cloudburst-app's Introduction

Cloudburst Detection System

The Cloudburst Detection System is an innovative Android app designed to predict and mitigate the impact of cloudbursts using machine learning models and real-time weather data. This app aims to save lives and protect property by providing accurate and timely cloudburst predictions.

Key Features

Real-Time Data Integration: Integrates real-time weather data from the OpenMeteo API to provide users with up-to-the-minute information, critical for cloudburst prediction.
Machine Learning Models: Utilizes advanced machine learning models trained on meteorological datasets to enhance the accuracy of cloudburst predictions.
User-Centric Design: Features a user-friendly interface tailored for mobile devices, with functionalities such as push notifications and location-based alerts to enhance user engagement.
Ease of Accessibility: Easily accessible with a single tap on a smartphone, making it convenient for users to check weather updates on the go.

Model Selection

1. Random Forest Classifier

Accuracy: 91.756%
Benefits:
- Strong at capturing non-linear patterns in meteorological data.
- Robust performance even with noisy data.

2. Gradient Boosting Classifier

Accuracy: 89.134%
Benefits:
- Effective in handling class imbalance, crucial for rare cloudburst events.
- Provides valuable insights into the importance of meteorological parameters.

3. Convolutional Neural Network (CNN)

Accuracy: 82.542%
Benefits:
- Specialized in handling spatial meteorological data, such as grid-based information.
- Automatically learns hierarchical features, accommodating diverse scales.

4. Linear Regression

Benefits:
- Simple and interpretable model for understanding linear relationships.
- Offers insights into the significance of individual meteorological parameters.