文章基本信息

标题：Predição do desempenho de Matemática e Suas Tecnologias do ENEM utilizando técnicas de Mineração De Dados
本地全文：下载
作者：Rafael Damiani Alves ; Cristian Cechinel ; Emanuel Queiroga 等
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：469
DOI：10.5753/cbie.wcbie.2018.469
语种：Portuguese
出版社：Anais dos Workshops do Congresso Brasileiro de Informática na Educação
摘要：The objective of this research is to find patterns and generate a predictive model of the performance indicator of the marks of the Mathematics test and its Technologies of the secondary schools, through the open data referring to the National High School Examination (ENEM) of 2015. The objective in question is based on data from the Program for International Student Assessment (PISA) of the year 2015, which demonstrate a worrying scenario of low performance of Brazilian elementary school students. The various techniques of data mining have been used to discover patterns that allow improvement in many areas. In this context, experiments were performed through educational data mining (EDM), where the data were categorized, to obtain a better result in the application of the algorithms. The predicted class was the school average that was categorized as: low, medium, high. The final models were trained and tested using the Naive Bayes and J48 algorithms. These algorithms were used through the WEKA software package.