Welcome to Siwei’s homepage

I am an Assistant Professor at the School of Natural and Computing Science, University of Aberdeen, UK. Before this position, I was a Postdoctoral researcher at MBZUAI working with Dr. Shangsong Liang. In addition, I work with Dr. Zaiqiao Meng and we lead the BioKR2 (Bio Knowledge Representation and Reasoning) Lab. The ultimate goal of BioKR2 is to inspire bioinformatics research by using meaningful representation learning methods.

I used to be a PhD student of Terrier team, supervised by Prof. Iadh Ounis and Prof. Craig Macdonald.

I am one of the co-founders and main contributors of the Beta-Recsys project, an open-source project for Building, Evaluating and Tuning Automated Recommender Systems.

News

Nov, 2023.          Our paper entitled A Social-aware Gaussian Pre-trained Model for Effective Cold-start Recommendation is accepted by Information Processing and Management.

Nov, 2023.          Our paper entitled Messages Are Never Propagated Alone: Collaborative Hypergraph Neural Network for Time-series Forecasting is accepted by TPAMI.

Aug, 2023.          Our paper entitled Sequence pre-training-based graph neural network for predicting lncRNA-miRNA associations is accepted by Briefings in Bioinformatics (IF=9.5).

Aug, 2023.          Our paper entitled HMCDA: a novel method based on the heterogeneous graph neural network and metapath for circRNA-disease associations prediction is accepted by BMC Bioinformatics (IF=3.327).

May, 2023.          Our paper entitled Leveraging Relational Graph Neural Network for Transductive Model Ensemble is accepted by SIGKDD (CCF A).

Nov, 2022.          I gave a talk on Graph Neural Networks at UN Big Data Hackathon 2022. The powerpoint is available here.

Oct, 2022.         I start my Postdoc at MBZUAI (Abu Dhabi) with Dr Shangsong Liang.

Oct. 2022.         Our paper entitled Graph neural pre-training for recommendation with side information is accepted by ACM TOIS (CCF A).

Mar. 2022.          Our paper entitled An MLP-based Algorithm for Efficient Contrastive Graph Recommendations is accepted by ACM SIGIR (CCF A).