DSI Inaugural "Analytics for Social Good Seed Grants "

Saturday, September 3, 2016

In early September of this year the Data Science Initiative (DSI), through the newly launched “Partnership for Social Good,” awarded two $20,000 grants to UNC Charlotte faculty and staff.  The purpose of the awards to provide start-up funds for new, innovative and interdisciplinary research, projects that engaged multiple colleges.  Several outstanding proposals were put forward.  Proposals were evaluated on several metrics including perceived "Social Good" impact and the likelihood for future external funding.  The two winning teams were:  

Leveraging Social Media and Online Resources to Combat Human Trafficking

  • Wenwen Dou, Department of Computer Science, College of Computing and Informatics
  • Matthew Phillips, Department of Criminal Justice and Criminology, College of Liberal Arts and Sciences
  • Jean-Claude Thill, Department of Geography and Earth Sciences, College of Liberal Arts and Sciences

 

Understanding the Factors that Influence Food Choice among UNC Charlotte Students

  • Elizabeth Racine, Department of Public Health Sciences, College of Health and Human Services
  • Wlodek Zadrozny, Department of Computer Science, College of Computing and Informatics
  • Lisa Schulkind, Department of Economics, Belk College of Business
  • Arthur Zillante, Department of Economics, Belk College of Business

The first proposal focused on human trafficking, a form of modern slavery where people profit from the control and exploitation of others, forcing them to engage in commercial sex or to provide services against their will. Modern technology has enabled Human Trafficking to flourish in the 21st century. Ubiquitous access to high speed and private information/communications channels enables the targeting and recruiting of victims and their subsequent exploitation. The research effort seeks to make use of the traffickers' own reliance on this technology to gain a more complete awareness of the scope and nature of human trafficking recruitment activities and gain a visual analytic tool to potentially facilitate the creation of strategies and interventions to combat this problem. Their visual and text analytic approach will explore the fusion of data from a variety of sources – including Social Media and websites in order to enable data-driven analysis of human-trafficking related activities.

The second project intends to explore young adults transition from the home environment to a university environment and their subsequent risk of overeating and becoming obese. One powerful resource, to help in understanding and predicting food choice, is utilization of the food transaction data generated when students purchase food with their meal plan (49er card).  Step 1 of their project was to obtain university approvals (legal, IRB, Auxiliary Services, etc.) to share the data. Their next step will be to take the food transaction data, add nutrient data, and build an analytic dataset. Graduate students in three colleges (CHHS, CCI, BCOB) will be hired to clean and construct the analytic dataset. The analytic dataset will then be used to answer two preliminary research questions:

1. How do food purchases differ by demographic characteristics (gender, race/ethnicity, financial aid status, etc.)?

2. How do student food purchases change over the academic year?

Findings will be used to develop interventions and/or educational materials to help students make healthier food purchasing decisions. External funding will be sought to further analyze the data and, initially, to measure the impact of the Affordable Care Act Nutrition Labeling policy on food purchases.

All told the DSI received 14 outstanding submissions from across the campus.  Interdisciplinary project teams included representation from six colleges: Belk College of Business, College of Arts + Architecture, College of Computing and Informatics, College of Education, College of Health and Human Services, College of Liberal Arts & Sciences).  The DSI plans to announce a new competition in mid- November.