http://repositorio.ufc.br/handle/riufc/22107 2026-04-02T14:48:33Z http://repositorio.ufc.br/handle/riufc/83758 Título: Índice de criminalidade municipal: construção e análise de um índice composto a partir de indicadores criminais Autor(es): Freitas, Romulo Barros de Abstract: The Municipal Crime Index developed and presented in this study aims to measure and monitor the level of criminality across the 184 municipalities of Ceará over time. It draws on the concept of the Municipal Development Index calculated by the Institute for Research and Economic Strategy of Ceará. Crime indicators were organized into dimensions according to the characteristics of the victims, resulting in dimensional subindices and a global composite crime index. Standardization took into account the polarity of the indicators, and the resulting values were aggregated using the simple geometric mean for the dimensions, followed by the geometric mean for calculating the global composite crime index. The scale was inverted to facilitate the interpretation of the sub-indices and the global index, where values close to zero indicate municipalities with lower crime incidence, and values close to one indicate municipalities with higher crime occurrence. The classification of municipalities was carried out in ranges defined using the Jenks natural breaks method, with subsequent adjustment to rounded cut-off points. The results presented in this study cover the years 2021 to 2024, highlighting the temporal evolution of Ceará’s municipalities during this period, enabling detailed analyses and comparisons over time, and allowing for the identification of whether municipalities are progressing or regressing in their performance. Furthermore, through heat maps, it is possible to identify regional patterns, which can assist public managers in formulating more targeted public security strategies. Tipo: TCC 2025-01-01T00:00:00Z http://repositorio.ufc.br/handle/riufc/83725 Título: Uso de redes neurais convolucionais no diagnóstico de leucemia linfoblástica aguda Autor(es): Esteves, Pietro de Oliveira Abstract: Acute Lymphoblastic Leukemia (ALL) is an aggressive hematologic cancer that predominantly affects children and requires early diagnosis to improve survival rates. Traditional diagnostic methods, such as bone marrow aspiration and flow cytometry, are invasive, expensive, and often inaccessible in resource-limited settings (Pui et al., 2015). Studies such as (Ghaderzadeh et al., 2022) propose the use of Artificial Intelligence (AI) models to assist physicians in identifying the disease through imaging, focusing on the differentiation between benign cells (hematogones) and malignant lymphoblasts. However, these approaches are often complex and resource-intensive, including longer processing times. This monograph also proposes the use of Convolutional Neural Networks (CNNs) to support the screening and diagnosis of ALL from Peripheral Blood Smear (PBS) images, employing the EfficientNet-B3 architecture, known for balancing performance and efficiency (Tan e Le, 2019). Supported by the FastAI library, a classification algorithm was implemented and achieved an accuracy of 98.92% on the test set. The results were compared with those of the reference study, which used DenseNet201 combined with HSV color segmentation, indicating that the proposed model has the potential to achieve competitive performance with lower complexity. The adopted approach aims to support medical diagnosis with an intelligent, reliable, and more accessible tool for clinical settings with limited infrastructure, while maintaining computational efficiency. Tipo: TCC 2025-01-01T00:00:00Z http://repositorio.ufc.br/handle/riufc/82743 Título: Modelagem preditiva de partidas de league of legends usando aprendizado supervisionado Autor(es): Sousa, Francisco Luan Rodrigues de Abstract: This study explores the application of supervised learning models to predict the outcomes of League of Legends matches, one of the most popular esports worldwide. Various approaches were analyzed, including K-Nearest Neighbors (KNN), Logistic Regression, Gradient Boosting, and Decision Trees, using a dataset extracted from Oracle’s Elixir. The methodology involved careful feature selection, removal of irrelevant metadata, and model validation through metrics such as AUC-ROC and accuracy. Results indicate that Logistic Regression and KNN achieved the best performance, with an AUC of up to 0,930, demonstrating strong predictive capability. Beyond model accuracy, the study highlights the strategic importance of early-game resource differences and champion selection in decision-making within competitive settings. Tipo: TCC 2025-01-01T00:00:00Z http://repositorio.ufc.br/handle/riufc/72578 Título: Relatório de estágio supervisionado obrigatório II realizado na empresa FortBrasil administradora de cartões de crédito Autor(es): Barros, José Wariston Abstract: This report aims to describe the activities developed during the internship compulsory supervised II, at the company Fortbrasil Gestora de Cartão de Crédito, located in Fortaleza, Ceará. This company operates mainly with credit cards since the year 2005. This internship was guided by Professor Dr. Gualberto Segundo Agamez Montalvo and was supervised by the coordinator of the Data Science sector Lucas Pinheiro de Goes Carneiro. During the period from January 2, 2020 to April 7, 2020, various activities, such as the construction of management reports aimed at monitoring of the models that were used in the sector and treatment of the database for the availability of information for the sector. For the development of these activities, some computational tools already known during graduation, the most used being RStudio, Power BI and SQL Server. In addition to having developed myself in relation to the knowledge of these tools, I was also able to develop myself as a professional, having practical experiences of Desktop. Tipo: TCC 2020-01-01T00:00:00Z