Predicting Students’ Dropout and Academic Success Using R

This project is created to determine factors and predict Academic status of students using machine learning techniques in R. By analyzing the dataset, we aim to identify key predictors of student dropout, enrollment, and graduation, ultimately supporting educational institutions in making data-driven decisions to improve student outcomes.

The objectives are as follows:

1. Build a predictive model to classify students into one of the three categories: Dropout, Enrolled, or Graduate, based on their academic and demographic features.
2. Analyze the importance of different variables in predicting student outcomes, providing insights into which factors most influence academic success or dropout.
3. Evaluate and compare the performance of various machine learning algorithms using appropriate metrics (e.g., accuracy, F1-score).
4. Determine the statistical significance of observed differences in model performance.

Data

Data used in this project is retrieved from UCI Machine Learning Repository titled "Predict Students' Dropout and Academic Success". The dataset comprises of 37 variables, from which are 36 features with a target variable of:

• Dropout
• Enrolled
• Graduate

Dataset Link:

• https://archive.ics.uci.edu/dataset/697/predict+students+dropout+and+academic+success

Note: Variable names & descriptions are also provided by the authors