|
Title
|
A Method To Predict Academic Performance With Combination Of Xgboost And Random Forest
|
|
Type
|
Presentation
|
|
Keywords
|
academic,performance,prediction,machine learning, random forest, XGboost, PCA
|
|
Abstract
|
Educational institutions collect vast amounts of student data, offering opportunities to improve academic achievement prediction. However, identifying the most influential factors and optimal algorithms remains a challenge. This research addresses this gap by proposing a novel method using XGBoost and Random Forest algorithms to identify key factors impacting prediction accuracy. Educational Data Mining (EDM) offers a powerful approach to analyze this data and understand student behavior. While EDM adoption lags behind other fields, this study tackles the unique challenges associated with educational data, including its sequential nature and temporally distributed training data. Our approach aims to provide valuable insights for educational administrators seeking to develop data-driven strategies to enhance student success
|
|
Researchers
|
Arash Khosravi (Second Researcher), Ahmad Azarnik (First Researcher)
|