Ntambara-Rukaka-Steven/.github

NTAMBARA Rukaka Steven is an Information Technology professional based in Rwanda, with a specialty spanning Web Testing, Data Visualization, and Analytics. # Student-Dropout-Prediction-University-of-Kigali # AUTHOR — NTAMBARA RUKAKA STEVEN # SUPERVISOR — DR MUSABE JEAN BOSCO This notebook provides a comprehensive exploratory data analysis of the Student Dropout dataset, exploring factors that contribute to student dropout rates. Dataset Characteristics: 10,000 student records 19 features (12 numeric, 7 categorical) Binary target: Dropout (0=No, 1=Yes) 23.54% dropout rate The dataset contains student data from University of Kigali Main Campus (and comparable institutional records) used to predict `Dropout` status, addressing the complex reasons behind student attrition. It includes 10,000 samples, focusing on scenarios that help predict student dropout risk. Attributes include age, gender, family income, academic performance, attendance, stress index, and behavioral factors. Student Dropout Prediction — University of Kigali Big Data Analytics for Early Dropout Risk Detection & Intervention --- ## 📌 Overview **University of Kigali (UoK)** — like most higher-education institutions — loses a meaningful share of enrolled students to dropout every academic year. Today, that loss is only visible *after the fact*: a student stops appearing on attendance sheets, stops submitting assignments, and is eventually withdrawn — with no unified, real-time view of *who* is at risk and *why*, until it is too late to intervene. This project delivers: 1. A **structured Big Data analytics pipeline** — cleaning, imputation, outlier treatment, and feature engineering — applied to a 10,000-record student dataset (`student_dropout_dataset_v3.csv`). 2. **Exploratory data analysis (EDA)** uncovering the demographic, academic, and behavioral patterns behind dropout at UoK-scale institutions. 3. A **prediction layer** (Logistic Regression, Decision Tree, Random Forest, SVM) that flags students at risk of dropout early enough for academic/student-affairs staff to intervene — before withdrawal is finalised. 4. A f …