How do you use technology/policy to improve citizens’ lives? Tell us about your role or organisation.
People experience and express emotions in response to events that are important to their concerns – be it feeling scared of a deadly disease or showing anger when being treated unfairly by others. Emotional reactions are important information that illuminate people’s underlying concerns, values, desires and opinions, which are useful sources of information for both public policy makers and companies seeking to understand their people, customers and stakeholders and make more informed decisions.
I am a social technologist with a passion for advancing human communications through building and applying useful technologies. I currently lead the Affective and Social Intelligence Group as Group Manager of the Social and Cognitive Computing Department at the Agency for Science, Technology and Research’s (A*STAR’s) Institute of High Performance Computing (IHPC).
My team and I develop and apply social technologies – in particular fine-grained sentiment and emotion analytic engines — to help policy makers understand the different types and intensities of emotions, topics and their patterns and trends, using data from digital and social media that cannot be easily extracted or analysed using conventional tools.
What was the most impactful project you worked on this year?
Since January this year, I started working on a project centred on the understanding of population and media responses to the Covid-19 pandemic, focusing on the trends and insights in Singapore and worldwide.
This project applies our in-house multidimensional emotion analysis technology and social listening capabilities in the context of the Covid-19 pandemic. In collaboration with communication scientists and public health researchers from the local research ecosystem, our studies aim to provide data and insights surrounding the Covid-19 pandemic, such as studying the global trends of sentiments and fine-grained emotion changes over time during the early phases of the Covid-19 outbreak, making sense of the sentiments’ geographic differences across countries, applying a theoretical lens to understand the cultural differences in peoples’ emotional responses, as well as tracking mental well-being patterns and media responses.
What is one unexpected learning from 2020?
What I really didn’t expect was the degree of the complexity and implications from the Covid-19 social listening application studies.
The data and insights turned out to be well-received by the research and public health communities, and fostered new collaborations with communication scientists, psychologists, and public health researchers. Equally valuable to us, this project enlightened my team and I with new inspirations for deeper new research directions.
For example, in our first study analysing the world’s emotion trends through social media, we discovered that there was a significant change of people’s emotional responses to the pandemic — that anger has overtaken fear as the dominant emotion over the course of the pandemic, while sadness and joy also surfaced.
This study was published in collaboration with health communication researchers from Nanyang Technological University’s Wee Kim Wee School of Communication and Information, Tianjin University’s School of New Media and Communication, University of Lugano’s Institute of Communication and Health, and University of Melbourne’s School of Culture and Communication.
Furthermore, we found that there are significant differences in the public’s sentiments and emotional responses to Covid-19 across countries such as Singapore, Korea, Italy and the United States. These discoveries motivated my team to pursue more in-depth ongoing studies to understand the significant differences of such emotional responses across different countries, cultures and varying dimensions of topics surrounding the Covid-19 pandemic.
What tool or technique particularly interests you for 2021? What are your priorities for 2021?
One translational research priority I am planning for 2021 is to work closely with research collaborators and relevant public agencies in Singapore to build a social ground listening system that may potentially inform infectious disease surveillance and transportation ground sensing for Singapore. I hope the tools we are going to build can be a valuable framework and source of insights to help inform policy making and communication.
For the longer term, I plan to extend the social technologies capable of extracting fine-grained information to be reflective of more long-term changes in people’s evolving priorities, desires and values, based on naturalistic social media data. The research outcomes can provide valuable information for future national and corporate planners, to give a ground for foresight exercises with answers to questions such as how the relative values such as “family”, “work-life balance”, “achievement”, “equality”, “open communication”, “environmental awareness” and “compassion” are changing over time, especially on a longitudinal time scale over every year or periods of 5 years.
What advice would you give to women looking to start a career in GovTech?
I hope more women will consider a career in tech, developing and applying technology and data science to help governments and policy makers, such as the case of social ground sensing using more holistic data sources. Ground sensing for policy decision and communication, especially in challenging times such as during the present Covid-19 pandemic, is complex in nature but is also increasingly more challenging in today’s social and digital media era. Women are very capable of being sensitive to details, which enables us to be very good at developing and applying fine-grained tools that help to uncover nuances and new insights about people and policies in a way that conventional tools and approaches may not uncover.
Write a message for your future self.
You have to lose some battles to win the war. Being great is by choice. Do not regret any choices or trade-offs you have to make to do the right thing.
References and Links
 Gupta, R.K. and Yang, Y. (2018). CrystalFeel at SemEval-2018 Task 1: Understanding and Detecting Intensity of Emotions using Affective Lexicons, In Proceedings of the 12th International Workshop on Semantic Evaluation (NAACL-SemEval).
 CrystalFeel – Multidimensional emotion intensity analysis. (2018-2020). Institute of High Performance Computing, A*STAR. Accessible at: https://socialanalyticsplus.net/crystalfeel
 Lwin MO, Lu J, Sheldenkar A, Schulz PJ, Shin W, Gupta R, & Yang Y. (2020). Global Sentiments Surrounding the COVID-19 Pandemic on Twitter: Analysis of Twitter Trends, JMIR Public Health Surveillance. Accessible at: https://publichealth.jmir.org/2020/2/e19447
 Heartbeat for COVID-19 – Live Global Sentiment and Emotion Trend Dashboard. (2020). Institute of High Performance Computing, A*STAR. Accessible at: http://corona2019.socialanalyticsplus.net
 Sensing Emotions in a Crisis, A*STAR Research, 30 Oct 2020. Accessible at: https://research.a-star.edu.sg/articles/features/sensing-emotions-in-a-crisis