Recently, the paper “Understanding Diverse Usage Patterns from Large-Scale Appstore-Service Profiles” was selected as the spotlight paper of the April 2018 issue of IEEE Transactions on Software Engineer (TSE), one of the top academic journals in software engineering.

This work set up a situation model for mobile Internet applications from three different aspects, which are application information, human-computer interaction, and system runtime log respectively. Combined with situational data from tens of millions of Android users, this paper utilizes a variety of machine learning technologies to deeply mine the behavioral patterns of users using mobile Internet applications and services, which reveals a variety of important rules, such as power law distribution of interaction frequency, service community structure, chaining calls, and so on. Further, this work raised several new and important research problems related to the development, testing, release, and maintaining of mobile Internet application which have not been studied yet. This expands new research domains for the research of software engineering.

Professor Xuanzhe Liu, Ph.D. student Huoran Li, and Ph.D. student Xuan Lu are the main authors of this work. Professor Tao Xie from UIUC, Professor Qiaozhu Mei from University of Michigan, Academician Hong Mei, and the co-founder of Wandoujia Feng Feng are the co-authors of this work. According to investigations, this paper has also been the second paper, the first author of which is from Mainland China, that is featured as the spotlight paper since TSE was founded 44 years ago. In fact, ever since this paper was able to online access on May 2017, it has received great attention from researchers in the related domain. It has been downloaded hundreds of times in a short period, which indicates that the research on Internet Software of Peking University has reached the international leading level.