GPC2024

ABSTRACT: Clinical trials generate highly relevant evidences for effective disease treatments. The extraction of necessary information from a large-scale clinical trial text through natural language processing for patient characteristic aggregation remains a research problem due to the complex of the investigator-authored free-text. This talk will introduce the recent research on: an extensible approach for automated semantic tag mining, clinical trial clustering by similar eligibility criteria, disease named entity recognition, temporal expression extraction and normalization, transgender identification for enhancing clinical trial recruitment, and measurable quantitative information extraction and normalization.

BIO: Dr. Tianyong Hao is a full Professor at South China Normal University. He received Ph.D. degree at City University of Hong Kong in 2010. He studied at York University in 2008 and Emory University in 2009. After that, he worked at University of New South Wales in 2012 and at Columbia University until 2014. Dr. Hao is a senior member of IEEE, an outstanding member of CCF, and the committee member of ISO TC37, SAC TC52 (Associate secretary), AACI (Associate secretary), CIPS (Council member), CIPS Health Informatics section (Vice chair), CCF YOCSEF Guangzhou (Chair, 19-20), CCF Guangzhou (Vice chair) etc. He is also the leader of a provincial level research team on NLP for big data. He currently is the committee director of a provincial engineering-technology center, and the director of a big data center under a provincial research institute. He is the lead guest editor of several SCI journals including JMIR Medical Informatics, BMC Medical Informatics and Decision Making, etc. He has published more than 200 SCI/EI indexed papers. He is the PI of 3 grants from NSFC and more than 20 grants in various levels (over 10 million RMB in total). He owns 9 best paper awards, 8 ISO/national standards (the first ISO standard in the history of SCNU), and 20 patents.

ABSTRACT: In the era of big data, the transformation of research paradigms is affecting various disciplines, and visualization and visual analyses, which closely integrate human intelligence with the powerful computing power of machines, play an important role in it. The report will introduce several recent works carried out by the Visualization Laboratory in close collaboration with various humanities and social science researchers, including maps of the evolution of prehistoric painted pottery patterns and maps of Han nationality spatiotemporal transmission. On the one hand, these interactive maps targeting historical information provide new tools for researchers in the field of history and culture, supporting more efficient exploration and discovery. On the other hand, they also provide a mean for the public to understand specialized knowledge in the field and offer suggestions for developing new exhibition methods.

BIO: Xiaoru Yuan is a researcher at the School of Intelligence Science and Technology, Peking University, Long term Associate Professor, Doctoral Supervisor, and Executive Deputy Director of the National Engineering Laboratory for Big Data Analysis and Applications. Xiaoru is committed to researching general basic methods and domain application systems for visualization and visual analysis. His achievements have been applied in fields such as simulation computing, transportation and aviation, media communication, humanities and social sciences. He has won the Best Paper or Nomination Award at domestic and international visual chemistry conferences such as IEEE VIS and ChinaVis multiple times. He served as the Chair of IEEE VIS Conference Papers in 2017/2021. He is currently a member of the IEEE VIS, PacificVis, ChinaVis Steering Committee, and an editorial board member for domestic and international journals such as IEEE TVCG. He advocated and co founded the China Visualization and Visual Analysis Conference. He is also a member of the Chinese Society of Image and Graphics and the director of the Visualization and Visual Analysis Professional Committee. His visual works have been selected for Beijing Design Week, Zhejiang Art Museum Art Exhibition, as well as conference art project exhibitions such as IEEE VIS, PacificVis, and ChinaVis.

ABSTRACT: With the explosive growth of Internet of Things (IoT) devices and the widespread adoption of 5G networks, the volume of data has surged, increasing the demand for real-time processing and low latency. The traditional cloud computing model faces challenges when handling massive amounts of data generated at the edge, due to transmission delays and bandwidth limitations. To address these issues, collaborative edge intelligence has emerged as a promising paradigm and is becoming a hot topic in research. This report aims to provide an overview of its key concepts, task offloading strategies, resource allocation algorithms, technical challenges, and application cases.

BIO: Tian Wang is a tenured professor at Beijing Normal University and serves as the Engineering Research Center of Cloud-Edge Intelligent Collaboration on Big Data, Ministry of Education. He is a PhD advisor and a recipient of the National Youth Top-notch Talent Support Program. Prof. Wang is the principal investigator for National Key Research and Development Program of China and leads the Innovation Team for Higher Education Institutions in Guangdong Province. He received his PhD from City University of Hong Kong and is recognized among the world’s top 2% of scientists. Additionally, he is part of the Leading Talent Program at Beijing Normal University.
Prof. Wang'sresearch focuses on the Internet of Things and edge intelligence. He has published over 50 papers in CCF A conferences and IEEE/ACM Transactions journals. His work has garnered more than 15,000 citations, with an H-index of 71, and includes 10 ESI Highly Cited Papers (three of which are ESI Hot Papers). He holds 30 authorized patents, with one patent successfully transferred. He has led one National Key R&D Program of China and five National Natural Science Foundation of China. His work has been honored with several awards, including the Hunan Provincial Natural Science Award (Second Prize), the Fujian Provincial Science and Technology Progress Award (Second Prize), and the Fujian Provincial Natural Science Award (Third Prize).

ABSTRACT: As a typical security threat that is easy to generate, widely spread, and difficult to eliminate, crosstalk attacks pose the greatest risk of potential information leakage to security-sensitive services in optical networks. Therefore, how to effectively avoid security threats from crosstalk attacks during the network planning stage is particularly important. In order to effectively protect security-sensitive services to maximize the avoidance of crosstalk attacks, this report introduces a crosstalk attack-aware routing and spectrum allocation technology based on routing avoidance. By analyzing the potential attack risks of different routing and spectrum solutions, a mixed integer linear programming model and heuristic algorithm are used to solve the problem, thereby reducing the risk of information leakage of security-sensitive services in optical networks and improving the security performance of the network.

BIO: Jianqing Li is currently a Professor and an Assistant Director of School of Computer Science and Engineering, Macau University of Science and Technology. He received the Ph. D. degree from the Beijing University of Posts and Telecommunications, Beijing, China, in 1999. From 2000 to 2002, he was a Visiting Professor with the Information and Communications University, Daejeon, South Korea. From 2002 to 2004, he was a Research Fellow with Nanyang Technological University, Singapore. He joined the Macau University of Science and Technology, Macau, China, in 2004. He won Third Prize in Technology Invention of 2016 and 2018 Macao Science and Technology Awards, respectively. He won Third Prize in Natural Science of 2022 Macao Science and Technology Awards. His main research interests include optical networks, wireless networks, fiber sensors, and the Internet of Things.

ABSTRACT: Recent advancements in generative artificial intelligence (AI) and federated learning (FL) have opened new avenues for efficient edge intelligence. This talk will explore the convergence of generative AI and FL within the realm of edge computing. Generative AI has demonstrated remarkable capabilities in synthesizing realistic data. In parallel, FL facilitates collaborative model training across distributed edge devices while ensuring data privacy. This talk will introduce a framework for on-demand generative AI services tailored for edge computing. The discussion will focus on how generative AI can be delivered to heterogeneous edge devices, accommodating varying latency and quality requirements. Additionally, the talk will highlight the potential benefits of integrating generative AI with FL to enhance resource efficiency in edge intelligence. By leveraging generative models to create synthetic training data, we can augment limited local datasets on edge devices. This balance between local and synthetic data allows FL systems to achieve fast and accurate convergence.

BIO: Yuan Wu is currently the Associate Professor with the State Key Laboratory of Internet of Things for Smart City, University of Macau, Macao, China, and also with the Department of Computer and Information Science, University of Macau. His research interests focus on edge computing and edge intelligence. He received the Best Paper Award from IEEE ICC'2016, IEEE TCGCC'2017, and WCNC'2023. He served as the symposium TPC co-chairs for GLOBECOM'2024, ICCC'2023, VTC'2022-Spring, etc. He is currently on the editorial board of IEEE TWC, TVT, and TNSE. He is a senior member of IEEE and distinguished member of CCF, and he currently serves as the Vice Chair of IEEE ComSoC Macau Chapter.

ABSTRACT: In recent years, the amount of bibliometric data available to researchers has grown rapidly with the expansion of coverage and accessibility of bibliometric databases such as Web of Science, Scopus and RISmed PubMed/MedLine. By analyzing of these data, researchers can quickly identify hot trends, author collaborative networks and knowledge graph. Multidisciplinary Digital Publishing Institute (MDPI) is a publisher of open-access scientific journals. It publishes over 390 peer-reviewed, open access journals, and is the largest publisher of open access articles. Due to its business model that prioritizes operational speed and business interests, many of MDPI journals are considered low-quality journals. Utilizing articles published by MDPI between 2014-2024 from the Web of Science website, the current study aims to explore the characteristics and dynamic evolution of the authors' co-authorship network in MDPI's journals. Our findings indicate that the author-ship network is seeing significant growth in size over time, while the overall structure is characterized by a loose configuration and lacks a central position held by influential researchers. The collaborative partnership among researchers mostly relies on the resemblance of geographical and institutional characteristics, indicating a certain level of strategic inclination towards using “internal” resources. We also found that interdisciplinary corporations are quite common, particularly in disciplines characterized by low publication rates within their specific fields, which tend to engage in interdisciplinary research and is likely driven by career advancement goals rather than academic innovation motivations.

BIO: Tianji Cai received his PhD degree at University of North Carolina at Chapel Hill in 2010. His research interests concentrate on new forms of data and new methods of analysis. Reflecting on his broad intellectual pursuits, his research topics are diverse, ranging from methodological, such as quantitative methods and data mining, to substantive ones, such as gene-environmental interplay and adolescent health behaviors. He has published widely in leading international journals including American Journal of Sociology, American Sociological Review, Chinese Sociological Review, Demography, Journal of Quantitative Criminology, Sociological Methodology, Sociological Methods & Research, etc.