学术报告

报告题目：Understanding Users from TheirUse of Smartphone Applications

地点：电信学院西一楼第一会议室（344）

时间：2017年6月23日星期五10:00—10:40

报告人：Sha Zhao (赵莎)

报告人简介：

Sha Zhao earned the Ph.D. degree in computer science from ZhejiangUniversity in June 2017.She visited the Human-Computer Interaction Institute at Carnegie MellonUniversity as a visiting PhD student from September 2015 to September 2016.Herresearch interests include ubiquitous computing, mobile sensing, and datamining. Zhao received the Best Paper Award of ACMUbiComp’16.

摘要：

The increasing prevalence of smartphonesin our everyday life has opened up a window into the lives of smartphone users.Smartphones serve a wide variety of functions, and users can install and usemobile applications to achieve nearly every imaginable purpose. Since asmartphone is linked to an individual user, its applications can effectivelycapture large amounts of a user’s personal information. This information couldbe used to improve mobile devices, applications and services in various ways,such as personalized recommendation and adaptation, targeted advertising, anddesigning application interfaces according to users’ interests and preferences,so as to improve the mobile user experience. In this talk, I will show how tomine user attributes from app lists on smartphones, and how to discover usergroups from smartphone app usage behaviors. Two large-scale real-world datasetsconsisting of over 100,000 users were used for evaluation.

欢迎教师和学生参加

计算机系

2017年6月20日

学术报告

报告题目: DeepLearning of Graphs with Diagonal Ngram Convolutional Neural Networks

地点：电信学院西一楼第一会议室（344）

时间：2017年6月23日星期五10:40—11:10

报告人：罗智凌

摘要:

ConvolutionalNeural Network (CNN) has gained attractions in image analytics and speechrecognition in recent years. However, employing CNN for classification ofgraphs remains to be challenging. This talk presents the Ngram graph-blockbased convolutional neural network model for classification of graphs. Ourresults show that the Ngram approach outperforms existing methods with highaccuracy and comparable performance.

个人简介:

Zhiling (Bruce) Luo receivedhis B.S. and Ph.D. degree in Computer Science from Zhejiang University in 2012and 2017, respectively. He was the visiting scholar of Georgia Institute ofTechnology, US, in 2016.

Hisresearch interests include service computing, machine learning and data mining.His works are published on IEEE Trans on Service Computing, Trans on Knowledgeand Data Engineering, and Enterprise Information System.

欢迎教师和学生参加

计算机系

2017年6月20日