Featured Researcher: Tianjie Deng

Tianjie Deng

Each week, Daniels is featuring a researcher who conducts meaningful research that impacts their field and the wider community. Learn more about their work in Q&As with the Daniels Research team and email them to nominate yourself or a colleague for a future Q&A. 

Tianjie Deng is an associate professor in the Department of Business Information and Analytics. Her primary research objective is to leverage digital trace data to benefit individuals and businesses, helping them create value. Her work has been published in esteemed academic journals and conference proceedings, including Information and Management, ACM Transactions on Management Information Systems, International Journal of Information Management, Information Technology & People, Journal of Economics and Business, Americas Conference on Information Systems, Hawaii International Conference on System Science, and International Conference on Global Perspectives on Design Science Research. 

What do you study? 

I’m interested in looking at digital trace data generated by online communities or businesses so that I can help businesses, communities and individuals create value. I can give three examples. First, I’ve looked at unstructured data from online communities of open-source software development teams to look for patterns in their work that might help them become more popular in terms of attracting and maintaining developers. When do they submit code? When do they review and comment on the codes, etc.? 

Another example: looking at hotel reviews and analyzing how a manager should respond to negative comments. Should they be apologetic? Should they provide an explanation? Which reaction will provide more consumer satisfaction and increased sales? 

Lastly, and what I’m most passionate about, is studying online communities of homeless youth and using some of these data-mining techniques to understand if individuals are in increased danger of substance use. I’m currently working on this study.  

How are you conducting this study? 

Young people experiencing homelessness rely heavily on social media to stay connected with family, friends and social workers. We’re using text mining to extract sentiment and topics from online posts to predict the likelihood of whether or not they will engage in substance use. The goal is to create a model with high accuracy that facilities can use to offer more resources to kids at higher risk and provide those resources earlier.  

We’re working with DU’s Graduate School of Social Work and the downtown shelters to connect with youth. They sign a consent form for us to collect data from their social media platforms and fill out a comprehensive questionnaire that addresses health behavior and family history. This is focused on Denver right now but ultimately we are hoping to apply this model to a larger data set.  

Could you talk about how the teacher-scholar model applies to your work? 

I teach a class called Complex Data Analytics in which we look at text mining and social network analysis, so my research is directly applicable to the classroom. The research allows me to teach my students better, because I can introduce them to new methodologies that I have come across in my research and stay updated that way. I also share my research with them to show them the practical relevance of the skills they’re learning. They get more hooked on the concept and motivated to learn because, before coming into the class, they don’t really understand what the topic is.  

And then lastly, I learn from them too. As college students, they are more similar to the youth we are studying than I am, so they can tell me, you know, this isn’t really the way that we talk, or you need to go here and look for this in the slang dictionary. For example, I can ask them to take some time to come up with the best way to extract information from the texts because social media text is full of typos, slang, emojis and codes. So it’s a good exercise for them, and it makes the model more accurate.  

 How do you hope your work might impact the business community or society as a whole? 

As I shared earlier, I am really passionate about using technology to help underrepresented communities to achieve better outcomes, whether that outcome is better physical or mental health, more robust social support systems or more job opportunities. I firmly believe that technology is most valuable when it is employed to empower and elevate unrepresented communities.