What is it about?
We analyze several variables to predict student performance measured by final grades obtained in an online higher education course of the first semester. We applied to prediction models: Genetic programming and statistical regression. The first obtained better accuracy results. With this work, we provide with an accurate model to predict student performance as well as a methodology to build prediction models.
Photo by Maik Jonietz on Unsplash
Why is it important?
This paper proves a better accuracy of genetic programming as well as the suitability of this type of techniques over the classical statistical regression. This has possible implications in dealing with the prediction of human-centered problems or when a human impact is a strong factor The methodology followed is systematic and allows the reproduction of the empirical analysis. This is one of the major contributions to science.
Read the Original
This page is a summary of: Prediction of Online Students Performance by Means of Genetic Programming, Applied Artificial Intelligence, September 2018, Taylor & Francis, DOI: 10.1080/08839514.2018.1508839.
You can read the full text:
The following have contributed to this page