CSSCS @ DSAA 2022

Liang Zhao

Emory University

Graph Neural Networks for Dynamic Network Modeling and Inverse Problems

Deep learning techniques have achieved tremendous success in continuous data like image and audio. They then extended such success into other data such as network- structured data which are ubiquitous in many scientific such as molecules and societal domains such as Internet of Things. Networks are commonly evolving over time, resulting in temporal networks. Analyzing the time-variant and time-invariant aspects of networks are critical research issues in temporal networks, which benefit wide range of applications such as mobility network simulation, information diffusion, and graph topology evolution. Historically, their modeling and analyses typically rely on the network generation principles predefined by human heuristics and prior knowledge. Such methods usually fit well towards the properties that the predefined principles are tailored for, but usually cannot do well for the others. However, in many application domains like the aforementioned ones, the network properties and generation principles are largely unknown. The recent research frontier on graph neural network provides a data-driven alternative for network representation learning and analysis. In this talk, I will first give a background of graph neural networks and then introduce our recent works on temporal networks. Finally, I will also mention our recent work on end-to-end source localization of graph diffusion process.

Dr. Liang Zhao is an assistant professor at the Department of Compute Science at Emory University. Before that, he was an assistant professor in the Department of Information Science and Technology and the Department of Computer Science at George Mason University. He obtained his Ph.D. degree as Outstanding PhD student in 2016 from Computer Science Department at Virginia Tech in the United States. His research interests include data mining and machine learning, with special interests in spatiotemporal and network data mining, deep learning on graphs, nonconvex optimization, and interpretable machine learning. He has published over a hundred papers in top-tier conferences and journals such as KDD, TKDE, ICDM, ICLR, NeurIPS, Proceedings of the IEEE, TKDD, CSUR, IJCAI, AAAI, and WWW. He won NSF Career Award in 2020 and Jeffress Trust Award in 2019. He also won Amazon Research Award in 2020 and Meta Research Award in 2022. He was honored to be a “CIFellow Mentor” in 2021. He was ranked as “Top 20 Rising Star in Data Mining” by Microsoft Search in 2016. He won several the Best Paper Award and Candidates such as Best Paper Award in ICDM 2019, Best Paper Candidate in ICDM 2021, Best Paper Award Shortlist in WWW 2021, and Best Paper Candidate in ACM SIGSPATIAL 2022. He is an IEEE senior member.

• 8. PATE: Property, Amenities, Traffic and Emotions Coming Together for Real Estate Price Prediction
• 106. Large-Scale Traffic Congestion Prediction based on Multimodal Fusion and Representation Mapping
• 179. Positive emotions help rank negative reviews for sellers and producers in e-commerce
• 191. What Really Drives the Spread of COVID-19 Tweets: A Revisit from Perspective of Content
• 374. Disentangling the Growth of Blockchain-based Networks by Graph Evolution Rule Mining
• 381. Simulating Spreading of Multiple Interacting Processes in Complex Networks
• 384. Modeling the impact of external influence on green behavior spreading in multilayer financial networks
• 404. Improving Source Localization by Perturbing Graph Diffusion
• 432. H4M: Heterogeneous, Multi-source, Multi-modal, Multi-view and Multi-distributional Dataset for Socioeconomic Analytics in the Case of Beijing

• Paper Submission Deadline: June 1, 2022
• Paper Notification: July 31, 2022 August 3, 2022 (extended)
• Camera-ready submission: August 15, 2022 August 20, 2022 (extended)

In the last two decades, there is a surge of interest to computational social scientists to leverage the data explosion in explaining and predicting social phenomena. Here, we aim to bring together state-of-the-art interdisciplinary research and invite researchers from different disciplines and methodological backgrounds to exchange ideas, research questions, recent results, and future challenges in this emerging area of research and public interest.

• Social network analytics
• Online communication and (mis)information diffusion
• Cultural, opinion, and normative dynamics
• Social media data mining
• Social media visual analytics
• Modelling, tracking and forecasting dynamics in social media
• Patterns in social-ecological systems
• Social, economic, environmental simulations or analysis
• Agent-based social models
• Peer production and mass collaboration
• E-democracy and government-citizen interaction
• Cooperative platforms
• Crowd-sourcing, herding behaviour, wisdom of crowds
• Group formation, evolution analysis

• Step 1: Login and enter DSAA conference in CMT3. Website: https://cmt3.research.microsoft.com/DSAA2022/
• Step 2: Click the "Create new submission" button and then "Special Sessions".
• Step 3: Enter your paper information and select "Computational Social Science and Complex Systems" as the subject area.

• The length of each paper submitted should be no more than 10 pages, and formatted following the standard 2-column U.S. letter style of IEEE Conference template. See the IEEE Proceedings Author Guidelines for further information and instructions.
• All submissions will be double-blind reviewed by the Program Committee on the basis of technical quality, relevance to the scope of the special session, originality, significance, and clarity. The names and affiliations of authors must not appear in the submissions, and bibliographic references must be adjusted to preserve author anonymity. Submissions failing to comply with paper formatting and authors anonymity will be rejected without reviews.
• Authors are also encouraged to submit supplementary materials, i.e., providing the source code and data through a GitHub-like public repository to support the reproducibility of their research results.

• All accepted full-length special session papers will be published by IEEE in the DSAA main conference proceedings under its Special Session scheme. All papers will be submitted for inclusion in the IEEEXplore Digital Library. The conference proceedings will be submitted for EI indexing through INSPEC by IEEE.
• Organizers of Special Session may additionally, arrange for special issues to further publish the extended journal versions of the papers. Several past special sessions have published special issues with the International Journal of Data Science and Analytics (JDSA, Springer).

• (i) information necessary for reproducing the experimental results reported in the paper (e.g., various algorithmic and model parameters and configurations, hyper parameter search spaces, details related to data set filtering and train/test splits, software versions, detailed hardware configuration, etc.).
• (ii) any data, pseudo-code and proofs that could not be included in the main page of the manuscript due to space limitations.

The list of authors at the time of submission is considered final and any further changes of the authorship are not allowed.

DSAA is an archival publication venue as such submissions that have been previously published, accepted, or are currently under consideration at other peer-review publication venues (i.e., journals, conferences, workshops with published proceedings, etc) are not permitted.

COIs must be declared at the time of submission. COIs include employment at the same institution within the past three years, collaborations during the past three years, advisor/advisee relationships, plus family and close friends.

At least one of the authors of each accepted paper must register in full and attend the conference to present the paper. No-show papers will be removed from the IEEE Xplore proceedings.

• Satoshi Kurihara, Keio University
• Qi Liu, University of Science and Technology of China

• Yaping Zhao, the University of Hong Kong
Inquiries about this special session should be sent to: zhaoyp (at) connect.hku.hk

• Ali Hosseiny, Shahid Beheshti University
• Andreia Sofia Teixeira, University of Lisbon
• Bodong Zhou, Individual Researcher
• Briane Paul V. Samson, De La Salle University
• Fabiola Pereira, Federal University of Uberlandia
• Haitian Zheng, University of Rochester
• Hyunuk Kim, Boston University
• Jiahui Liu, The University of Hong Kong
• Kaiping Chen, University of Wisconsin-Madison
• Mengli Yu, Nankai University
• Peggy Lindner, University of Houston
• Rene C. Batac, De La Salle University
• Shan Lu, Central University of Finance and Economics
• Songyi Cui, The University of Hong Kong
• Stephan Leitner, University of Klagenfurt
• Subhayan Mukerjee, National University of Singapore
• Xiaoqian Hu, Capital University of Economics and Business
• Yang Yang, Beihang University
• Yuwei Chuai, Beihang University
• Zhenkun Zhou, Capital University of Economics and Business