University of Southern Queensland (USQ)张吉教授学术报告 12月25日下午
发布时间: 2017-12-21 访问次数: 172

报告题目:A Parallelized Graph Mining Approach for Efficient Fraudulent Phone Call Detection
时间:2017-12-25 (星期一) 15:30 ~ 17:00
地点:成功楼602报告厅
主办:数学与信息学院, 福建省网络安全与密码技术重点实验室
主讲:张吉University of Southern Queensland (USQ), Australia


报告人简介:Prof. Ji Zhang is currently a A/Prof, Reader (Equivalent to Professor in the US universities) in Computer Science, a PhD supervisor and the leader of the research group on data engineering and big data analytics at the University of Southern Queensland (USQ), Australia. Prof. Zhang is an IEEE Senior Member, Australian Endeavour Fellow, Queensland Fellow and Izaak Walton Killam Fellow (Canada) and the recipient of the USQ Research Excellence Award in 2011. He served the Principal Advisor for Research in Division of ICT Services, USQ from 2010-2013. He was a Research Fellow in CSIRO ICT Center, Australia from 2008 to 2009. He received his degree of Ph.D. from the Faculty of Computer Science at Dalhousie University, Canada in 2008.  Prof. Zhang's research interests include knowledge discovery and data mining (KDD), Big Data analytics, bioinformatics, information privacy and security, and health informatics. He has published over 110 papers, some appearing in top-tier international journals including IEEE Transactions on Dependable and Secure Computing (TDSC), Information Sciences, Knowledge-based Systems, WWW Journal, Bioinformatics, Knowledge and Information Systems (KAIS), Soft Computing, Journal of Database Management and Journal of Intelligent Information Systems (JIIS) and international conferences such as VLDB, ACM CIKM, ACM SIGKDD, IEEE ICDE, IEEE ICDM, WWW, DASFAA, PAKDD, DEXA and DaWak. He has also authored 1 monograph and 6 book chapters. Prof. Zhang is holding a patent with CSIRO, Australia.


报告摘要:In recent years, fraud is becoming more rampant internationally with the development of modern technology and global communication. Due to the rapid growth in the volume of call logs, the task of fraudulent phone call detection is confronted with Big Data issues in real-world implementations.  In this talk, I will discuss a highly- efficient parallelized graph-mining based fraudulent phone call detection framework, namely PFrauDetector, which is able to automatically label fraudulent phone numbers with a “fraud” tag, a crucial prerequisite fordistinguishing fraudulent phone call numbers from the normal ones. PFrauDetector generates smaller, more manageable subnetworks from the original graph and performs a parallelized weighted HITS algorithm for significant speed acceleration in the graph learning module.  We conduct a comprehensive experimental study based on a real dataset collected through an anti-fraud mobile application, Whoscall. The results demonstrate a significantly improved efficiency of our approach compared to FrauDetector and superior performance against other major classifier-based methods.