Enhancing the selection of a model-based clustering with external categorical variables

In cluster analysis, it can be useful to interpret the partition built from the data in the light of external categorical variables which are not directly involved to cluster the data. An approach is proposed in the model-based clustering context to select a number of clusters which both fits the data well and takes advantage of the potential illustrative ability of the external variables. This approach makes use of the integrated joint likelihood of the data and the partitions at hand, namely the model-based partition and the partitions associated to the external variables. It is noteworthy that each mixture model is fitted by the maximum likelihood methodology to the data, excluding the external variables which are used to select a relevant mixture model only. Numerical experiments illustrate the promising behaviour of the derived criterion. © 2014 Springer-Verlag Berlin Heidelberg.

Modelo preditivo da criminalidade – georeferenciação ao concelho de Lisboa

Dissertação apresentada como requisito parcial para a obtenção do grau de Mestre em Estatística e Gestão de Informação

Enhancing the selection of a model-based clustering with external categorical variables

In cluster analysis, it can be useful to interpret the partition built from the data in the light of external categorical variables which are not directly involved to cluster the data. An approach is proposed in the model-based clustering context to select a number of clusters which both fits the data well and takes advantage of the potential illustrative ability of the external variables. This approach makes use of the integrated joint likelihood of the data and the partitions at hand, namely the model-based partition and the partitions associated to the external variables. It is noteworthy that each mixture model is fitted by the maximum likelihood methodology to the data, excluding the external variables which are used to select a relevant mixture model only. Numerical experiments illustrate the promising behaviour of the derived criterion.

Estimação de Modelos de Adopção e Difusão de Tecnologias de Informação e Comunicação em Portugal

Dissertação apresentada como requisito parcial para obtenção do grau de Mestre em Estatística e Gestão de Informação

Exploraty Multivariate Statistical Methods Applied to Pharmaceutical Industry CRM Data

Dissertação apresentada como requisito parcial para obtenção do grau de Mestre em Estatística e Gestão de Informação

SPATIOTEMPORAL TRENDS OF CASES OF PANDEMIC INFLUENZA A(H1N1)PDM09 IN ARGENTINA, 2009-2012

The aim of this paper was to analyze the spatiotemporal variations of cases of influenza A(H1N1)pdm09 in Argentina. A space-time permutation scan statistic was performed to test the non-randomness in the interaction between space and time in reported influenza A(H1N1)pdm09 cases. In 2009, two clusters were recorded in the east of Buenos Aires Province (May and June) and in the central and northern part of Argentina (July and August). Between 2011 and 2012, clusters near areas bordering other countries were registered. Within the clusters, in 2009, the high notification rates were first observed in the school-age population and then extended to the older population (15-59 years). From 2011 onwards, higher rates of reported cases of influenza A(H1N1)pdm09 occurred in children under five years in center of the country. Two stages of transmission of influenza A(H1N1)pdm09 can be characterized. The first stage had high rates of notification and a possible interaction with individuals from other countries in the major cities of Argentina (pattern of hierarchy), and the second stage had an increased interaction in some border areas without a clear pattern of hierarchy. These results suggest the need for greater coordination in the Southern Cone countries, in order to implement joint prevention and vaccination policies.

CO2 Concentration in Day Care Centres is Related to Wheezing in Attending Children

Poor ventilation at day care centres (DCCs) was already reported, although its effects on attending children are not clear. This study aimed to evaluate the association between wheezing in children and indoor CO2 (a ventilation surrogate marker) in DCC and to identify behaviours and building characteristics potentially related to CO2. In phase I, 45 DCCs from Lisbon and Oporto (Portugal) were selected through a proportional stratified random sampling. In phase II, 3 months later, 19 DCCs were further reassessed after cluster analysis for the greatest difference comparison. In both phases, children’s respiratory health was assessed by ISAAC-derived questionnaires. Indoor CO2 concentrations and building characteristics of the DCC were evaluated in both phases, using complementary methods. Mixed effect models were used to analyze the data. In phase I, which included 3,186 children (mean age 3.1±1.5 years), indoor CO2 concentration in the DCC rooms was associated with reported wheezing in the past 12months (27.5 %) (adjusted odds ratio (OR) for each increase of 200 ppm 1.04, 95 % CI 1:01 to 1:07). In phase II, the association in the subsample of 1,196 children seen in 19 out of the initial 45 DCCs was not significant (adjusted OR 1.02, 95 % CI 0.96 to 1.08). Indoor CO2 concentration was inversely associated with the practices of opening Windows and internal doors and with higher wind velocity. A positive trend was observed between CO2 and prevalence of reported asthma (4.7 %). Conclusion: Improved ventilation is needed to achieve a healthier indoor environment in DCC.

Environment and Health in Children Day Care Centres (ENVIRH) - Study Rationale and Protocol

Background: Indoor air quality (IAQ) is considered an important determinant of human health. The association between exposure to volatile organic compounds, particulate matter, house dust mite, molds and bacteria in day care centers (DCC) is not completely clear. The aim of this project was to study these effects. Methods --- study design: This study comprised two phases. Phase I included an evaluation of 45 DCCs (25 from Lisbon and 20 from Oporto, targeting 5161 children). In this phase, building characteristics, indoor CO2 and air temperature/relative humidity, were assessed. A children’s respiratory health questionnaire derived from the ISAAC (International Study on Asthma and Allergies in Children) was also distributed. Phase II encompassed two evaluations and included 20 DCCs selected from phase I after a cluster analysis (11 from Lisbon and 9 from Oporto, targeting 2287 children). In this phase, data on ventilation, IAQ, thermal comfort parameters, respiratory and allergic health, airway inflammation biomarkers, respiratory virus infection patterns and parental and child stress were collected. Results: In Phase I, building characteristics, occupant behavior and ventilation surrogates were collected from all DCCs. The response rate of the questionnaire was 61.7% (3186 children). Phase II included 1221 children. Association results between DCC characteristics, IAQ and health outcomes will be provided in order to support recommendations on IAQ and children’s health. A building ventilation model will also be developed. Discussion: This paper outlines methods that might be implemented by other investigators conducting studies on the association between respiratory health and indoor air quality at DCC.

Digital Divide Across the European Union

Dissertação apresentada como requisito parcial para obtenção do grau de Mestre em Estatística e Gestão de Informação

Food Allergy and Anaphylaxis in Infants and Preschool-Age Children

Food allergy (FA) prevalence data in infants and preschool-age children are sparse, and proposed risk factors lack confirmation. In this study, 19 children’s day care centers (DCC) from 2 main Portuguese cities were selected after stratification and cluster analysis. An ISAAC’s (International Study of Asthma and Allergies in Childhood) derived health questionnaire was applied to a sample of children attending DCCs. Outcomes were FA parental report and anaphylaxis. Logistic regression was used to explore potential risk factors for reported FA. From the 2228 distributed questionnaires, 1217 were included in the analysis (54.6%). Children’s median age was 3.5 years, and 10.8% were described as ever having had FA. Current FA was reported in 5.7%. Three (0.2%) reports compatible with anaphylaxis were identified. Reported parental history of FA, personal history of atopic dermatitis, and preterm birth increased the odds for reported current FA. A high prevalence of parental-perceived FA in preschool-age children was identified. Risk factor identification may enhance better prevention.

Building clusters for CRM strategies by mining airlines customer data

Trabalho de Projeto apresentado como requisito parcial para obtenção do grau de Mestre em Estatística e Gestão de Informação

Dynamics of innovation patterns: some insights for Portugal

Dissertação apresentada como requisito parcial para obtenção do grau de Mestre em Gestão de Informação

Digital Divide Across the European Union

Dissertação apresentada como requisito parcial para obtenção do grau de Mestre em Estatística e Gestão de Informação

How can clusters help in the product type choice? The case of a specialized retail store

This work project (WP) is a study about a clustering strategy for Sport Zone. The general cluster study’s objective is to create groups such that within each group the individuals are similar to each other, but should be different among groups. The clusters creation is a mix of common sense, trial and error and some statistical supporting techniques. Our particular objective is to support category managers to better define the product type to be displayed in the stores’ shelves by doing store clusters. This research was carried out for Sport Zone, and comprises an objective definition, a literature review, the clustering activity itself, some factor analysis and a discriminant analysis to better frame our work. Together with this quantitative part, a survey addressed to category managers to better understand their key drivers, for choosing the type of product of each store, was carried out. Based in a non-random sample of 65 stores with data referring to 2013, the final result was the choice of 6 store clusters (Figure 1) which were individually characterized as the main outcome of this work. In what relates to our selected variables, all were important for the distinction between clusters, which proves the adequacy of their choice. The interpretation of the results gives category managers a tool to understand which products best fit the clustered stores. Furthermore, as a side finding thanks to the clusterization, a STP (Segmentation, Targeting and Positioning) was initiated, being this WP the first steps of a continuous process.

Parallelization of kinetic Monte Carlo algorithm to simulate AL3Sc precipitation

The present paper reports the precipitation process of Al3Sc structures in an aluminum scandium alloy, which has been simulated with a synchronous parallel kinetic Monte Carlo (spkMC) algorithm. The spkMC implementation is based on the vacancy diffusion mechanism. To filter the raw data generated by the spkMC simulations, the density-based clustering with noise (DBSCAN) method has been employed. spkMC and DBSCAN algorithms were implemented in the C language and using MPI library. The simulations were conducted in the SeARCH cluster located at the University of Minho. The Al3Sc precipitation was successfully simulated at the atomistic scale with the spkMC. DBSCAN proved to be a valuable aid to identify the precipitates by performing a cluster analysis of the simulation results. The achieved simulations results are in good agreement with those reported in the literature under sequential kinetic Monte Carlo simulations (kMC). The parallel implementation of kMC has provided a 4x speedup over the sequential version.