The Zombie Detector using ML is a machine learning project designed to predict the likelihood of individuals turning into zombies during a simulated zombie outbreak. By leveraging demographic data, location information, and available supplies, this project aims to raise awareness about emergency preparedness and enable informed decision-making in crisis scenarios.Using logistic regression as the core algorithm, the project has developed a trained machine learning model. Hyperparameter tuning techniques have been applied to optimize the model's performance, which has been evaluated using precision, recall, and F1-score metrics. The project emphasizes robust data preprocessing techniques, such as analyzing categorical features through contingency tables and Chi-square tests, and exploring numerical features using T-tests and density plot visualizations. To address imbalanced data, the SMOTE technique has been employed for oversampling the dataset.
The project features a user-friendly Flask web application that enables real-time predictions. Users can input their data, and the application leverages the predictive model to provide accurate and timely results. The aim is to empower individuals to take proactive measures and make informed decisions for their safety during a potential zombie outbreak.The associated GitHub repository hosts the complete source code, including data preprocessing scripts, model training, evaluation code, and the Flask web application. It serves as a valuable resource for developers, researchers, and enthusiasts interested in the intersection of machine learning and emergency preparedness. The repository includes a detailed README file with instructions for environment setup, application execution, and result interpretation.By merging machine learning with emergency preparedness, the Zombie Detector using ML project serves as a practical tool to raise awareness, enhance decision-making, and support emergency responders in crisis management. It opens avenues for collaboration, contributions, and further advancements to improve the project's functionality and real-world applicability.
DEMONSTRATION LINK : https://drive.google.com/file/d/1pVwpAD-U6852KqOn7uMYhcrQE6G1JQlG/view?usp=sharing