Liu, Xuanzhe is currently an associate professor in the School of Computer Science. He obtained his Ph.D. from Peking University in 2009. His major research interests include Web/mobile computing systems, applications, and services, user interactions/behaviors, most from a data-driven perspective.
Dr. Liu has published more than 50 referred papers, and most of them are published in premiere conferences such as WWW, ICSE, OOPSLA, UbiComp, and IMC, and high-impact journals such as TSE, TMC, TOIS, TOIT, and TSC. He has served in the Technical Program Committee of various international conferences including KDD, ICSE, ICDCS, CIKM, ICSOC, and ICWS, and as the General Co-Chair of CollaborateCom 2016 and the Track Chair of ICWS 2016 and MS 2016. He was awarded by the Nominee of National Outstanding Ph.D. Dissertation (2011) and the CCF Outstanding Ph.D. Dissertation (2010).
Dr. Liu has various research projects including NSFC, 973 programs, 863 project, etc. His research achievements are summarized as follows:
1) Performance measurement and optimization for mobile computing systems. Mobile devices such as smartphones, tablet computers, and wearables, establish the major user channel to access the Internet, and the Quality-of-Experiences (QoE) optimization becomes a challenging issue. He proposed new performance measurement and modeling techniques to anatomize factors that can probably affect the user-perceived QoE, including page load time, energy, and data traffic. He developed various system-level supports such as on-demand edge/cloud offloading, collaborative cache, and middlebox-based transfer prioritization, which have been demonstrated to significantly improve the performance of both Android apps and mobile Web apps.
2) Web APIs and mashups. Currently, various applications are delivered via the Web in a service-oriented fashion, i.e., in form of APIs. One major research topic in services computing and Web engineering is to dynamically integrate these applications and services and create new value-added applications, namely services mashups. He proposed new Web component model, designed domain-specific languages and API search algorithms, and developed end-user programming environments along with automated client-side code generation techniques to facilitate the interactive composition of services mashups.
3) User behavior/Interaction Analytics. Understanding the user behaviors when interacting computer systems is always important. Collaborated with various industrial Internet Service Providers such as Tencent, Wandoujia, and Kika, he proposed new analytic models and machine-learning techniques to infer the insights of “in-app” user interaction/behaviors from millions real-world users at scale. The derived knowledge can improve the user experiences from various aspects such as user adoption estimation, app/content recommendation, and so on.
