You can access the live demo of the app hosted on Streamlit's Community Cloud using the link: https://flowscope.streamlit.app/
FlowScope is an advanced web application designed to improve decision-making and prediction accuracy for web traffic analysis using the HybridFlow Forecast Model. The model combines several state-of-the-art time series forecasting techniques, including ARIMA, SARIMA, ETS, and LSTM, to deliver robust and accurate predictions. This project is particularly focused on the analysis and forecasting of web traffic, allowing businesses to make data-driven decisions to optimize their operations.
- 🔗 HybridFlow Forecast Model: Integrates multiple forecasting models (ARIMA, SARIMA, ETS, LSTM) to enhance prediction accuracy.
- 📁 Customizable Inputs: Allows users to upload their dataset, specify relevant columns, and configure model parameters.
- 📊 Interactive Dashboard: Provides an intuitive and user-friendly interface for visualizing raw data, model predictions, and evaluation metrics.
- 📈 Model Evaluation: Offers comprehensive performance metrics (MAE, MSE, RMSE, MAPE) for both testing and future predictions.
- 🔮 Future Predictions: Generates and visualizes future time series predictions based on the trained models.
- 📤 Export Functionality: Enables users to export the prediction results to a CSV file.
To run FlowScope locally, follow these steps:
- Clone the repository:
git clone https://github.com/your-username/flowscope.git
- Navigate to the project directory:
cd flowscope - Install the required dependencies:
pip install -r requirements.txt
- Run the Streamlit application:
streamlit run app.py
- Open the application in your web browser (usually http://localhost:8501).
- 📥 Data Upload: Users upload their time-stamped dataset in CSV format.
- 🧹 Data Cleaning: Missing values are filled with the mean of the column, and the data is sorted by the timestamp.
- 🔢 ARIMA: Suitable for short-term forecasting of stationary data.
- 📅 SARIMA: Ideal for capturing seasonal patterns and long-term trends.
- 🔄 ETS: Models error, trend, and seasonality without requiring differencing.
- 🧠 LSTM: Captures long-term dependencies and non-linear relationships in the data.
- 📊 Model Evaluation: Calculates MAE, MSE, RMSE, and MAPE for each model on the testing data.
- 🔮 Future Predictions: Generates predictions for future time steps and evaluates model performance.
- 📉 Visualizations: Displays actual vs. predicted values and future predictions using interactive charts.
This project is licensed under the MIT License - see the LICENSE file for details.
This project was developed as part of a research initiative at BVRIT, Narsapur. Special thanks to our mentors and colleagues for their support and guidance.