Ronghui Guo

Ronghui Guo

B.S. from Shandong University, M.S. from Tianjin University. Research: graph neural networks, anomaly detection, and large language models.

Publications

NAAST-GNN: Neighborhood Adaptive Aggregation and Spectral Tuning for Graph Anomaly Detection

Published:

A novel GNN model addressing heterophily in graph anomaly detection through neighborhood adaptive aggregation in the spatial domain and spectral tuning in the spectral domain.

Ronghui Guo, Xiaowang Zhang, Zhizhi Yu, Minghui Zou, Sai Zhang, and Zhiyong Feng. (2025). "NAAST-GNN: Neighborhood Adaptive Aggregation and Spectral Tuning for Graph Anomaly Detection." Proceedings of the 34th International Joint Conference on Artificial Intelligence (IJCAI '25), pp. 2847–2855.
Download Paper

Empowering Agile-Based Generative Software Development through Human-AI Teamwork

Published:

An agile-based generative software development framework (AgileGen) using Gherkin-based acceptance criteria and human-AI teamwork to ensure semantic consistency between requirements and code.

Sai Zhang, Zhenchang Xing, Ronghui Guo, Fangzhou Xu, Lei Chen, Zhaoyuan Zhang, Xiaowang Zhang, Zhiyong Feng, and Zhiqiang Zhuang. (2025). "Empowering Agile-Based Generative Software Development through Human-AI Teamwork." ACM Transactions on Software Engineering and Methodology (TOSEM), Vol. 34, No. 6, Article 156, pp. 1–46.
Download Paper

MG-GNN: Enhancing GNNs for Anomaly Detection via Minority Class Sample Generation

Published:

A method that generates minority class samples in the hidden space to address class imbalance in graph anomaly detection, using BWGNN backbone and SMOTE-based interpolation.

Ronghui Guo, Minghui Zou, Sai Zhang, Xiaowang Zhang, and Zhiyong Feng. (2024). "MG-GNN: Enhancing GNNs for Anomaly Detection via Minority Class Sample Generation." ISWC 2024 Posters, Demos and Industry Tracks, CEUR Workshop Proceedings, Vol. 3828.
Download Paper

Graph Local Homophily Network for Anomaly Detection

Published:

A novel graph anomaly detection method using local homophily as a metric to guide multi-frequency information fusion, revealing the camouflage phenomenon of anomalous nodes.

Ronghui Guo, Minghui Zou, Sai Zhang, Xiaowang Zhang, Zhizhi Yu, and Zhiyong Feng. (2024). "Graph Local Homophily Network for Anomaly Detection." Proceedings of the 33rd ACM International Conference on Information and Knowledge Management (CIKM '24), pp. 706–716.
Download Paper

Effective Hybrid Graph and Hypergraph Convolution Network for Collaborative Filtering

Published:

An interpretable hybrid framework combining graph and hypergraph convolution networks for collaborative filtering, with optimized dense information flow.

Xunkai Li, Ronghui Guo, Jianwen Chen, Youpeng Hu, Meixia Qu, and Bin Jiang. (2022). "Effective Hybrid Graph and Hypergraph Convolution Network for Collaborative Filtering." Neural Computing and Applications (NCA), Vol. 35, No. 3, pp. 2633–2646.
Download Paper