HR Analytics for Candidate Selection: Predicting Hiring Decisions with Supervised Learning (SVM, Random Forest, Logistic Regression)

This project presents the design, training, and evaluation of predictive models for optimizing the hiring process in a large industrial company (Ternium). Using a real anonymized dataset of 5,853 applicants and 56 attributes, the goal was to predict whether a candidate would be hired (“Yes/No”) through machine learning classification techniques.

A complete data science pipeline was implemented, including data cleaning, feature engineering, model training, hyperparameter tuning, and evaluation through accuracy, F1-score, and ROC-AUC. Among all tested models, Support Vector Machine (SVM), Random Forest, and Logistic Regression achieved the highest accuracy (≈97%), precision (≈0.79), and sensitivity (≈0.90), outperforming simpler baselines.

Additionally, a Flask-based web application was developed to operationalize the model, enabling HR personnel to input candidate data and receive automated recommendations. The system demonstrates how Human Resources Analytics (HR Analytics) can reduce bias, improve efficiency, and support evidence-based decision-making in recruitment.