Student Risk Identification Model Using Random Forest Algorithm
European Journal of Molecular & Clinical Medicine,
2020, Volume 7, Issue 8, Pages 3017-3024
AbstractThe main aim of this work is address the issues and predict students who fail or not complete their online graduation course within stipulated time. Training of the existing machine learning (ML) model is done from existence of data from previous course. This manuscript finds the solution for efficient learning when we don’t have learning data from previous years for a particular course (i.e., for the new course introduced which has no history). To address the problem mentioned the proposed work builds a machine learning model which uses data from newly introduced course. For this the proposed model uses newly introduced course data of already submitted task, Hence the model induces imbalanced data issues. For addressing this issue, this work presents a Random Forest (RF) classification algorithm. By the results obtained by experiments conducted we see that a significant outcome is attained by proposed model compared to existing ML models.
- Article View: 212
- PDF Download: 332