Selected Tutorials
1. Self-Supervised Learning for Recommendation: Foundations, Methods and Prospects
Speakers:
- Junliang Yu (The University of Queensland)
- Tong Chen (The University of Queensland)
- Hongzhi Yin (The University of Queensland)
Brief outline of the tutorial:
Recommender systems have become a necessity in this Internet era to offer personalization. However, in contrast to the increasing ease of model building and deployment, the lack of user behavioral data still remains a major pain point for modern recommender systems that constantly compromises recommendation performance. Recently, self-supervised learning (SSL), which can enable training on massive unlabeled data with automatic data annotation, has achieved tremendous success in many fields and been applied to an ever-expanding range of applications including recommendation. Many recent studies have demonstrated that all kinds of recommendation models can be significantly improved through learning with well-designed self-supervised tasks and data augmentations. In this tutorial, we will provide a panorama of the research efforts on self-supervised recommendation. Specifically, the content includes: (1) foundations and overview of self-supervised recommendation; (2) a comprehensive taxonomy of existing self-supervised recommendation methods; (3) how to apply SSL to various recommendation scenarios; (4) Challenges and future research directions.
Speakers Bio
Junliang Yu is a final-year Ph.D. student with the School of Information Technology and Electrical Engineering at the University of Queensland. His research interests include recommender systems, social media analytics, deep learning on graphs, and self-supervised learning. He has 10+ publications on top-tier international venues such as KDD, WWW, ICDM, CIKM, AAAI, SIGIR, VLDBJ, and TKDE. He has been actively providing professional services to many toptier conferences/journals such as AAAI, CIKM, IJCAI, etc. He has rich lecture experience and tutored one relevant course of social media analytics, and also has made oral presentations on multiple top-tier conferences.
Dr. Tong Chen is a lecturer with the Data Science Discipline at The University of Queensland. He received his PhD degree in Computer Science from The University of Queensland in 2020. Dr. Chen’s research interests include data mining, machine learning, business intelligence, recommender systems, and predictive analytics. He has 60+ publications on top-tier international venues such as KDD, SIGIR, ICDE, AAAI, IJCAI, ICDM, WWW, TKDE, IJCAI, TOIS, and CIKM. He has been actively providing professional services to over 20 world-leading international conferences/journals in the fields of data mining, information retrieval and AI. Dr. Chen has ample track records in lecturing, witnessed by his course design and delivery experience in business analytics, full-course teaching experience in social media analytics and database systems, as well as invited talks on cutting-edge recommender systems at the WWW’22 Tutorial, ICDM’20 NeuRec Workshop, Beihang University, and Zhejiang University.
Dr. Hongzhi Yin works as ARC Future Fellow and associate professor with The University of Queensland, Australia. He received his doctoral degree from Peking University in July 2014. His current main research interests include recommender systems, graph embedding and mining, chatbots, social media analytics and mining, edge machine learning, trustworthy machine learning, decentralized and federated learning, and smart healthcare. He has published 220+ papers with Hindex 52, including 22 most highly cited publications in Top 1% (CNCI) venues such as KDD, SIGIR, WWW, WSDM, SIGMOD, VLDB, ICDE, AAAI, TKDE,etc. He has won 6 Best Paper Awards such as Best Paper Award at ICDE 2019, Best Student Paper Award at DASFAA 2020, and Best Paper Award Nomination at ICDM 2018. Dr. Yin has rich lecture experience and taught 5 relevant courses such as information retrieval and web search, data mining, social media analytics, and responsible data science. He was nominated as Most Effective Teacher of EAIT Faculty in The University of Queensland for 2020, 2021 and 2022. He has delivered 12 keynotes, invited talks and tutorials at the top international conferences such as tutorials at WWW 2017, KDD 2017 and WWW 2022.
2. How AI Benefits Knowledge Graph Management? Approaches and Open Problems
Speakers:
- Hongzhi Wang (Harbin Institute of Technology)
- Zhixin Qi (Harbin Institute of Technology)
- Yu Yan (Harbin Institute of Technology)
Brief outline of the tutorial:
Since the knowledge graph is a specific type of graph data, we focus on discuss how AI benefits knowledge graph management in this tutorial. This tutorial aims to highlight the significance of AI for knowledge graph management and pays attention on the existing methods, open problems, and challenges. It consists of three main parts: knowledge graph management overview, physical design tuning for knowledge graph stores, and performance prediction for knowledge graph queries. The knowledge graph management part overviews frequently-used approaches for knowledge graph management. The part of physical design tuning for knowledge graph stores introduces physical design tuning methods for knowledge graph management based on AI techniques. The performance prediction for knowledge graph queries part discusses how AI techniques predict performance for knowledge graph queries. We conclude with a discussion of open problems and challenges for this issue. The tutorial is never presented anywhere else. From this tutorial, audiences could learn the state-of-the-art AI techniques for knowledge graph management, capture the research advances and be inspired to find new ideas in this area.
Speakers Bio
Hongzhi Wang is a professor and PhD supervisor of Harbin Institute of Technology, the secretary general of ACM SIGMOD China, CCF outstanding member. Research fields include big data management and analysis, database, knowledge engineering, and data quality. He was “starring track” visiting professor at MSRA and postdoctoral fellow at University of California, Irvine. Prof. Wang has been PI for more than 10 national or international projects including NSFC key project, NSFC projects and National technical support project, and co-PI for more than 10 national projects include 973 project, 863 project and NSFC key projects. He also serves as a member of ACM Data Science Task Force. He won first natural science prize of Heilongjiang Province, Microsoft Fellowship, IBM PhD Fellowship and Chinese excellent database engineer. His publications include over 300 papers including VLDB, SIGMOD, SIGIR papers, 6 books and 3 book chapters. His PhD thesis was selected to be outstanding PhD dissertation of CCF and Harbin Institute of Technology. He serves as the reviewer of more than 20 international journals including IEEE TKDE and PC member of over 30 top international conferences. His papers were cited more than 4,000 times.
Zhixin Qi is an assistant professor in the School of Transportation Science and Engineering at Harbin Institute of Technology. She received her PhD from the School of Computer Science and Technology at Harbin Institute of Technology in 2022. Her research interests include knowledge graph, AI4DB, and graph data management. She was awarded National Scholarship for PhD students in 2021, National Scholarship for master students in 2017, and National Scholarship for undergraduates in 2014. She has published more than 10 papers in international journals and conferences, including TKDE, KAIS, KBS, Neurocomputing, WWWJ, JCST, CIKM, and DASFAA.
Yu Yan is currently a PhD student in the School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China. She received her master degree from Harbin Institute of Technology in 2021. She committed to the research of database tuning, multi-model database, and database automanagement. She got National Scholarship for master students in 2020. She has published many papers in international conferences and journals, such as Information Sciences, ApWeb and etc.
3. Fusion of Relational and Graph Database Techniques: An Emerging Trend
Speakers:
- Yu Liu (Beijing Jiaotong University)
- Qingsong Guo (North University of China)
- Jiaheng Lu (University of Helsinki)
Brief outline of the tutorial:
In this tutorial, we will give an overview of the recent advances in the fusion of relational and graph database techniques. In particular, we mainly focus on the key operations (i.e., relational join and subgraph matching) for both types of databases, recent query processing techniques to answer relational queries against a graph database and graph queries against a relational database. We will also conduct a brief review of the development history of graph databases and point out promising directions for future research.
Speakers Bio
Yu Liu is a Lecturer at Beijing Jiaotong University, China. His current research interests include key techniques for graph databases and multi-model databases such as scalable graph algorithms and graph-based learning. He has published more than 10 refereed papers in various journals and conferences.
Qingsong Guo is a Lecturer at the School of Computer Science and Technology (School of Data Science), North University of China (NUC). His research interests include multi-model data management and automatic management of big data with deep learning algorithms.
Jiaheng Lu is a Professor at the University of Helsinki. His main research interests lie in big data management and database systems. He has written four books on Hadoop and NoSQL databases, and more than 100 journal and conference papers published in SIGMOD, VLDB, TODS, TKDE, etc.
4. Cyber Attack Investigation Based on Provenance Graph
Speakers:
- Jing Qiu (Guangzhou University)
Brief outline of the tutorial:
Cyberspace attacks have grown in sophistication and stealth, which makes it difficult for traditional security devices to recover the full picture of the attack processes. Endpoint and traffic monitoring devices can be used to collect the log events and alerts from multiple sources and correlate them to evaluate attack behaviors. However, the information in audit logs often lacks an understanding of the complex relationships between alerts and actual intrusion instances and the precision needed for attack investigation on different hosts. The provenance graph-based cyber attack investigation approaches can supply the causation and time-space relationships between events contained in the audit logs, reconstruct the high-level attack scneriao graph with rich semantics, and complete the investigation task of the entire chain and multi-path of complicated attacks. This tutorial introduces the development of the provenance graph-based cyber attack investigation method, analyzes how the powerful causal expression ability of the provenance graph plays a role in the attack investigation task, and presents to audience the entire life cycle of the attack investigation task by summarizing work in the field.
Speakers Bio
Qiu Jing, professor and doctoral supervisor of the Institute of Advanced Technology in Cyberspace of Guangzhou University, part-time researcher of the Pengcheng Laboratory (Shenzhen, China), and CCF Distinguished Member. She is engaged in the research of basic theory and advanced intelligent algorithm design in the field of cyberspace security threat awareness. She has published more than 70 SCI/EI papers, including 5 highly cited papers. Hosted over 12 projects including National Natural Science Foundation of China and Sub-topic of National Key Research and Development Plan. 23 patents were applied and 10 were authorized. Relevant achievements have been applied in government functional departments and enterprises such as China Academy of Cyberspace Research and Beijing Aerospace Automatic Control Institute
5. Approximate Computation for Big Data Analytics
Speakers:
- Shuai Ma (Beihang University)
Brief outline of the tutorial:
Over the past a few years, research and development has made significant progresses on big data analytics. A fundamental issue for big data analytics is the efficiency. However, there are cases in practice that the optimal solution is impossible to obtain or not required or has a high price to pay, and it is reasonable to sacrifice optimality with a “good” feasible solution that can be computed efficiently. Approximate computation is one important way to achieve this goal for big data analytics, and has been paid more and more attentions. In this tutorial, we shall cover different approximation techniques, i.e., approximation algorithms, approximate query processing, and approximation computing, and introduce approximate computation for efficient and effective big data analytics.
Speakers Bio
Shuai Ma is a full professor in the SKLSDE Lab at the School of Computer Science and Engineering, Beihang University, China. He obtained two PhD degrees: University of Edinburgh in 2010 and Peking University in 2004, respectively. His research interests include big data, and database theory and systems. He is/was an Associate Editor of the VLDB Journal, Knowledge and Information Systems and IEEE Transactions on Big Data. He is a recipient of the best paper award of VLDB 2010, the special award of Chinese Institute of Electronics for progress in science and technology in 2017, the best paper candidate of ICMD 2019, the National Science Fund of China for Excellent Young Scholars in 2019, and an outstanding speaker of in ICA3PP 2021.
6. Sequential Recommendation: Formulation, Technologies and Future Directions
Speakers:
- Jiajie Xu (Soochow University)
- Pengpeng Zhao (Soochow University)
Brief outline of the tutorial:
Sequential recommendation is an essential part of modern recommmender systems due to its ability of capturing users’ dynamic preference from their historical activities. With increasing research interests in sequence models recently, this tutorial aims to give a comprehensive and up-to-date introduction of sequential recommendation methods, including its background and problem formulation. We provide an overview on the fundamental sequence modeling techniques. In particular, we systematically summarize the representative and latest research developments on this topic. Finally, we identify some existing challenges and future directions in this area.
Speakers Bio
Dr. Jiajie Xu is a professor with Soochow University, China. His research interests include recommender systems, spatio-temporal data management and mining. He has published over 100 high quality papers in prestigious conferences and journals including TKDE, SIGMOD, PVLDB, ICDE, SIGKDD, IJCAI, AAAI, WWWJ, etc. He received the Best Paper Award and Honorable Mention in CIKM, WISE, APWebWAIM, etc. He has served as program committee member of international conferences such as ICDE, AAAI, IJCAI, SIGKDD, CIKM, etc.
Dr. Pengpeng Zhao is a professor with Soochow University, China. His research interests include data mining, deep learning, and recommender systems. He has published more than 100 papers in international conferences and journals, and 60+ papers were published in CCF rank A/B conferences, such as TKDE, AAAI, IJCAI, ACM MM, WWW, ICDM, CIKM, DASFAA, and ICME. He has served as editorial board member of the International Journal of Intelligent Information Systems, and program committee member of international conferences such as AAAI, IJCAI, WSDM, CIKM, DASFAA and PAKDD.
7. "Redesign of Distributed Relational Databases" Perspective after Thirty Years!
Speakers:
- Kamalakar Karlapalem (IIIT)
- Satyanarayana R Valluri (Databricks)
Brief outline of the tutorial:
The advent of cloud-based databases with an emphasis on scalability and elasticity lead to a new trend of disaggregating storage and compute. The data read from storage is further distributed across multiple nodes during query execution for scale-out to handle increased data volume and query complexity. The tutorial discusses how the distributed design concepts are relevant to the latest cloud-based distributed databases. A distributed design can make the physical layout of the data on storage as efficient as possible to prune/skip data at the source. This tutorial presents a brief history of distributed database design (redesign) problems, some challenges, and how current research and products handle these problems and drive opportunities for future work in this area.
Speakers Bio
Kamal Karlapalem is a Professor and Head of the Data Science and Analytics Center IIIT-Hyderabad. He had worked for over three decades on specific problems of distributed relational database design, object-oriented database partitioning and allocation, and larger data warehouse design. He introduced the problem of a total redesign of distributed relational databases in 1992, and it has been thirty years since.
Satya Valluri is a Software Engineer at Databricks, USA. He is part of the query optimizer group and focuses on optimizing SQL queries for Spark and Databricks SQL. Previously, he worked at Meta Platforms Inc, USA, where he was involved in managing a highly scalable and distributed database that stores the operational data of Meta. Before Meta, Satya worked in the Query Optimizer group of Oracle. His main areas of interest are query processing and optimization, query execution and manageability, and debuggability of features in DBMS systems. Satya did a postdoctoral fellowship at EPFL, Switzerland, and has a Ph.D. from IIIT, Hyderabad.