https://repositorio.ufpb.br/jspui/handle/123456789/2351 2026-04-30T21:08:57Z https://repositorio.ufpb.br/jspui/handle/123456789/35395 Título: Diferenciação dos fatores estáticos na população brasileira: uma abordagem demográfica e estatística Autor(es): Almeida, Jéssika Helen de Orientador: Silva, Everlane Suane de Araújo Abstract: The description of a population in the present helps to understand its behavior in the future. Accordingly, this study aimed to analyze the static demographic factors of the Brazilian population through a comparative approach between the 2010. Demographic indicators such as dependency ratio, aging index, median age, and sex ratio were used to understand regional variations in age structure and sex composition. Age pyramids were analyzed as a visual tool to represent population growth by age group and sex, contributing to the identification of changes in the patterns observed over time. The research included statistical tests aimed at verifying the normality of the variables, a necessary condition for applying subsequent methods. Then, a Multivariate analysis of Variance (MANOVA) was conducted to simultaneously examine the effects of time and location on the selected indicators. A univariate analysis of variance (ANOVA) was also applied to each indicator in order to identify, individually, which indicators showed statistically significant changes between the years studied. The results revealed relevant differences in the static factors of the population across Brazilian regions, such as the increase in the proportion of elderly people in certain areas, a greater presence of young people in others, and variations in the sex ratio. The data analysis allowed the observation of consistent changes in the population structure, reflecting distinct social, economic, and historical transformations among the territories. Thus, the study provided valuable insights for understanding changes in the static factors of the Brazilian population, serving as a basis for planning public policies aimed at the country’s demographic reality. Editor: Universidade Federal da Paraíba Tipo: TCC 2025-05-05T00:00:00Z https://repositorio.ufpb.br/jspui/handle/123456789/35138 Título: Mortalidade por doenças crônicas no Semiárido Brasileiro: hipertensão arterial e diabetes mellitus Autor(es): Nascimento, Teresa Tamires Bezerra do Orientador: Silva, Everlane Suane de Araújo da Abstract: This study analyzed the patterns of proportional mortality and the crude and agestandardized mortality rates due to Diabetes Mellitus and Hypertension in the Brazilian Semi-arid region, for the years 2018 and 2021, with breakdowns by State and sex. The general objective was to analyze the behavior of mortality due to Hypertension and Diabetes Mellitus in the Brazilian semi-arid region in the years 2018 and 2021, considering regional and sex-related differences. The results revealed a mortality profile marked by a concentration of deaths in older age groups, typical of non-communicable chronic diseases, but with relevant nuances between States and between men and women. The proportional mortality curves showed that, despite the predominance of deaths from the age of 60 onward—especially in the 80+ age group—some States still reported significant proportions in intermediate age ranges, suggesting early and potentially avoidable deaths, particularly among men. The States of Espírito Santo and Sergipe presented atypical patterns, with abrupt fluctuations and irregular concentrations, indicating either inconsistencies in records or specific demographic realities. Meanwhile, the States of Piauí, Maranhão, and Minas Gerais showed persistently high standardized mortality rates, reflecting a more intense chronic disease burden possibly associated with structural inequalities in access to continuous care. The cluster analyses reinforced the heterogeneity among the States: while Ceará, Bahia, and Rio Grande do Norte formed a group with lower rates and temporal stability, others such as Piauí and Maranhão remained on the opposite end, with persistently high indicators for both men and women. The exclusion of Sergipe in some analyses was necessary to preserve the robustness of the clusters, due to its statistical outlier behavior in 2018. By integrating proportional and standardized mortality analyses, it was found that mortality from these diseases in the Brazilian Semi-arid region is not merely a result of population aging, but rather an outcome of the interplay between social, healthcare, and territorial factors. The findings highlighted the need for differentiated public policies that consider local specificities, prioritize prevention, and strengthen primary healthcare—especially in settings of higher epidemiological vulnerability. Editor: Universidade Federal da Paraíba Tipo: TCC 2025-05-06T00:00:00Z https://repositorio.ufpb.br/jspui/handle/123456789/35103 Título: Uso de aprendizado de máquina na modelagem preditiva dos valores de imóveis na cidade de João Pessoa Autor(es): Pereira, Gabriel de Jesus Orientador: Marinho, Pedro Rafael Diniz Abstract: Brazil’s urban trajectory underwent significant changes throughout the 20th century. In 1940, only 31% of the population lived in urban areas, but this figure had already surpassed 50% by 1970 (WAGNER; WARD, 1980). The accelerated growth of cities increased the demand for housing, leading to the emergence of public policies and financing mechanisms such as the Sistema Financeiro de Habitação (SFH) and, later, the Sistema de Financiamento Imobiliário (SFI). This process directly influenced the development of the real estate market in various cities, including João Pessoa, capital of the state of Paraíba, which has been gaining national prominence. By November 2024, the city recorded an accumulated appreciation of 16.13% in residential properties over the previous 12 months, ranking among the capitals with the highest growth in the sector. In light of this scenario of expansion and appreciation, this study aimed to develop a machine learning-based model to predict property values in João Pessoa, using data collected through web scraping from real estate websites. Various machine learning techniques were applied to build and evaluate predictive models, with the final model based on the Gradient Boosting algorithm showing satisfactory performance, with an 𝑅2 of 85.42%, a MAPE of 18.21%, and an RMSE of 227,677.44. In addition to modeling, interpretability techniques such as SHAP, LIME, ICE, and PDP were used to assess variable importance and behavioral patterns in price estimation. The results indicated that the property area was the most influential factor in predictions, followed by the ratio between the number of bedrooms and area, number of parking spaces, geographic coordinates, number of bedrooms, and the average rental value in the neighborhood. Finally, the data and predictive model were integrated into a web application developed to assist in the property valuation process in João Pessoa. The application allows users to make predictions based on provided characteristics and visualize descriptive maps and charts, offering a data-driven approach to the local real estate market. Editor: Universidade Federal da Paraíba Tipo: TCC 2025-05-05T00:00:00Z https://repositorio.ufpb.br/jspui/handle/123456789/34881 Título: Análise da fecundidade adolescente no semiárido brasileiro por regressão logística binária Autor(es): Tomaz, Tânia Barbosa Orientador: Silva, Everlane Suane de Araújo da Abstract: Adolescent fertility is a significant demographic phenomenon for understanding social and territorial inequalities in Brazil and constitutes one of the indicators of SDG 3 of the 2030 Agenda. Its implications go beyond reproductive aspects, impacting maternal and child health, educational trajectories, access to the labor market, and the perpetuation of intergenerational poverty, especially in vulnerable contexts. Despite the national trend of declining fertility rates, this reduction has not occurred uniformly across age groups and regions. This study aimed to analyze the sociodemographic factors associated with fertility levels among adolescents aged 15 to 19 in Brazil’s Semi-Arid region in 2023. A quantitative approach was adopted using microdata from SINASC (2019–2023) and population estimates from IBGE, covering 1,473 municipalities. The methodology involved four main steps: data processing, assessment of data quality, analysis of adolescent fertility trends, and application of a Binary Logistic Regression model. The dependent variable was the adolescent-specific fertility rate (ASFR), coded as 1 for municipalities below the national average (0.0391) and 0 otherwise. Explanatory variables included maternal education, marital status, number of previous children, race/skin color, and territorial typology. Results indicated that the presence of adolescents with lower secondary education increased the likelihood of a municipality having an ASFR below the national average by 3.75%. Additionally, municipalities classified as adjacent rural, adjacent intermediate, and remote rural areas showed a higher probability of low adolescent fertility compared to urban areas, suggesting that factors such as migration, cultural changes, or possible underreporting may have influenced these outcomes, particularly in remote rural areas. The presence of married adolescents increased the likelihood by only 1.7%, indicating that while marriage remains associated with fertility, its effect was limited. Variables such as race/skin color and previous childbirths showed no statistically significant association, reflecting the complexity of sociodemographic interactions with adolescent fertility in the Semi-Arid region. The model demonstrated good performance, with an AUC of approximately 73%, sensitivity of 54.2%, and specificity of 80.8%, supporting its applicability. It is concluded that, although adolescent fertility is declining in the Brazil’s Semi-Arid, territorial disparities persist and require context-sensitive and targeted public policies. Editor: Universidade Federal da Paraíba Tipo: TCC 2025-05-06T00:00:00Z