Predicting Students’ Performance in University Courses: A Case Study and Tool in KSU Mathematics Department

Ghada Badr, Afnan Algobail, Hanadi Almutairi, Manal Almutery

Procedia Computer Science · 2016 · 67 citations · 8 references

DOIFull text

Open access

Concepts

TL;DR

Educational data mining uses institutional data to uncover factors behind high dropout rates and poor academic performance. The study aims to build a predictive model that estimates students’ performance in a programming course from prior grades to reduce dropout rates. The model uses an association‑rules classifier to predict programming course performance. The model enables instructors to tailor teaching strategies by accurately estimating students’ learning abilities.

Abstract

Educational data mining is a growing field that uses the data obtained from educational information systems to discover knowledge and find answers to questions and problems concerning the education system. High dropout rates and poor academic performance among students are examples of the most common issues that affect the reputation of an educational institution. Students’ academic records can be analyzed to explore the factors behind these phenomena. This paper discusses the building of a model to predict the performance of students in a programming course based on their grades in courses in other subjects. A classification based on an association rules algorithm is used to build a classifier to help evaluate the student's performance in the programming course. This model aims to reduce dropout levels by helping student predict their likelihood of success in a course before they enroll in it. In addition, course instructors will be able to enhance student performance in the course by better estimating their abilities to learn the subject matter and adjusting their teaching strategies and methods.

References

8