https://github.com/murilochianfa/network-traffic-time-series-forecasting

Predict the future of your network using the best time series ML model that fit with your traffic.
https://github.com/murilochianfa/network-traffic-time-series-forecasting

dataset ddos forecasting lstm machine-learning network-traffic prophet sarimax streamlit

Last synced: about 14 hours ago
JSON representation

Predict the future of your network using the best time series ML model that fit with your traffic.

• Host: GitHub
• URL: https://github.com/murilochianfa/network-traffic-time-series-forecasting
• Owner: MuriloChianfa
• License: mit
• Created: 2024-10-27T14:14:09.000Z (3 months ago)
• Default Branch: main
• Last Pushed: 2024-11-09T22:49:57.000Z (2 months ago)
• Last Synced: 2025-01-04T12:55:48.691Z (9 days ago)
• Topics: dataset, ddos, forecasting, lstm, machine-learning, network-traffic, prophet, sarimax, streamlit
• Language: Python
• Homepage:
• Size: 6.05 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
 Network Traffic Time Series Forecasting 


> Predict the future of your network using the best time series ML model that fit with your traffic.

- You can try with your own data builded from raw netflow.

- Fork this project to your Github account.

- This software is created under [MIT License](https://github.com/MuriloChianfa/network-traffic-time-series-forecasting/blob/main/LICENSE)

## Facebook Prophet network traffic forecasting

![Prophet-forecasting](images/prophet-forecasting.png)

## SARIMAX network traffic forecasting

![SARIMAX-forecasting](images/forecasting.png)

## ML model fitting

![SARIMAX](images/banner.png)



# Written paper

> [!IMPORTANT]

>

> The following work and its results are the result of a project presented at the end of the subject of the pattern recognition class of the Master's in Science course Computing at the State University of Londrina (UEL) and does not have the objective of being published as a scientific article.

![preview](images/network-traffic-time-series-forecasting-paper.png)

You can read the [written paper](network-traffic-time-series-forecasting.pdf) here, please notice to the above advice.

# Install instructions

We've some methods to get up and running the application:

  Using pure Python

  ### Dependencies

  - *Python v3.11.10.*

  #### 1° - Clone the project

  ```bash

  git clone [email protected]:MuriloChianfa/network-traffic-time-series-forecasting.git

  cd network-traffic-time-series-forecasting

  ```

  #### 2° - Install dependencies into a new virtual environment

  ```bash

  virtualenv -p python3.11 venv

  . ./venv/bin/activate

  pip install -r requirements.txt

  ```

  #### 3° - Running the application

  ```bash

  streamlit run forecasting/main.py

  ```

  #### 4° - Access the application

  [http://localhost:8501](http://localhost:8501)

  Using Docker Compose

  ### Dependencies

  - *Docker v24.0 or higher.*

  - *Docker Compose v2.13 or higher.*

  - *Your may need nvidia-container-toolkit.*

  #### 1° - Clone the project

  ```bash

  git clone [email protected]:MuriloChianfa/network-traffic-time-series-forecasting.git

  cd network-traffic-time-series-forecasting

  ```

  #### 2° - Running project

  ```bash

  docker compose -f docker-compose.yml up -d

  ```

  #### 3° - Access the application

  [http://localhost](http://localhost)

  Building your own dataset

  #### 1° - Preparing go modules

  ```bash

  cd preprocess

  go mod init network-traffic-time-series-forecasting

  go mod tidy

  go env -w GO111MODULE=on

  go get github.com/phaag/go-nfdump@d2ff6042cb5186ede4064cbd50253ab97a78a89e

  ```

  #### 2° - Running traffic extractor

  ```bash

  go run extract-traffic.go

  ```