Machine-learning

Student Dropout Prediction

This repository contains a machine learning project that predicts whether a student is likely to graduate or drop out using supervised learning techniques in R. The project compares multiple classification algorithms to identify the model with the best predictive performance.

Project Overview

Student dropout is a significant challenge for higher education institutions. Early identification of students at risk allows universities to provide timely support and improve student retention. This project applies machine learning methods to predict student outcomes based on demographic, academic, and socioeconomic characteristics.

Dataset

The project uses the Predict Students' Dropout and Academic Success dataset from the UCI Machine Learning Repository.

• Source: UCI Machine Learning Repository
• Observations: 4,424 students
• Predictors: 36 student-related variables
• Target Variable: Graduate or Dropout (binary classification)

Machine Learning Models

The following classification models were implemented and compared:

• K-Nearest Neighbors (KNN)
• Decision Tree
• Naive Bayes
• Support Vector Machine (Linear Kernel)
• Support Vector Machine (Radial Kernel)

Data Preprocessing

The project includes:

• Data cleaning
• Removal of the "Enrolled" class
• Feature selection
• Conversion of categorical variables to factors
• 80/20 train-test split
• Model training and evaluation

Model Evaluation

Models were evaluated using:

• Accuracy
• Confusion Matrix
• Sensitivity
• Specificity
• Balanced Accuracy

Among the models tested, the Linear Support Vector Machine (SVM) achieved the highest overall prediction accuracy, outperforming KNN, Decision Tree, Naive Bayes, and Radial SVM. Studnt dropout predictions.Rmd.txt

Tools & Technologies

• R
• R Markdown
• caret
• tidyverse
• e1071
• tree
• MASS
• corrplot

Skills Demonstrated

• Machine Learning
• Predictive Modeling
• Classification
• Data Preprocessing
• Exploratory Data Analysis (EDA)
• Model Evaluation
• Statistical Learning
• Data Mining
• R Programming

Repository Contents

Student_Dropout_Prediction.Rmd
README.md

Future Improvements

• Hyperparameter tuning
• Random Forest implementation
• XGBoost comparison
• Feature importance analysis
• Cross-validation
• Interactive visualizations

Author

Esther Koffi

Bachelor of Science in Mathematics & Statistics

Data Analyst | Business Intelligence | Machine Learning