Hao Do Phuc

I'm a PhD student in Computer Science at SUT (Federal State Budgetary Educational Institution of Higher Education "Saint Petersburg State University of Telecommunications named after prof. M.A. Bonch-Bruevich") currently working on machine learning applications in IoT security and network traffic analysis. My research focuses on developing novel approaches for detecting malicious traffic and botnet attacks in IoT networks using federated learning and deep learning techniques. I specialize in applying advanced machine learning models like Graph Neural Networks, Convolutional Neural Networks, and Transformers to network security challenges. My recent work has explored horizontal federated learning approaches for IoT malware detection, designing efficient feature extraction methods for attack classification, and leveraging pretrained models for various ML tasks.

Prior to my current research, I worked on human action recognition in video sequences using hybrid deep learning architectures combining FFT and EfficientNet models. I've published several papers in international conferences like ICACT, RIVF, and ICUMT, collaborating with researchers from multiple institutions to advance the fields of network security and machine learning. My current interests span federated learning, IoT security, computer vision, and natural language processing.

Publications

A Horizontal Federated Learning Approach to IoT Malware Traffic Detection: An Empirical Evaluation with N-BaIoT Dataset

P. H. Do, D. Le, V. Vishnevsky, A. Berezkin, R. Kirichek

International Conference on Advanced Communication Technology 2024

Unveiling the Power of Pretrained Models for Neural Machine Translation in Vietnamese Language: A Comparative Analysis

Unveiling the Power of Pretrained Models for Neural Machine Translation in Vietnamese Language: A Comparative Analysis

P. H. Do, Pham Van Quan, D. Le, T. Dinh, Nang Hung Van Nguyen, Minh Tuan Pham

Conference on Research, Innovation and Vision for the Future in Computing & Communication Technologies 2023

Leveraging FFT and Hybrid EfficientNet for Enhanced Action Recognition in Video Sequences

Leveraging FFT and Hybrid EfficientNet for Enhanced Action Recognition in Video Sequences

Nguyen Nang Hung Van, P. H. Do, Van Nam Hoang, Aleksandr Borodko, Tran Duc Le

Symposium on Information and Communication Technology 2023

Graph Neural Networks for Traffic Classification in Satellite Communication Channels: A Comparative Analysis

P. H. Do, T. Le, A. Berezkin, R. Kirichek

Proceedings of Telecommunication Universities 2023

A Horizontal Federated-Learning Model for Detecting Abnormal Traffic Generated by Malware in IoT Networks

A Horizontal Federated-Learning Model for Detecting Abnormal Traffic Generated by Malware in IoT Networks

P. H. Do, D. Le, V. Vishnevsky, A. Berezkin, R. Kirichek

International Conference on Advanced Communication Technology 2023

An Efficient Feature Extraction Method for Attack Classification in IoT Networks

P. H. Do, T. Dinh, D. Le, V. Pham, L. Myrova, R. Kirichek

International Conference on Ultra Modern Telecommunications 2021

Classifying IoT Botnet Attacks with Kolmogorov-Arnold Networks: A Comparative Analysis of Architectural Variations

P. H. Do, Tran Duc Le, T. Dinh, Van Dai Pham

IEEE Access 2025