Home

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Ma, Xiuli

Assistant Professor

Research Interests: Data mining

Office Phone: 86-10-6275 4911

Email: xlma@pku.edu.cn

Ma, Xiuli is an assistant professor in the Department of Intelligence Science, School of EECS since 2005. She obtained her B.Sc. and M.S. from Hebei University in 1993 and 1996, and Ph.D. from Peking University in 2003 respectively. She had been visiting scholar of Department of Computer Science, University of Illinois at Urbana-Champaign during Sep. 2014 to Jan. 2016, and Department of Computer Science, Hong Kong University of Science and Technology from Mar. to May 2003. Her research interests include data mining, network-based mining, sensory data mining and bioinformatics.

Dr. Ma has published more than 40 research papers, and some of them are published in top-tier conferences and journals, such as KDD, AAAI, ICDE, ICDM in data mining domain and RECOMB in bioinformatics. She has served as reviewer of various international conferences and journals including SIGMOD, VLDB, ICDE, KDD, EDBT and TKDE. She was awarded the scholarship by China Scholarship Council (CSC) in 2013 (No. 201308110276).

Dr. Ma has been granted by National Natural Science Foundation of China, 973 programs, IBM SUR, etc. His research achievements are summarized as follows:

1) Continuous, Online, Large-Scale Monitoring and Analysis: The major goal is to facilitate continuous, online mining and analysis in large-scale, real-time monitoring networks. She proposes to discover and track summarization patterns within the sensory data and to reveal the intricacies of the system or environment in a continuous, online way. Also, she proposes to summarize massive streams based on the correlation among them by a few typical representative trends, thus reducing the monitoring scale effectively. Then, based on these two orthogonal kinds of summarization, outlier detection can be independent of pre-specified pattern.

2) Complexes Detection in Biological Networks through Diversi?ed Dense Subgraphs Mining: Protein-protein interaction (PPI) networks enable us to explore biological processes and cellular components at multiple resolutions. Proteins densely interact with each other, forming large molecular machines or cellular building blocks. Identification of such densely interconnected clusters or protein complexes from PPI networks enables people to obtain a better understanding of the hierarchy and organization of biological processes and cellular components. She proposes to formulate the problem of detecting complexes from biological networks into finding a diverse set of dense subgraphs. By comparing with existing algorithms on human and yeast PPI networks, she demonstrates that her method can detect more putative protein complexes and achieves better prediction accuracy.

3) Multi-Dimensional Data-oriented Automatic Navigation and Knowledge Discovery: Her aim is to investigate deeply the theory and methods on automated navigation of multidimensional analysis, and knowledge discovery in multi-dimensional data. She proposes several automated navigation mechanisms such as intelligent rollup, intelligent drilldown, intelligent slice, and virtual cubes. The goal is to automate much of the manual effort spent in analysis, navigate the analysis to the interesting part. Knowledge discovery includes several mechanisms such as summarization of likelihood, exceptional slices mining, and maximal correlated member clusters mining.