Li, Hongyan
professor
Research Interests: Database, data mining
Office Phone: 86-10-6275 4911
Email: lihy@cis.pku.edu.cn
Li, Hongyan is a professor in the Department of Intelligence Science, School of EECS. She obtained her B.Eng. and Ph.D. from Northwestern Polytechnical University in 1993 and 1999 respectively. Her research interests include data management, business process control, data mining and knowledge discovery, with a focus on big data management, large-scale data analysis and business process anomalies detecting.
Dr. Li has published more than 100 research papers, an academic monograph, a teaching material and a translation. She has seven patents for the invention of the country. She has served in Database Specialized Committee of China Computer Federation and Cloud Computing & SaaS Specialized Committee of China Institute of Communications.
Dr. Li also has strong collaboration with Industry including projects and patents. Her research has been supported by NSFC, China Mobile, Singapore PeridotHealth Systems Pte. Ltd., etc. Her research achievements are summarized as follows:
1) Data management techniques on Cloud data: Cloud storage has served as the basic platform for a large number of applications in different fields, which bring great challenge to researchers in data management. She proposed a layered and configurable storage model named “Dynamic Table” to describe mixed dataset with different characteristics and to cope with hybrid workload of OLTP and OLAP-like queries, a 4-Valued Logic based extended relation algebra to distinguish the semantic of absent values in the sparse data, a scalable and elastic secondary index to improve Ad-hoc queries, and a declarative Structured Query Language to turn the No SQL Cloud data management system to Not Only SQL system.
2) Data mining techniques on Big data: With the advent of big data era, how to acquire and derive the abundant information and embedded knowledge form big data have always been a hot spot as well as challenge in the research field of data mining. She proposed new data mining techniques to deal with the characteristics of big data, including large volume, heterogeneity, rich structure, correlation and uncertainty. These methods have applied successfully in China Launch Vehicle, China Mobile, Singapore National Healthcare Group, etc.
3) Business process control: In recent years, Web Information System (WIS) has evolved from the pure information exhibition system to the business system supporting application logic. This evolution demands WIS to support not only data storage and retrieval but also the awareness of process. She focused on the ad hoc process awareness in WIS, and proposed a datalog based language to describe the process model which emphasizes the knowledge associated with the process and enriches the logic structure, a verification method to verify the consistency between process model and data model, a data model based anomalies detection method for business process model with parallel structure, etc. Some methods have applied successfully in CCTV asset management system and the integrated medical information system of Singapore Ministry of National Defense for military and logistics management.
