Aranha, Eduardo Henrique da SilvaSilva, Francisco Genivan2018-11-222018-11-222018-07-27SILVA, Francisco Genivan. Análise do comportamento de estudantes em videoaulas. 2018. 96f. Dissertação (Mestrado em Sistemas e Computação) - Centro de Ciências Exatas e da Terra, Universidade Federal do Rio Grande do Norte, Natal, 2018.https://repositorio.ufrn.br/jspui/handle/123456789/26119Distance Education and the use of e-learning systems contribute to the great generation of educational data. In this way, the use of databases and the storage of execution logs make the information more easily accessible and conducive to the investigation of educational processes. Methodologies for automatic extraction of useful information from large volumes of data, especially data mining, have significantly contributed to improvements in the field of education. However, most traditional methods are focused solely on the data or how they are structured, with no major concern with the educational process as a whole. In addition, little attention has been given to information about student behavior during resource use and educational media. Taking this into account, we have observed that video lesson have been used as a significant part of several courses offered in educational institutions, demonstrating that the video culture is increasingly disseminated and is part of students' daily lives. Therefore, we understand that analyzing their behavior during the execution of the videos can contribute to a more precise evaluation of the quality of the subjects covered and the way they were worked out. Thus, this master's work was constituted by conducting studies conducted in order to investigate how students behave during the use of video lessons, which is done in order to highlight the benefits of this type of analysis for education. The evaluation of video lessons occurs through a process that involves extracting information from log files and modeling actions through process mining. In this sense, the results demonstrate that the number of visualizations, the time spent and the time of abandonment of the video are variables that have great capacity to offer useful information about the students' learning. This demonstrates that evaluating the educational resource through the analysis of its actions can contribute substantially to the educational area, benefiting the treatment of issues such as the identification of bottlenecks in the learning process and the anticipation of problems, especially in distance education. The results obtained with the use of Process Mining in data provided greater clarity about student behavior during video lessons, providing the necessary guidance for the actions to be taken by teachers and content producers. In view of this, the work contributes to the improvement of key aspects of video learning from a multidisciplinary approach, directly helping educators and managers to promote a more complete educational training, based on better understood resources.Acesso AbertoVideoaulasComportamentoMineração de processosmasterThesisCNPQ::CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO::SISTEMAS DE COMPUTACAO