Poseidon is a python-based application that leverages software defined networks (SDN) to acquire and then feed network traffic to a number of machine learning techniques. The machine learning algorithms classify and predict the type of device.

using deep learning to classify the encrypted network traffic

Machine learning plugins for network traffic

Self-attentive deep learning method for online traffic classification and its interpretability (CN21 & NetAI20)

基于IDS 2018数据集使用LightGBM和XGBoost实现DDoS流量分类 sFlow RT&Mininet流量采集与分类 杭电综合项目实践

🛜→🖼️ Replication of the model set forth in "FlowPic: Encrypted Internet Traffic Classification is as Easy as Image Recognition" by Tal Shapira and Yuval Shavitt

Network traffic classifier based on Apache Spark and MLlib

An ML-based classification method to classify the traffic and intrusions in a network which improves the overall performance of the network system.

Network traffic Simulation & Classification system

Supporting page for the manuscript titled, "Incremental Federated Learning for Traffic Flow Classification in Heterogeneous Data Scenarios."

🐳📡🐶 Generate network communication data for target tasks in diverse network conditions.

Code resource (R) for the paper: https://onlinelibrary.wiley.com/doi/abs/10.1002/cpe.4181

Code Repo for Leveraging Federated Learning and XAI for Private and Lightweight Edge Training In Network Traffic Classification

A Network Traffic Processing and Feature Extraction Tool

This is a code repository for a paper with title "Mitigating Adversarial Attacks in Federated Learning Based Network Traffic Classification Applications using Secure Hierarchical Remote Attestation and Adaptive Aggregation Framework"

Honours Year Project on: Network Traffic Classification of 5G networks for Improvement of QoS

Supporting page for the manuscript titled, "AutoFlow: An Autoencoder-based Approach for IP Flow Record Compression with Minimal Impact on Traffic Classification."

AI-powered Network Traffic Classifier developed for the IBM Z Datathon.

A model trained for network traffic prediction. problem and data provided by ETRI(Electronics and Telecommunications Research Institute), Korea, Republic of.