Caicedo Castro, Isaac BernardoPinto Lucas, José David2022-03-032022-03-032022-02-28https://repositorio.unicordoba.edu.co/handle/ucordoba/4869In this work, I aim at finding the functional dependency between the engineering students’ performance in computer programming courses and competencies that they achieved during the secondary. The context of this work consists of engineering students with a major in computer science at University of Córdoba, in Colombia. I assume the test known as Saber 11, which is used to evaluate students at secondary level throughout Colombia, measures the extent that secondary students have attained competencies in four areas as follows: i) Mathematics, ii) critical reading, iii) English, and iv) social science and citizen competencies. In fact, the Saber 11 test outcomes are considered to study applications for undergraduate careers at Colombian universities, such as, e.g., the University of Córdoba. Attaching the results of the saber 11 test together with the results obtained by the programming courses (computational logic, programming I, II, II) by means of computational learning algorithms predicts the future performance of the new students of the systems engineering career, the computer programming courses are assumed to be fundamental for the rest of the courses of the curriculum, which require students to have skills to solve problems through the writing of computer programs. In conclusion, the main contribution of this work is an intelligence system which predicts the students’ average grade of all computer programming courses given their outcomes achieved from the Saber 11 test, in the context of the B.Sc. in Engineering with a major in computer science at University of Córdoba. From the evaluation carried out on this systems, it reaches a RMSE and R2 about 0.29 and 0.98, respectively.RESUMEN ........................................................................................................................... 12ABSTRACT ......................................................................................................................... 131. INTRODUCCIÓN ........................................................................................................ 141.1 PLANTEAMIENTO DEL PROBLEMA ................................................................. 162. OBJETIVOS ................................................................................................................. 232.1 OBJETIVOS GENERAL: .............................................................................. 232.2 OBJETIVOS ESPECÍFICOS: ........................................................................ 233. REVISIÓN BIBLIOGRÁFICA .................................................................................... 243.1 APRENDIZAJE COMPUTACIONAL. ......................................................... 243.2 REGRESIÓN LINEAL .................................................................................. 243.3 VALIDACIÓN CRUZADA. .......................................................................... 243.4 MÉTODOS DE MACHINE LEARNING IMPLEMENTADOS-SUPPORT VECTOR MACHINE REGRESION. ...................................................................... 253.5 PROCESO DE GAUSS (GAUSSIAN PROCESS). ...................................... 253.6 RED NURONAL ............................................................................................ 263.7 JUSTIFICACIÓN DEL PROBLEMA. .......................................................... 263.8 ALCANCES ................................................................................................... 283.9 CONTRIBUCIÓN DE LA TESIS .................................................................. 293.10 ESTRUCTURA DE LA TESIS .................................................................. 304. ESTADO DEL ARTE .................................................................................................. 315. MATERIALES Y MÉTODOS ..................................................................................... 355.1 MATERIALES ............................................................................................... 355.2 MÉTODOS ......................................................................................................... 366. RESULTADOS Y DISCUSIONES ............................................................................. 436.1 RESULTADOS: KFOLD VALIDATIÓN. ....................................................... 446.2 COMPARATIVA DE LOS RESULTADOS. .................................................... 566.3 PRESENTACIÓN DE RESULTADOS PRUEBAS TTEST (COMPARATIVA DE MEDIAS DE R2 OBTENIDAS) ....................................................................... 576.4 PRUEBA INDIVIDUAL T-TEST-1 .................................................................. 596.5 PRUEBA T-TEST-2 ........................................................................................... 606.6 PRUEBA T-TEST-3 ........................................................................................... 616.7 CUARTILES ...................................................................................................... 626.8 PROMEDIO Y MEDIDAS CENTRALES. ....................................................... 636.9 MEDIDAS DE DISPERSIÓN. .......................................................................... 657. CONCLUSIONES ........................................................................................................ 668. RECOMENDACIONES .............................................................................................. 679. BIBLIOGRAFÍA .......................................................................................................... 68ANEXOS .............................................................................................................................. 69application/pdfspaCopyright Universidad de Córdoba, 2022Trabajo de grado - Pregradoinfo:eu-repo/semantics/openAccessAtribución-NoComercial-SinDerivadas 4.0 Internacional (CC BY-NC-ND 4.0)Algoritmos, predicción, resultados.Competences, SAT, dependency.