SICSS-Singapore

Location: Conference Room; Level 1, 3 Research Link, Innovation 4.0, 117602

9:00 - 9:30 Registration

9:30 - 10:00 Self-introduction (participants)

10:00 - 11:30 Keynote Lecture by Noshir Contractor, Northwestern University / Title: Using Digital Exhaust from the Web to Leverage Network Insights in the Algorithmically Infused Workplace / Abstract: Social network research can enhance performance in the algorithmically infused workplace. The evidence of social networks to identify who has good ideas, who is influential, and what teams will get work done efficiently and effectively is well established. The challenge has been collecting network data via time-consuming surveys, which elicit low response rates and have high obsolescence. This talk presents empirical examples to demonstrate how we can leverage people analytics to mine data created by individuals in their digital transactions to address challenges they face with issues such as team assembly and team conflict.

11:30 - 13:00 Lunch (The Green Wall Lounge, Level 1)

13:00 - 14:40 Talk by Chao Yu, National University of Singapore / Title: Exploring Subtle Motivations and Unintended Consequences Using Text Sentiment Analysis / Abstract: Communication behavior is observable to an unprecedented degree, primarily through raw text, although visual content is becoming increasingly prevalent. In this workshop, we will focus on the analysis of online reviews, a common form of raw text, through two focused tutorials. Each session begins with an empirical study to ground our exploration. The first tutorial covers sentence-level sentiment analysis, which provides insights into how sentiments are expressed across different parts of a text. The second tutorial advances to a more targeted analysis of sentiments associated with specific aspects within a text (e.g., sentiment related to food or service, or the environment in a restaurant review). This approach not only helps understand individual-level motives but also explores macro-level phenomena that offer a comprehensive view of communication behaviors as they manifest online.

14:40 - 15:10 Tea Break

15:10 - 16:10 Research Sharing (participants) / Xueyan Cao, The Chinese University of Hong Kong / Yi Ting Chen, University of Wisconsin-Madison / Victoria Chua, Nanyang Technological University / Jianing Deng, Duke University / Min Gong, National University of Singapore / Kunmei Han / National University of Singapore

Location: Boardroom, #04-04 The Centre for Trusted Internet Community (CTIC), 3 Research Link, Innovation 4.0, 117602

10:00 - 11:30 Talk by Chris Chao Su, Boston University (online) / Title: Decoding the Dynamics of Media Platforms: Two Computational Approaches / Abstract: This talk will examine the dynamics of media platforms through two research projects employing computational methods. In the first project, a network analysis approach is employed to investigate the citation and co-citation network behind fact-checking content in the United States, revealing the selective way online fact-checkers cites sources. The second project analyzes the value and regulation systems embedded within the public-facing policies (community guidelines) of media platforms through lexical analysis and network analysis. Using digital datasets that are not commonly employed in communication research, the findings of these two projects aims to understand the complexity of media platforms and the impact on their audiences.

11:30 - 13:00 Lunch

13:00 - 14:40 Talk by Tianyu He, National University of Singapore / Title: Thinking and Working with Algorithms in Social Sciences / Abstract: In this talk, I will argue that computational methods will have a profound impact on the future of social science research. This involves incorporating algorithms into two key areas of the research process. First, leveraging algorithms to generate novel theoretical conjectures and validate these hypotheses. Second, utilizing algorithms to create better experimental paradigms and operationalize previously elusive concepts. The talk will provide an overview of algorithm-aided research, illustrated with specific empirical examples.

14:40 - 15:10 Tea Break

15:10 - 16:10 Research Sharing (participants) / Yuhan Hu, University of Oxford / Tengjiao Huang, DSO National Laboratories / Xiaoyun Huang, The Chinese University of Hong Kong / Jianfeng Lan, Shanghai Jiao Tong University / Gionnieve Lim, Singapore University of Technology and Design / Xinyi Liu, Northwestern University / Siyuan Brandon Loh, Institute of High Performance Computing (Astar)

Location: Boardroom, #04-04 The Centre for Trusted Internet Community (CTIC), 3 Research Link, Innovation 4.0, 117602

10:00 - 11:30 Talk by Jiaxin Pei, University of Michigan (online) / Title: Scaling-up Social Research with Efficient Data Annotation / Abstract: Computational Social Science research often involves analyzing large-scale datasets in different modalities (e.g. text and images). Despite recent advances in AI, data annotation is often needed to build task-specific machine-learning models for automatic data analysis. This talk covers practical guides on how to effectively collect data annotation for various social NLP tasks. Furthermore, I will introduce how to use POTATO for efficient data labeling in various settings.

11:30 - 13:00 Lunch

13:00 - 14:40 Talk by Prasanta Bhattacharya; Agency for Science, Technology and Research (ASTAR) / Title: Analysing Social Networks: Concepts, Methods and Applications / Abstract: Social networks influence nearly every aspect of our social and professional lives, often invisibly. The quantitative analysis of social networks has become increasingly systematized over the years, and is now widely considered a mature and frontline discipline in the study of human behaviour. This talk will introduce participants to foundational concepts, methods and a few relevant studies concerning the analysis of dynamic social networks – focusing particularly on the challenge of inferring causal relationships in large networks using both experimental and non-experimental approaches. We will conclude with a brief discussion on some of the emerging challenges and untapped research opportunities in this space.

14:40 - 15:10 Tea Break

15:10 - 16:10 Research Sharing (participants) / Qianfeng Lu, Università della Svizzera italiana / Tianqi Song, National University of Singapore / Jiajun Tang, Nanjing University / Youyi Wei, The Chinese University of Hong Kong / Tianyi Yang, University of Massachusetts Amherst / Xuzhen Yang, Michigan State University

Location: Boardroom, #04-04 The Centre for Trusted Internet Community (CTIC), 3 Research Link, Innovation 4.0, 117602

10:00 - 11:30 Talk by Koustuv Saha, The University of Illinois Urbana-Champaign (online) / Title: Measuring Wellbeing in Situated Contexts with Social Media and Multimodal Sensing: Promises and Perils / Abstract: In this talk, Koustuv Saha will present theory-driven computational and causal methods for leveraging social media in concert with complementary multisensor data to examine wellbeing, particularly in situated communities such as college campuses and workplaces. He will also interrogate the meaningfulness of the data and inferences and reflect on how these approaches can potentially be misinterpreted or misused without additional considerations. To bridge the gap between the theoretical promise and practical utility, he will present the importance of evaluating the needs, benefits, and harms of wellbeing sensing technologies in practice. This talk will propel the vision toward questioning the underlying assumptions and in responsible design and deployment of wellbeing sensing technologies (if at all) for situated communities and the future of work.

11:30 - 13:00 Lunch

13:00 - 14:40 Workshop (Part I): Joe Simons, Prasanta Bhattacharya; Agency for Science, Technology and Research (ASTAR) / Title: An introductory workshop on using large language models in social science / Abstract: Recent developments in language modelling have opened up new opportunities for social science research. Taking social psychology as a focal case, this workshop will discuss some promising directions for using these new tools. This will be followed by a hands-on session on setting up and using LLMs for commonly encountered research tasks.

14:40 - 15:10 Tea Break

15:10 - 16:10 Workshop (Part II): Wong Liang Ze; Agency for Science, Technology and Research (ASTAR) / Title: An introductory workshop on using large language models in social science / Abstract: Recent developments in language modelling have opened up new opportunities for social science research. Taking social psychology as a focal case, this workshop will discuss some promising directions for using these new tools. This will be followed by a hands-on session on setting up and using LLMs for commonly encountered research tasks.

Location: Boardroom, #04-04 The Centre for Trusted Internet Community (CTIC), 3 Research Link, Innovation 4.0, 117602

10:00 - 11:30 Talk by Yingdan Lu, Northwestern University / Title: From Pixels to Politics: Computational Visual Analysis in Political Communication / Abstract: The increasing accessibility of digital visual data and the proliferation of visual-based social media have motivated scholars to examine important political communication questions through analyzing large-scale visual data. Recent advances in computer vision, deep learning, and large multimodal models have provided researchers with powerful tools for such scalable visual analysis. This talk explores how computational visual analysis enhances our understanding of political communication by showcasing two empirical studies. The first study employs unsupervised image clustering on protest visuals from ten million tweets to understand the role of visual content in three prominent social-mediated protests. The second study examines over five million Douyin videos from over 18,000 regime-affiliated accounts in China, using computer vision methods to uncover a decentralized model of modern propaganda production and dissemination. In addition to demonstrating how computational analysis of visuals can shed light on political processes and outcomes, I will discuss the opportunities, challenges, and future directions for this approach in communication research.

11:30 - 13:00 Lunch

13:00 - 14:40 Grouping (participants)

14:40 - 15:10 Tea Break

15:10 - 16:10 Grouping (participants)

Location: Conference Room; Level 1, 3 Research Link, Innovation 4.0, 117602

10:00 - 11:30 Talk by Diyi Yang, Stanford University (online) / Title: Can Large Language Models Transform Computational Social Science? / Abstract: Large language models (LLMs) have created unprecedented opportunities for analyzing and generating language data on a massive scale, which has the potential to transform the field of social sciences since language data play a central role in all areas. In this talk, we present a road map for using LLMs as computational social science tools. To do so, we first analyze the zero-shot performance of 13 LLMs on 24 representative computational social science benchmarks. We then demonstrate how they can enhance the CSS research pipeline and significantly reduce costs and increase the efficiency of social science analysis. Finally, we outline a few major concerns about the application of LLMs to social sciences, and make recommendations for investments that may help to address them.

11:30 - 13:00 Lunch (The Green Wall Lounge, Level 1)

13:00 - 14:40 Closing Talk by Alvin Zhou, University of Minnesota Twin Cities / Title: We Are Computational Storytellers: Forging Synergy Between Question, Data, and Method to Create a Compelling Research Design / Abstract: Instead of blindly applying computational methods to social data, computational social scientists should forge synergy between question, data, and method to create a research design that tells a compelling story. The talk will introduce three case studies, discuss some behind-the-scene thinking processes, and explore how CSS projects — starting with an interesting research question, or some invaluable data, or an innovative methodological framework — can forge synergy between the three elements to create a research design that tells an engaging social science story.

14:40 - 15:10 Tea Break

15:10 - 16:10 Presentations (participants)