Prediction of University Dropout Using Machine Learning
Autor: | Manuel Ayala-Chauvin, Aracelly Fernanda Núñez-Naranjo, Genís Riba-Sanmartí |
---|---|
Rok vydání: | 2021 |
Předmět: |
Higher education
Computer science business.industry k-means clustering 020206 networking & telecommunications 02 engineering and technology Support vector machine ComputingMilieux_COMPUTERSANDEDUCATION 0202 electrical engineering electronic engineering information engineering Mathematics education 020201 artificial intelligence & image processing Cluster analysis business Dropout (neural networks) |
Zdroj: | Advances in Intelligent Systems and Computing ISBN: 9783030682842 |
Popis: | University dropout is a complex issue that affects all higher education institutions worldwide. This phenomenon is shown by the high proportion of students that never finish their university training, with the associated economic and social costs. The challenge for higher education institutions is to design and improve policies to increase student retention, specially within the first years. This study uses data mining to find patterns and student clustering that help explaining university dropout. The data for the analysis was gathered from the students that signed up on two admission periods of the Universidad Tecnologica Indoamerica of Ambato, Ecuador. A k-means algorithm is used to classify and define the performance patterns, and predictions for new students are made using a support-vector machine (SVM) model. The results allow institutions and the faculty to focus in high risk groups during the first terms and amend their future learning behaviour. To sum up, this study presents a models to explain and predict university dropout, and to design actions to reduce it. |
Databáze: | OpenAIRE |
Externí odkaz: |