DisasterSense:

Advanced Neural Network for Predicting Natural Catastrophes and Impacts

DisasterSense is an advanced C++ project designed to analyze natural disaster risks and predict their potential impacts using artificial neural networks (ANNs) such as multilayer perceptrons (MLPs). The project integrates libraries like Eigen for matrix operations, JSON for data manipulation, and SFML for graphical visualizations.

Features

• Neural Network (MLP): Custom implementation of a multilayer perceptron for predictive analysis.
• Matrix Computation: Powered by the Eigen library for efficient mathematical operations.
• JSON Data Handling: Supports structured data input and output through the JSON for Modern C++ library.
• Visualization: Real-time graphical representation of predictions and risk factors using SFML.

Project Structure

GeoImpactAI/
├── build/          # Compiled binaries
├── data/           # Data files (e.g., JSON, CSV)
├── include/        # Header files (.h)
├── libs/           # External libraries (Eigen, JSON, SFML)
├── src/            # Source files (.cpp)
├── README.md       # Project documentation

Requirements

• C++ Compiler (G++ or equivalent)
• Visual Studio Code with extensions:
  • C/C++ (Microsoft)
  • Code Runner
  • CMake Tools (optional)
• Libraries:
  • Eigen (for matrix computations)
  • JSON for Modern C++ (for JSON handling)
  • SFML (for graphics)

Installation

1. Clone the repository or create a project directory:

   git clone https://github.com/DisasterSense/DisasterSense.git
   cd DisasterSense

2. Set up external libraries:

   • Download Eigen and place it in libs/eigen/.
   • Download JSON for Modern C++ and place json.hpp in libs/json/.
   • Download SFML and extract it to libs/SFML/.

3. Configure the project in Visual Studio Code:

   • Add include and library paths in tasks.json (see below for configuration).

Configuration

tasks.json Example

{
    "version": "2.0.0",
    "tasks": [
        {
            "label": "Build Project",
            "type": "shell",
            "command": "g++",
            "args": [
                "-I", "${workspaceFolder}/libs/eigen",
                "-I", "${workspaceFolder}/libs/json",
                "-I", "${workspaceFolder}/libs/SFML/include",
                "-L", "${workspaceFolder}/libs/SFML/lib",
                "-lsfml-graphics",
                "-lsfml-window",
                "-lsfml-system",
                "-g",
                "${workspaceFolder}/src/*.cpp",
                "-o",
                "${workspaceFolder}/build/GeoImpactAI"
            ],
            "group": {
                "kind": "build",
                "isDefault": true
            }
        }
    ]
}

launch.json Example

{
    "version": "0.2.0",
    "configurations": [
        {
            "name": "Run GeoImpactAI",
            "type": "cppdbg",
            "request": "launch",
            "program": "${workspaceFolder}/build/GeoImpactAI",
            "args": [],
            "stopAtEntry": false,
            "cwd": "${workspaceFolder}",
            "environment": [],
            "externalConsole": true,
            "MIMode": "gdb",
            "setupCommands": [
                {
                    "description": "Enable pretty-printing for gdb",
                    "text": "-enable-pretty-printing",
                    "ignoreFailures": true
                }
            ],
            "preLaunchTask": "Build Project"
        }
    ]
}

Usage

1. Place your input data in the data/ directory (e.g., data/input.json).
2. Open the project in VS Code.
3. Build the project:
   • Press Ctrl+Shift+B to compile.
4. Run the project:
   • Press F5 to execute the binary.

Example

Here is an example of running the project with sample JSON input:

{
    "region": "Coastal Area",
    "temperature": 28.5,
    "humidity": 75.2
}

The program will predict the risk level and impact severity based on the input data.

Future Enhancements

• Implement additional machine learning models.
• Add support for real-time data streaming.
• Enhance the visualization capabilities with 3D mapping.

License

This project is open-source and available under the MIT License.