Xiaojie Guo, a doctoral student in the Department of Information Sciences and Technology (IST), along with several faculty members in IST and the Department of Electrical and Computer Engineering, won the Best Paper Award at the IEEE International Conference on Data Mining (ICDM 2019) earlier this month in Beijing, China. ICDM is a top conference in data mining, with an acceptance rate lower than 10 percent for full research papers.
Her paper, “Deep Multi-attributed Graph Translation with Node-Edge Co-evolution,” explores the transformation between arbitrary graph-structured data from a source domain to a targeted domain by using new techniques in deep-graph neural networks and interpretable machine learning.
Liang Zhao, Guo’s advisor and one of the paper’s co-authors, says this research can benefit many real-world applications that involve modeling and understanding the transformation among graph-structured data in areas such as neuroscience, social networks, biochemistry, cybersecurity, and hardware design. “I regard this award as a big encouragement for our work on this topic.”
Guo agrees, “In the realm of deep-graph learning, this paper represents our work under a new research direction called ‘deep-graph translation,’ which we have been exploring since last year. I am really happy to receive this important award, which is not only an honor but also a strong recognition of this research direction.”