Academic Performance Predicting Model based on Machine Learning and Keller's Motivation Measure

This article investigates a model for predicting the academic performance of university students using Machine Learning techniques based on the level of motivation achieved with the implementation of the ARCS instructional model and the use of a technological tool called Arduino Science Journal used...

Full description

Saved in:
Bibliographic Details
Published in:Proceedings - International Conference of the Chilean Computer Science Society pp. 1 - 7
Main Authors: Laurens, Luis, Garcia, Ruber Hernandez
Format: Conference Proceeding
Language:English
Published: IEEE 21.11.2022
Subjects:
Academic performance
ARCS
IMMS
Instruments
Kinematics
Machine learning
Predicting model
Prediction algorithms
Predictive models
Python
Support vector machines
ISSN:2691-0632
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This article investigates a model for predicting the academic performance of university students using Machine Learning techniques based on the level of motivation achieved with the implementation of the ARCS instructional model and the use of a technological tool called Arduino Science Journal used for learning Topics related to the kinematics of bodies. Analready validated methodology focused on motivation was implemented,which was quantified through the Instructional Material Motivational Survey (IMMS) instrument, which was applied toa group of 36 students of the Kinematics and Dynamics subject from a Civil Industrial Engineering career. Machine learning techniques were used to predict academic performance based on regression algorithms. The results show that Confidence was the IMMS dimension with the best prediction results. At the sametime, the Support Vector Regression algorithm achieves the lowest mean absolute error in the estimated academic performance. This research provides a prediction model of academic performance through emotional variables of the students, showing the potential to act as an early warning system, helping teachers to manage students' academic performance, and allowing students to self assess their performance.
ISSN:2691-0632
DOI:10.1109/SCCC57464.2022.10000282