Mestrado em Ciência da Computação (INF)

URI Permanente para esta coleção

http://200.137.215.59/tede/handle/tde/225

Navegar

Navegando Mestrado em Ciência da Computação (INF) por Por Orientador "Brancher, Jacques Duílio"

Agora exibindo 1 - 4 de 4
  • Nenhuma Miniatura disponível
    Item
    Análise dos microdados do Enade: proposta de uma ferramenta de exploração utilizando mineração de dados
    (Universidade Federal de Goiás, 2019-12-20) Araújo, Rodrigo Alexandrino; Brancher, Jacques Duílio; http://lattes.cnpq.br/7909976127880843; Brancher, Jacques Duílio; Sanches, Danilo Sipoli; Camos, Vitor Valério Souza
    One way to analyze higher education institutions and student performance is through the National Student Performance Examination (ENADE). From its results it is possible to make intelligent decisions for the improved teaching-learning process. However, in the analysis reports provided by Anísio Teixeira National Institute for Educational Studies and Research (INEP) only descriptive analyzes are available. Although the Institute provides ENADE’s evidence-related micro-data, advanced knowledge in data analysis and statistics is required to obtain more in-depth information about candidates. That said, this paper aims to use KDD techniques to develop an exploratory analysis tool for Enade microdata, together with a classification model capable of predicting student performance. For the elaboration of rules, several decision tree classification algorithms were used, in which CART stood out. The end result was a data analysis tool, which allows comparing higher education courses and institutions, and providing the best view of this information for the purpose of assisting decision making. Finally, an online questionnaire was distributed so that teachers, students and coordinators could evaluate and validate the developed system. After this study, the tool proved to be satisfactory and fulfills what is promised, and serves as a motivation to improve the work developed.
  • Nenhuma Miniatura disponível
    Item
    Fatores e evidências sobre o Exame Nacional do Ensino Médio (Enem): uma abordagem exploratória e experimental com mineração de dados
    (Universidade Federal de Goiás, 2021-03-29) Franco, Jacinto José; Brancher, Jacques Duílio; http://lattes.cnpq.br/7909976127880843; Brancher, Jacques Duílio; Ferreira, Deller James; Góis, Lourival Aparecido
    Enem is a test required by most Brazilian universities in order to select the best candidates to fill the vacancies available in undergraduate courses. Because it has been widely applied for 22 years, this assessment is an important thermometer of Brazilian education, thus allowing the identification of the factors that most contribute to student performance. The questionnaire applied at the time of Enem registration changes considerably over the years, but not everything that is collected is important to say whether the candidates' performance is high or low. Therefore, this work, based on this gap identified in the literature, we intend to list the most important factors for detecting high and low performance students. To achieve this goal, an experimental and exploratory approach is used regarding the use of attribute selection algorithms, validated with classifiers and replicated. The results indicate that from a limited set of 10 factors, it is possible to classify student performance with an average accuracy above 80\%. This result is interesting for researchers who aim to study Brazilian students, as up to 327 characteristics were collected in Enem in a single year. It is concluded that the combination of attribute selectors and classifiers allow the analysis of educational data to be facilitated, as it enables to focus on a restricted set of more significant factors. The main contributions of this work are constituted by the systematized experimental scheme, the experiments and patterns identified in the data, which subsidizes the realization of reflections on the student public. The main elements of this dissertation were condensed in an article, submitted and accepted for publication, the rest is in the process of being edited for publication in events in the area.
  • Nenhuma Miniatura disponível
    Item
    Mineração de dados educacionais baseada em grafos: uma análise em cursos de computação com alto índice de retenção
    (Universidade Federal de Goiás, 2021-04-19) Oliveira, João Lucas dos Santos; Brancher, Jacques Duílio; http://lattes.cnpq.br/7909976127880843; Brancher, Jacques Duílio; Silv, Nádia Félix Felipe da; Barros, Rodolfo Miranda de
    Student evasion and retention is a recurring problem in all areas of education. In Area ssuch as Educational Data Mining (MDE) have been used to mitigate such problems. In particular, the area of Graph-based Educational Data Mining (G-EDM) uses unconventional data mining techniques to represent student behavior. This analysis of students can be done both in physical and virtual environments, through complex networks and graphs. The students’ behavior shown by the graphs can express dimensional patterns that would not be expressed by tabular and statistical analyzes. The present work investigated three different techniques of representing student history to investigate the possible causes of retention and dropout in computer courses. The results show that it is possible to identify retention problems in curriculum and that the modeling of the curriculum in the form of agraph can show patterns that would not be possible to describe in tabular representation.
  • Nenhuma Miniatura disponível
    Item
    Fatores de impacto no desempenho acadêmico: um estudo de caso em cursos de computação
    (Universidade Federal de Goiás, 2021-09-23) Oliveira, Michelle Christiane da Silva; Brancher, Jacques Duílio; http://lattes.cnpq.br/7909976127880843; Brancher, Jacques Duílio; Ferreira, Deller James; Barros , Rodolfo Miranda de
    Through the computerized systems of universities, it is possible to have access to a lot of student data, from demographic, socioeconomic, admission, egress and performance data. Transforming these data into useful information for the academic society, both management and students, is a challenge. One of the ways to identify the impact factors on the academic performance of higher-level students is Educational Data Mining. Based on the results, it is possible to make academic, managerial and administrative decisions based on evidence. This study aims, through the use of Educational Data Mining techniques, to identify which factors impact the performance of higher education students in computing courses, having as a case study, the computing courses of the Instituto de Informática da Universidade Federal de Goiás, with a database of 2.501 incoming students between the years 2009 to 2019. Through Systematic Literature Review, the main algorithms used for educational data mining (analysis and prediction) were identified. The data base went through the data mining process (selection, pre-processing, data transformation, datamining), where a data set was initially defined, which allowed the generation of graphical views of various aspects of the profile of the data students. This dataset was then adjusted to be applied to the algorithms identified in the SLR, where it was possible to define a data model. With the application of these algorithms to the data model, it was possible to identify the algorithms that had the best performance (accuracy). And also analyze, through feature importance techniques, such as SHAP and correlation maps between Heatmaps attributes, which factors had the greatest impact on student performance.