A Comparative Study of Predicting Student’s Performance by use of Data Mining Techniques

Aysha Ashraf, Muhammad Gufran Khan

2018 · 23 citations · 16 references

DOIFull text

Open access

Concepts

Abstract

Educational systems need innovative ways to improve quality of education to achieve the best results and decrease the failure rate.  Educational Data Mining (EDM) has boomed in the educational systems recently as it enables to analyze and predict student performance so that measures can be taken in advance. Due to lack of prediction accuracy, improper attribute analysis, and insufficient datasets, the educational systems are facing difficulties and challenges exist to effectively benefit from EDM. In order to improve the prediction process, a thorough study of literature and selection of the best prediction technique is very important. The main objective of this paper is to present a comparative study of various recently used data mining techniques, classification algorithms, their impact on datasets as well as the prediction attribute’s result in a clear and concise way. The paper also identifies the best attributes that will help in predicting the student performance in an efficient way.

References

16