Gestão de Conhecimento de uma Instituição de Educação

• Main Author: Cabeda, José Eduardo Barreira
• Publication Date: 2018
• Format: Master thesis
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10400.22/14055
• topic: Machine Learning; Business Intelligence; Educational Data Mining; Learning Analytics; R Language

With the emergence of the Internet, bigger computation capacity and the fall of costs in storing information came a flood of new data and the capacity to analyze it. Many areas have developed and progressed due to this, such as fintech and online advertising, but others have only now started to develop. One of these is the educational data mining and learning analytics which has enormous potential to empower the teachers and students to be more successful. To this end, the present work analyzes data relating the students of Software Engineer Department (DEI) of Instituto Superior de Engenharia do Porto (ISEP) with the goal of improving students’ success rate, by facilitating the comprehension of the information, detect patterns in the data and predict future events directly related to the students’ behavior. This work has proposed and implemented an architecture which presents the results to the user through a data portal. This portal has been divided into two components. The first agglomerates the analysis of the data while the second presents models built with Random forests, Decision Trees and Naive Bayes to predict the students’ behavior.