indústria recolhidos em diferentes datas e após tratamento biológico, designados por A, B, C e D apresentando características diferentes. Sendo que o efluente A apresentava valores de CQO, fósforo total e turvação de 500 mg O2/L;32 mg P/L e 40 NTU respectivamente, o efluente B 1400 mg O2/L; 120 mg P/L e 80 NTU respectivamente, o efluente C 12300 mg O2/L; 87 mg P/L e 350 NTU respectivamente e o efluente D 340 mg O2/L; 33 mg P/L e 42 NTU respectivamente. Os coagulantes estudados foram hidróxido de cálcio (HC), sulfato de alumínio (SA) e tricloreto de ferro (TF). Verificou-se que o coagulante com maior eficácia nos efluentes estudados foi o TF. As maiores remoções de CQO, fósforo total e turvação, 89%, 99,9% e 99%, respectivamente, foram obtidas para o efluente C, com uma dosagem de TF de 4 g/L e com um pH entre 6 e 7. Entre os efluentes estudados este era o que apresentava valores iniciais mais elevados para qualquer um destes parâmetros. As melhores percentagens de remoção obtidas com o HC, para a CQO, fósforo total e turvação foram de 59%, 99% e 91%, respectivamente, com uma dosagem de HC de 1 g/L e com um pH entre 10 e 11,5 foram conseguidas no tratamento do efluente D, com o qual se alcançaram também as melhores remoções de CQO, fósforo total e turvação de 65%, 99% e 87%, respectivamente, quando se utilizou o coagulante SA, com uma dosagem de 2 g/L e com um pH entre 7 e 7,5. Relativamente ao volume de lamas produzido neste processo pela utilização dos diferentes coagulantes no tratamento dos efluentes referidos concluiu-se que o coagulante que gera menor volume de lamas é o HC, sendo o SA aquele que origina um maior volume. Submeteu-se posteriormente o efluente D, pré-tratado por coagulação/floculação, a um processo de adsorção em batch utilizando o CAG Aquasorb 2000, onde se conseguiu uma remoção de CQO de 48%, alcançando para este parâmetro o valor de 63 mg O2/L, nas condições operatórias que correspondem a uma massa de CAG de 12,5 g/L e um tempo de contacto de 3 horas. Quanto aos custos associados com os coagulantes, o que menores custos apresenta é o HC (150 €/ton), seguido pelo TF (250 €/ton) e por ultimo o SA (340 €/ton). Sendo que o efluente quando tratado com TF e SA é necessário uma correcção do pH do meio para que estes coagulantes actuem eficazmente, em que essa correcção de pH é realizada com hidróxido de sódio (540 €/ton). Realizaram-se ainda estudos de equilíbrio de adsorção com o carvão activado referido e o azul-de-metileno usando diferentes concentrações deste (50 mg/L; 100 mg/L e 200 mg/L) e diferentes massas de CAG (0,1g; 0,2g; 0,3g; 0,4g e 0,5g). A temperatura a que se realizaram estes ensaios foi de 28,7ºC e o volume de azul-de-metileno foi de 200 mL. Verificou-se que os melhores resultados obtidos foram para uma concentração de adsorvato de 100 mg/L. Ajustaram-se os modelos de Langmuir e Freundlich às isotérmicas obtidas tendo correlações mais elevadas para a concentração de 100 mg/L de corante (azul de metileno), sendo o modelo de Freundlich aquele que melhor se ajustou apresentando uma correlação quadrática de 0,9744 e os seguintes parâmetros Kf = 6,59 e n = 5,33, enquanto que o de Langmuir apresentou uma correlação quadrática de 0,9583 e os seguintes parâmetros qmáx = 83,3 mg/g de adsorvente e K = 20 L/mg de adsorvato.. Verificou-se que a capacidade de adsorção promovida pelo CAG, em relação ao azul-demetileno, obtida experimentalmente, 83,3 mg/g, é muito inferior à capacidade de adsorção teoricamente prevista pela ficha técnica deste carvão, 280 mg/g a uma temperatura de 25ºC, o que pode indiciar que o carvão utilizado não estaria nas melhores condições.

Exploração de dados de uma solução mobile wallet usando as técnicas de data mining

A descoberta de conhecimento em dados hoje em dia é um ponto forte para as empresas. Atualmente a CardMobili não dispõe de qualquer sistema de mineração de dados, sendo a existência deste uma mais-valia para as suas operações de marketing diárias, nomeadamente no lançamento de cupões a um grupo restrito de clientes com uma elevada probabilidade que os mesmos os utilizem. Para isso foi analisada a base de dados da aplicação tentando extrair o maior número de dados e aplicadas as transformações necessárias para posteriormente serem processados pelos algoritmos de mineração de dados. Durante a etapa de mineração de dados foram aplicadas as técnicas de associação e classificação, sendo que os melhores resultados foram obtidos com técnicas de associação. Desta maneira pretende-se que os resultados obtidos auxiliem o decisor na sua tomada de decisões.

Copper, nickel and zinc removal by peanut hulls: batch and column studies in mono, tri-component systems and with real effluent

The main goal of this research study was the removal of Cu(II), Ni(II) and Zn(II) from aqueous solutions using peanut hulls. This work was mainly focused on the following aspects: chemical characterization of the biosorbent, kinetic studies, study of the pH influence in mono-component systems, equilibrium isotherms and column studies, both in mono and tri-component systems, and with a real industrial effluent from the electroplating industry. The chemical characterization of peanut hulls showed a high cellulose (44.8%) and lignin (36.1%) content, which favours biosorption of metal cations. The kinetic studies performed indicate that most of the sorption occurs in the first 30 min for all systems. In general, a pseudo-second order kinetics was followed, both in mono and tri-component systems. The equilibrium isotherms were better described by Freundlich model in all systems. Peanut hulls showed higher affinity for copper than for nickel and zinc when they are both present. The pH value between 5 and 6 was the most favourable for all systems. The sorbent capacity in column was 0.028 and 0.025 mmol g-1 for copper, respectively in mono and tri-component systems. A decrease of capacity for copper (50%) was observed when dealing with the real effluent. The Yoon-Nelson, Thomas and Yan’s models were fitted to the experimental data, being the latter the best fit.

Removal of Cd(II), Zn(II) and Pb(II) from aqueous solutions by brown marine macro algae: kinetic modelling

Specific marine macro algae species abundant at the Portuguese coast (Laminaria hyperborea, Bifurcaria bifurcata, Sargassum muticum and Fucus spiralis) were shown to be effective for removing toxic metals (Cd(II), Zn(II) and Pb(II)) from aqueous solutions. The initial metal concentrations in solution were about 75–100 mg L−1. The observed biosorption capacities for cadmium, zinc and lead ions were in the ranges of 23.9–39.5, 18.6–32.0 and 32.3–50.4 mg g−1, respectively. Kinetic studies revealed that the metal uptake rate was rather fast, with 75% of the total amount occurring in the first 10 min for all algal species. Experimental data were well fitted by a pseudo-second order rate equation. The contribution of internal diffusion mechanism was significant only to the initial biosorption stage. Results indicate that all the studied macro algae species can provide an efficient and cost-effective technology for eliminating heavy metals from industrial effluents.

Copper and lead removal by peanut hulls: equilibrium and kinetic studies

This research work aims to study the use of peanut hulls, an agricultural and food industry waste, for copper and lead removal through equilibrium and kinetic parameters evaluation. Equilibrium batch studies were performed in a batch adsorber. The influence of initial pH was evaluated (3–5) and it was selected between 4.0 and 4.5. The maximum sorption capacities obtained for the Langmuir model were 0.21 ± 0.03 and 0.18 ± 0.02 mmol/g, respectively for copper and lead. In bi-component systems, competitive sorption of copper and lead was verified, the total amount adsorbed being around 0.21 mmol of metal per gram of material in both mono and bi-component systems. In the kinetic studies equilibrium was reached after 200 min contact time using a 400 rpm stirring rate, achieving 78% and 58% removal, in mono-component system, for copper and lead respectively. Their removal follows a pseudo-second-order kinetics. These studies show that most of the metals removal occurred in the first 20 min of contact, which shows a good uptake rate in all systems.

Adsorption kinetics of removal of yellow lanasol dyestuff using gallinaceous feathers

Gallinaceous feathers are an abundant solid waste from the poultry processing industries, which poses disposal problems. A kinetic study dealing with the adsorption process of wool reactive dye, Yellow Lanasol 4G (CI Reactive Yellow 39), on gallinaceous (Gallus gallus, Cobb 500) feathers was carried out. The main research goals of this work were to evaluate the viability of using this waste as adsorbent and to study the kinetics of the adsorption process, using a synthetic effluent. The characterization of feathers was performed by scanning electron microscopy, mercury porosimetry and B.E.T. method. The study of several factors (stirring, particles size, initial dye concentration and temperature) showed their influence over the adsorption process. An adapted version of the Schumckler and Goldstein´s unreacted core model fitted the experimental data. The best fit was obtained when the rate-limiting step was the diffusion through the reacted layer, which was expected considering the size of the dyestuff molecules. The comparison with the granular activated carbon (GAC) Sutcliffe GAC 10-30 indicate that in spite of the high adsorption capacities shown by feathers the GAC presented higher values, the values obtained were respectively 150 and 219 mg g-1, for an initial concentration of 500 mg L-1. The results obtained might open future perspectives both to the valorization of feathers and to the economical treatment of textile wastewaters.

Mining protein structure data

The principal topic of this work is the application of data mining techniques, in particular of machine learning, to the discovery of knowledge in a protein database. In the first chapter a general background is presented. Namely, in section 1.1 we overview the methodology of a Data Mining project and its main algorithms. In section 1.2 an introduction to the proteins and its supporting file formats is outlined. This chapter is concluded with section 1.3 which defines that main problem we pretend to address with this work: determine if an amino acid is exposed or buried in a protein, in a discrete way (i.e.: not continuous), for five exposition levels: 2%, 10%, 20%, 25% and 30%. In the second chapter, following closely the CRISP-DM methodology, whole the process of construction the database that supported this work is presented. Namely, it is described the process of loading data from the Protein Data Bank, DSSP and SCOP. Then an initial data exploration is performed and a simple prediction model (baseline) of the relative solvent accessibility of an amino acid is introduced. It is also introduced the Data Mining Table Creator, a program developed to produce the data mining tables required for this problem. In the third chapter the results obtained are analyzed with statistical significance tests. Initially the several used classifiers (Neural Networks, C5.0, CART and Chaid) are compared and it is concluded that C5.0 is the most suitable for the problem at stake. It is also compared the influence of parameters like the amino acid information level, the amino acid window size and the SCOP class type in the accuracy of the predictive models. The fourth chapter starts with a brief revision of the literature about amino acid relative solvent accessibility. Then, we overview the main results achieved and finally discuss about possible future work. The fifth and last chapter consists of appendices. Appendix A has the schema of the database that supported this thesis. Appendix B has a set of tables with additional information. Appendix C describes the software provided in the DVD accompanying this thesis that allows the reconstruction of the present work.

Data mining languages for business intelligence

Doctoral Thesis in Information Systems and Technologies Area of Engineering and Manag ement Information Systems

Modeling radionuclides dispersion and deposition downwind of a coal-fired power plant

In this study the inhalation doses and respective risk are calculated for the population living within a 20 km radius of a coal-fired power plant. The dispersion and deposition of natural radionuclides were simulated by a Gaussian dispersion model estimating the ground level activity concentration. The annual effective dose and total risk were 0.03205 mSv/y and 1.25 x 10-8, respectively. The effective dose is lower than the limit established by the ICRP and the risk is lower than the limit proposed by the U.S. EPA, which means that the considered exposure does not pose any risk for the public health.

Recomendações personalizadas de alunos em sistemas de hipermédia adaptativa educacional usando Data Mining

O aumento de tecnologias disponíveis na Web favoreceu o aparecimento de diversas formas de informação, recursos e serviços. Este aumento aliado à constante necessidade de formação e evolução das pessoas, quer a nível pessoal como profissional, incentivou o desenvolvimento área de sistemas de hipermédia adaptativa educacional - SHAE. Estes sistemas têm a capacidade de adaptar o ensino consoante o modelo do aluno, características pessoais, necessidades, entre outros aspetos. Os SHAE permitiram introduzir mudanças relativamente à forma de ensino, passando do ensino tradicional que se restringia apenas ao uso de livros escolares até à utilização de ferramentas informáticas que através do acesso à internet disponibilizam material didático, privilegiando o ensino individualizado. Os SHAE geram grande volume de dados, informação contida no modelo do aluno e todos os dados relativos ao processo de aprendizagem de cada aluno. Facilmente estes dados são ignorados e não se procede a uma análise cuidada que permita melhorar o conhecimento do comportamento dos alunos durante o processo de ensino, alterando a forma de aprendizagem de acordo com o aluno e favorecendo a melhoria dos resultados obtidos. O objetivo deste trabalho foi selecionar e aplicar algumas técnicas de Data Mining a um SHAE, PCMAT - Mathematics Collaborative Educational System. A aplicação destas técnicas deram origem a modelos de dados que transformaram os dados em informações úteis e compreensíveis, essenciais para a geração de novos perfis de alunos, padrões de comportamento de alunos, regras de adaptação e pedagógicas. Neste trabalho foram criados alguns modelos de dados recorrendo à técnica de Data Mining de classificação, abordando diferentes algoritmos. Os resultados obtidos permitirão definir novas regras de adaptação e padrões de comportamento dos alunos, poderá melhorar o processo de aprendizagem disponível num SHAE.