SciELO - Scientific Electronic Library Online

 
vol.8 número5Concepciones sobre el Aprendizaje de Estudiantes de Pedagogía de la Universidad de Magallanes y Docentes en Ejercicio en la Educación Básica de la ciudad de Punta Arenas, Chile índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados

Revista

Articulo

Indicadores

Links relacionados

  • En proceso de indezaciónCitado por Google
  • No hay articulos similaresSimilares en SciELO
  • En proceso de indezaciónSimilares en Google

Compartir


Formación universitaria

versión On-line ISSN 0718-5006

Resumen

ECKERT, Karina B  y  SUENAGA, Roberto. Analysis of Attrition-Retention of College Students Using Classification Technique in Data Mining. Form. Univ. [online]. 2015, vol.8, n.5, pp.03-12. ISSN 0718-5006.  http://dx.doi.org/10.4067/S0718-50062015000500002.

Academic information is analyzed to identify the factors that have more impact on desertion of students of Computer Science Engineering of the University Gastón Dachary in Argentina, by applying data mining techniques. The data source comes from the information provided by the student when they entered the university (personal and educational background) and information generated during the studies. Data are selected and analyzed using different criteria for the representation and application of classification algorithms such as decision trees, bayesian networks and rules. Influential variables on desertion are identified: passed courses, number and grades of courses, origin and age of student when he/she entered the university. Through this process it was possible to identify several variables that characterize the cases of desertion and its relation with academic achievement, especially during the first year of study.

Palabras clave : college retention; data mining; classification algorithms; university studies.

        · resumen en Español     · texto en Español     · Español ( pdf )

 

Creative Commons License Todo el contenido de esta revista, excepto dónde está identificado, está bajo una Licencia Creative Commons