Tutorial I: Indexing and Querying Metric Spaces

Time: 9:30-12:00, August 1, 2019

Location: 2F Yuegui Hall

Abstract:

With the rapid advances in Internet, wireless, and other techniques, there is an exploration of big data with three typical characteristics, i.e., volume, velocity, and variety. Volume denotes that the amount of data is extremely large; velocity represents that the speed of data input and output is extremely high; and variety indicates that the range of data types and sources is extremely wide. A lot of studies have been done on volume and velocity, but not as much has been reported on variety. To handle the variety of data, metric space can be used. Metric space is a general model that can represent any type of data as long as its distance metric satisfies the triangle-inequality. Thus, based on metric space model, we can develop a unified solution to process all various data types. In this tutorial, the speakers systematically present indexing and query processing technologies in metric spaces, including metric index structures, metric query processing, and metric space mining.

Biography. Yunjun Gao is now a full professor at the College of Computer Science, Zhejiang University (ZJU), Hangzhou, China. He received the Ph.D. degree in computer science from ZJU in 2008. Prior to joining ZJU in 2010, he was a research assistant or postdoctoral research fellow (scientist) or visiting professor/scholar in the City University of Hong Kong (CityU, China), Singapore Management University (SMU, Singapore), Simon Fraser University (SFU, Canada), and Nanyang Technological University (NTU, Singapore), respectively. His primary research areas are Database, Big Data Management and Analytics, and AI Interaction with DB Technology. He has published more than 130 papers on several premium/leading journals including TODS, VLDBJ, TKDE, TOIS, TMC, TFS, TITS, and DKE, and various prestigious international conferences such as SIGMOD, VLDB, ICDE, SIGIR, AAAI, EDBT, and DASFAA. He is a senior member of the CCF; a member of the ACM and the IEEE; a/an (young) editorial board member or associate editor of JCST, DAPD, and FCS; and a guest editor of WWWJ, IJDSN, and DSE. He is/was a referee/reviewer of several top/important journals such as TODS, VLDBJ, TKDE, TMC, TKDD, Information Sciences, GeoInformatica, etc.; and he is/has serving/served as an organization committee (e.g., PC co-chairs, workshop co-chairs, publication chair, publicity co-chair, etc.) or a program committee member for various conferences such as SIGMOD, VLDB, ICDE, CIKM, SIGSPATIAL GIS, DASFAA, etc. He won the Best Paper Award of APWeb-WAIM 2018, 2017 CCF Outstanding Doctoral Dissertation Award (Supervisor), the First Prize of the Ministry of Education Science and Technology Progress Award (2016), the Winner of National Outstanding Young Scientist Fund Project (2015), the Nomination of the Best Paper Award of SIGMOD 2015, One of the Best Papers of ICDE 2015, and The First Prize of Zhejiang Province Science and Technology Award First (2011).

Biography. Lu Chen is an Assistant Professor in Aalborg University, Denmark. She received the Ph.D. degree in computer science from Zhejiang University, China, in 2016, and then worked as a research fellow in Nanyang Technological University from Oct. 2016 to Sep. 2017. Her research concerns data management and data-intensive systems, and its focus is on metric data management. She has published more than 30 papers on several top/important database conferences (e.g., SIGMOD, VLDB, ICDE, SIGIR) and journals (e.g., VLDBJ, TKDE, Information Sciences). Her paper was selected as one of best papers in ICDE 2015, her thesis is selected as one of the excellent PHD theses by CCF, and her paper won APWeb-WAIM 2018 best paper award. She was also a publication chair of WISE 2017, a PHD colloquium co-chair of MDM 2019, a publicity co-chair of IEEE ICBK 2019, and a guest editor of WWWJ and DSE.

Biography. Keyu Yang received the BS degree in computer science from Zhejiang University of Technology, China, in 2016. He is currently working toward the Ph.D. degree in the College of Computer Science, Zhejiang University, Hangzhou, China. His research interests include metric data management and machine learning interaction with data management technologies.

Tutorial II: Hashing for Image Retrieval

Time: 14:00-15:00, August 1, 2019

Location: 2F Jingui Hall

Abstract:

Hashing has shown its efficiency and effectiveness in facilitating large-scale multimedia applications. In this tutorial, we will introduce the motivation and advantages of applying hashing in the task of image retrieval. In the first part, a number of classic hashing methods will be discussed, including PCA Hashing, Spectral Hashing, Supervised Hashing with Kernels, and Supervised Discrete Hashing. In this second part, we will present our recent work of zero-shot hashing, robust hashing, and scalable hashing, which are designed for different retrieval scenarios.

Biography. Zi Huang is an Associate Professor (Reader) and ARC Future Fellow in School of ITEE, The University of Queensland. She received her BSc degree from Department of Computer Science, Tsinghua University, China, and her PhD in Computer Science from School of ITEE, The University of Queensland in 2001 and 2007 respectively. Dr. Huang's research interests mainly include multimedia indexing and search, social data analysis and knowledge discovery. She has published 100+ papers in prestigious venues, and is currently an Associate Editor of The VLDB Journal. Dr. Huang has received 2016 Chris Wallace Award from Computing Research and Education (CORE) Australasia for a notable breakthrough or a contribution of particular significance in Computer Science, and Women in Technology (WiT) Infotech Research Award 2014, Queensland. She was also a recipient of the Excellence in Higher Degree by Research Supervision Award, University of Queensland, 2018.

Tutorial III: Cohesive Subgraph Search on Large Graphs: Concepts and Algorithms

Time: 14:00-17:00, August 1, 2019

Location: 2F Yuegui Hall

Abstract:

The problem of finding cohesive subgraphs from a large graph is a fundamental problem in graph data mining which has attracted much attention in recent years due to a large number of practical applications. In this tutorial, I will first introduce two widely-used cohesive subgraph models: k-core and k-truss. Then, I will present a peeling algorithm and an h-index iteration algorithm to efficiently compute the k-core and k-truss decomposition on large graphs. Finally, I will introduce some generalized k-core concepts and algorithms on attributed and temporal graphs.

Biography. Rong-Hua Li received the Ph.D. degree from the Chinese University of Hong Kong in 2013. He is currently an associate Professor at Beijing Institute of Technology (BIT), Beijing, China. Before joining BIT in 2018, he was an assistant professor at Shenzhen University. His research interests include graph data management and mining, social network analysis, graph computation systems, and graph-based machine learning.

Biography. Hongchao Qin is currently a Ph.D. Candidate in Northeastern University, China. He received the B.S. degree in mathematics and M.E. degree in computer science from Northeastern University in 2013 and 2015, respectively. His current research interests include social network analysis and data-driven graph mining.