|
报告题目：From Location to Intention: Context-aware Sensing for Mobile Computing
报告人：韩劲松
学术简历：
Jinsong Han is an associate professor at the Department of Computer Science and Technology, Xi’an Jiaotong University. He received his Ph.D. of Computer Science from Hong Kong University of Science and Technology in 2007. His work focuses on wireless network, mobile computing, and security. He was the winner of Hong Kong ICT Awards - Best Innovation & Research Award (Silver Award). He has published over 30 innovative works in highly ranked journals (TON, TMC, TKDE) and top-tier conference proceedings (ACM CCS, Ubicomp, Sensys, and IEEE INFOCOM, ICNP, ICDCS, etc.)。
报告摘要：
Today’s mobile computing often requires sensing contextual information, including environmental conditions and human activities, to provide better services. A wide variety of media, including images, radio frequency, acoustic, ultrasound, magnetic, etc., can be utilized for sensing environment contexts and recognizing subject activities, regardless of whether the subject wears devices.
In this talk, I will present different approaches to sense concrete contextual information ranging from location to intention. In particular, I will present a novel Radio Frequency Identification (RFID) based device-free intrusion detection scheme, which leverages a newly discovered phenomenon called critical state caused by interference among passive tags. Then, I will discuss the possibility of using RF signals to test the internal status of containers and detect abnormal internal changes. I will introduce our nondestructive testing system using backscatter communication, called Echoscope, based on commodity off-the-shelf devices. Furthermore, I will present a gaze based intention detection scheme using smartphones instead of specialized equipment such as Eye trackers.
|
|
报告题目：基于用户交互行为学习的移动应用与服务优化(暂定)
报告人：刘譞哲
学术简历：
博士，北京大学信息科学技术学院副教授。主要研究方向包括系统软件、移动计算、服务计算等。近年来在 TSE、TMC、TOIS、TOIT、WWW、UbiComp、ICSE、OOPSLA等国际期刊和会议上发表学术论文 30 余篇，其中CCF A类论文10余篇；获3次国际会议最佳论文奖和2次最佳论文奖提名；博士学位论文《基于社区的服务组装技术研究》获 2011 年全国优秀博士学位论文提名奖、2010 年中国计算机学会优秀博士论文奖；作为主要完成人获得教育部科技进步一等奖（排名第3）；研制的关键技术以通过专利转化方式应用于央企、政府、互联网等多个领域，取得良好社会经济效益。担任国际服务学会青年科学家论坛中国区主席；国际会议CollaborateCom 2016共同主席；国际会议KDD、ICSE、CIKM、ICDCS、ICSOC、ICWS等程序委员会成员；Journal of Computer Science and Technology特邀编辑（Guest Editor）；CCF学术工作委员会秘书、委员；CCF服务计算专委会委员；中国电子学会青年科学家俱乐部云计算大数据专委会秘书长。
报告摘要：
The prevalence of smart mobile devices has promoted the popularity of mobile applications (a.k.a. apps) and appstores such as Apple Appstore, Google Play, and numerous third-party appstores. The appstore has significantly changed the development and delivery model of software, and even established a new ecosystem. Understanding user behaviors over the appstores can help improve app development, deployment, delivery, revenue, etc. This talk presents our recent study of deriving various insightful knowledge by learning behavioral profiles collected from millions of users, including app management activities (i.e., downloading, updating, and uninstalling apps) and app network usage. Based on the findings as well as derived knowledge, we suggest some practical machine-learning based solutions to address challenges raised by the appstore-centric ecosystem.
|
|
报告题目：领域知识驱动的用户行为数据挖掘方法及应用
报告人：刘淇
学术简历：
博士，中国科学技术大学计算机学院副教授。主要研究数据挖掘与知识发现、机器学习方法及其应用，着重于针对用户行为数据（如消费数据、社交数据、教育数据等）的建模和应用研究。在重要国际学术会议和期刊共发表论文60余篇，2011年获得数据挖掘领域顶级国际会议之一IEEE ICDM的最佳研究论文奖，还获中科院院长特别奖、KSEM 2013最佳论文奖以及SDM 2015 最佳论文候选奖、中科院优博等重要学术奖励，入选了中科院青年创新促进会。主持了多项国家、省部级以及与知名公司（如微软、腾讯、科大讯飞）的合作研究项目。担任了CCF大数据学术会议(BigData)2015-2017的宣传主席、是包括IJCAI、KDD、WWW、AAAI、ICDM、CIKM等国际会议的程序委员会委员以及国际期刊TKDE、TKDD、TC、TSMC-C、TIST等的审稿人、是FCS青年AE。
报告摘要：
在当今大数据时代，社会个体在诸多领域的行为轨迹能够被大量的记录和保存，如消费行为数据、社交行为数据、教育相关的行为数据等，这些社会化行为数据是学习用户行为模式和了解社会运行规律的重要信息来源，有着迫切的研究价值和应用需求。然而，应用场景的多样性、领域特征的复杂性等，使得传统的数据挖掘结果难以有效满足不同场景的实际需求，在已有方法基础上，必须发展新的理论框架和处理算法。为此，本报告将以各领域记录的用户行为数据为研究对象，以用户理解为核心，从数据挖掘方法与领域知识联合建模的角度，介绍我们近期在用户行为数据挖掘方法及应用方面的一些探索工作。
|
