https://github.com/adityajhakumar/earthquake-detection-system
https://github.com/adityajhakumar/earthquake-detection-system
Last synced: about 1 month ago
JSON representation
- Host: GitHub
- URL: https://github.com/adityajhakumar/earthquake-detection-system
- Owner: adityajhakumar
- License: apache-2.0
- Created: 2024-09-14T06:54:41.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2024-09-14T07:20:12.000Z (over 1 year ago)
- Last Synced: 2024-09-14T18:30:57.886Z (over 1 year ago)
- Language: Jupyter Notebook
- Size: 147 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Earthquake Detection System 🌍🚨
## Welcome! 👋
Welcome to the Earthquake Detection System project! 🎉 This initiative aims to leverage machine learning to detect earthquakes using data from gyroscopes and accelerometers. Whether you're a researcher, developer, or just curious about the intersection of technology and natural disaster detection, this project is for you. Dive in to explore how we've built and fine-tuned a system to keep our communities safe from the tremors of Earth!
## What’s Inside? 🧩
### 1. Data Generation 🛠️
We’ve crafted a dataset that's both realistic and synthetic to train and test our models. Here’s what you’ll find:
- **Accelerometer Data (Ax, Ay, Az)**: Tracks acceleration in the X, Y, and Z axes.
- **Gyroscope Data (Gx, Gy, Gz)**: Measures rotational changes in the X, Y, and Z axes.
- **Labels**: Simple binary indicators - `0` for normal conditions and `1` for earthquakes.
#### How We Generate Data
- **Normal Conditions**: Smooth, steady data with minimal noise.
- **Earthquake Conditions**: Features significant spikes and variations to mimic real earthquakes.
### 2. Model Training 🤖
We’ve experimented with several powerful models to find the best fit for our task:
#### Random Forest 🌳
- **What It Is**: A collection of decision trees working together to make predictions.
- **Why It’s Great**: Handles diverse data well and avoids overfitting.
#### XGBoost 🏆
- **What It Is**: An advanced boosting technique that’s both fast and flexible.
- **Why It’s Great**: Delivers exceptional performance and scales effortlessly.
#### Support Vector Machine (SVM) 🚀
- **What It Is**: Finds the best boundary between different classes.
- **Why It’s Great**: Perfect for high-dimensional data and complex decision boundaries.
#### Long Short-Term Memory (LSTM) ⏳
- **What It Is**: A type of neural network designed to understand sequences.
- **Why It’s Great**: Excellent for capturing patterns over time, ideal for time-series data.
### 3. Evaluating Our Models 📊
We don’t just train models - we thoroughly evaluate them using:
- **Accuracy**: How often our model gets it right.
- **Precision**: How many of the predicted positives are truly positive.
- **Recall**: How many of the actual positives our model managed to catch.
- **F1 Score**: A balanced measure combining precision and recall.
### 4. Ensuring Robustness 🔍
We use cross-validation to make sure our models are reliable:
- **K-Fold Cross-Validation**: Splits data into multiple folds to ensure robust training.
- **Stratified K-Fold Cross-Validation**: Maintains class balance in each fold for more accurate results.
### 5. Visualizing Performance 📈
See how well our models perform through visualizations:
- **Training and Validation Accuracy**: Tracks accuracy improvements over epochs.
- **Training and Validation Loss**: Shows how loss decreases during training.
## Flowchart
Here’s a simplified flowchart of our process:
1. **Data Generation**
- Create synthetic and realistic sensor data for both normal and earthquake conditions.
2. **Data Preprocessing**
- Scale features and split data into training and test sets.
3. **Model Training**
- Train various models (Random Forest, XGBoost, SVM, LSTM) using the training data.
4. **Model Evaluation**
- Evaluate models on test data, focusing on key metrics like Accuracy, Precision, Recall, and F1 Score.
5. **Cross-Validation**
- Apply K-Fold and Stratified K-Fold cross-validation to ensure our models are robust and generalizable.
6. **Visualization**
- Plot training and validation metrics to monitor and understand model performance.
## Getting Started 🚀
### Installation 🛠️
1. **Clone the Repository**
```bash
git clone https://github.com/adityajhakumar/earthquake-detection.git
cd earthquake-detection
```
2. **Install Dependencies**
Set up a virtual environment and install required packages:
```bash
python -m venv venv
source venv/bin/activate # On Windows use `venv\Scripts\activate`
pip install -r requirements.txt
```
3. **Requirements**
Our `requirements.txt` includes:
```plaintext
numpy
pandas
scikit-learn
keras
matplotlib
```
### Usage
- **Generate Data**: Run `generate_data.py` to create your dataset:
```bash
python generate_data.py
```
- **Train and Evaluate Models**: Use `train_evaluate_models.py`:
```bash
python train_evaluate_models.py
```
- **Cross-Validation**: Perform cross-validation with `cross_validation.py`:
```bash
python cross_validation.py
```
## Evaluation Metrics
We use the following metrics to evaluate our models:
- **Accuracy**: Measures overall correctness.
- **Precision**: Evaluates the accuracy of positive predictions.
- **Recall**: Assesses how well positive cases are detected.
- **F1 Score**: Balances precision and recall.
- 
- 
## License 📝
This project is licensed under the Apache License 2.0. Check out the [LICENSE](LICENSE) file for details.
## Contact 📧
Have questions or feedback? Feel free to reach out to me, [Aditya Kumar Jha](mailto:your.email@example.com). I’m here to help!