期刊名称：Anais dos Workshops do Congresso Brasileiro de Informática na Educação
印刷版ISSN：2316-8889
出版年度：2018
卷号：7
期号：1
页码：389
DOI：10.5753/cbie.wcbie.2018.389
语种：Portuguese
出版社：Anais dos Workshops do Congresso Brasileiro de Informática na Educação
摘要：The purpose of this paper is to show how the Correspondence Analysis technique can be used to enhance the Exploratory Data Analysis on the evasion phenomenon for a set of educational data with a majority of the categorical type. To this, the independence calculation was implemented from the chi-square, and the heatmap and perceptual map were generated with the indexes generated on the socioeconomic data of students of courses of IFRN. As a result, the relations of attraction and repulsion between socioeconomic attributes and drop-out are presented, and the profile of the student most vulnerable to evasion is drawn. Some characteristics are: students who have already failed at least once, who do not live with their parents, elementary school in public school, the schooling of the lower financial officer.
