Machine LearningYEAR 2026

Student Placement Prediction

Predicting undergraduate career placement outcomes using supervised classification pipelines.

TECHNOLOGY STACK

PythonPandasScikit-LearnSeabornData Visualization

THE CHALLENGE

Academic institutions lack proactive diagnostic tools to identify students in need of career preparation support before the final placement season begins.

METHODOLOGY & ARCHITECTURE

Applied exploratory data analysis (EDA) to uncover primary predictive factors (CGPA, technical internships, soft skills, backlogs), followed by feature standardization and balanced sampling.

MEASURABLE RESULTS & FINDINGS

Attained an 88.5% testing accuracy and 0.91 ROC-AUC score, identifying CGPA thresholds and internship exposure as top variance drivers.