Non-government Organization (NGO) Collaborations
The Maternal and Child Healthcare Mobile Health program operated by ARMMAN aims to improve dissemination of health information to pregnant women and mothers with an aim to reduce maternal, neonatal and child mortality and morbidity. ARMMAN serves expectant/new mothers in disadvantaged communities with median daily family income of $3.22 per day, which is seen to be below the world bank poverty reporeted by The World Bank. The program is composed of multiple enrolled beneficiaries and a planner who schedules service calls to improve the overall engagement of beneficiaries; engagement is measured in terms of total number of automated voice (health related) messages that the beneficiary engaged with.
I work with ARMMAN to design algorithms that can use the domain knowledge from the service call scheduling process to predict engagement behavior in a way to maximize the downstream service call performance. Specifically, due to the limited data of historical engagement behavior and the uncertainty of human behavior, it is impossible to perfectly predict engagement behavior and therefore there is always predictive error involved. But predictive error can lead to different consequence in the service call scheduling, where we definitely do not want a small predictive error leading to a significant degrade in the service call performance. Therefore, it is important to integrate the effect of downstream planning process into machine learning pipelines. The model learned by our model leads to a significantly higher engagement performance using the same amount of service call budget. I am working on deploying the new algorithm and new model to further improve the maternal and child health program performance with ARMMAN.
This project run by The Citizen of the Earth, Taiwan, aims to identify illegal factory expansion using satellite imagery. To mitigate the environmental impact caused by illegal factory expansion, the Citizen of the Earth creates a website game with satellite imagery in different years of the some candidate locations to allow participants to help identify illegal factory and report to the government for further inspection.
This program has significantly raised the attention to the environmental issue caused by industrialization. My role is to further apply artificial intelligence to automatically skim through all potential locations before presenting these locations to the website game.
Theory and Math Clinic @ National Hsinchu University of Education
I believe in math and theory because math and theory are the least resource-dependent skills one could obtain. These are also solid skills that help me throughout my entire undergrad and PhD. I collaborated with National Hsinchu University of Education to teach and pass my experience about college math to high school and middle school students. I wish to provide resources to help more students like me during their early years to strengthen their solid skills before entering college.
On the other hand, during my PhD, I also realized that math and theory are great tools, but not necessarily the final destination. I realized that math and theory can be used to raise academic attention to certain social issues that I care. I talked to social workers in different social issues to understand the fundamental challenges, and built theory and model based on the conversation to further consolidate the challenges and bring more attention to the social issues. This is what drives me to work on interdisciplinary research with theory and social challenges combined.