Best student paper awarded to Dr Shirui Pan

Dr Shirui Pan's award-winning research, Open-World Graph Learning, has received the Best Student Paper Award at the IEEE International Conference on Data Mining (ICDM-20)

The ICDM is a premier research conference in data mining and provides an international forum for academics to present original projects and exchange innovative and practical research.

The conference covers all aspects of data mining, including algorithms, software, systems, and applications. ICDM is a highly competitive conference in data mining with an acceptance rate of 9.8% for full papers this year.

Dr Shirui Pan is a Lecturer in the Department of Data Science & AI in the Faculty of Information Technology at Monash University. Dr Pan's Open-World Graph Learning research pioneers a new direction in deep learning for graph data analysis.

Traditionally, machine learning on graph data learning is carried out in a closed-world setting where the number of classes and their training data are provided to help train models, and the learning goal is to correctly classify unlabelled nodes into classes already known. In reality, due to limited labelling capability and dynamic evolving of networks, some nodes in the networks may not belong to any existing/seen classes, and therefore cannot be correctly classified by closed-world learning algorithms.

This research is a collaborative project with Dr Shirui Pan from Monash University and PhD student Man Wu and Professor Xingquan Zhu from Florida Atlantic University. It proposes a new open-world graph learning paradigm, where the graph machine learning goal is to not only classify nodes belonging to seen classes into correct groups but also classify nodes not belonging to existing classes to an unseen class.

This research can be beneficial in many applications where deep graph analysis plays a key role, including bioassay activity prediction, credit assessment, traffic flow forecasting, and drug discovery.

Of the recent win, Dr Pan says, "We’ve been researching deep graph learning since 2016 and to receive a best paper award for this research is a strong recognition of our work in this field.

Early this year, Dr Pan also had an additional three research papers accepted by the Thirty-fourth Conference on Neural Information Processing Systems (NeurIPS), a top conference in machine learning.

To learn more about Dr Pan’s research, please visit here.