The keynote speaker at this year’s Fall Seminar will be Hsuanwei Michelle Chen, Assistant Professor at San Jose State University’s iSchool. In this interview, Prof. Chen tells us a little about her research, including a sneak peak at her upcoming presentation.
Prof. Chen, as a data scientist, your research focuses on big data visualization and social network analysis. What are some specific topics you are currently investigating?
I’m currently working on a few projects that focus on how to use information visualization to improve data analytics and management, which, in turn, enhances collection analysis, user engagement, and resource allocation. For example, one topic that I’m investigating is how information visualization can be adopted to analyze the images that Bob Ross painted and discussed on the PBS series The Joy of Painting. Examining his artistic output as a dataset suggested complex themes related to emotion, memory, and the very act of painting itself. The preliminary findings show how information visualization can be used for art collection analysis by exploring the trajectory of artistic creation. Another topic I’m investigating is visualizing social media interactions (e.g., Pinterest boards and Twitter “tweets”) to further understand how libraries can use social media interactions to engage and attract patrons. One of the initial findings is that library users prefer participating in “active” services created by the library; thus libraries need to spend much more time and effort contributing to a creative theme or design on their chosen topics, rather than simply pushing static posts like book covers.
Social networks reveal the interests of millions of individual consumers, as well as links within and among market segments. Besides targeted advertising, what benefits can social network analysis provide? How do these benefits differ between small and large enterprises?
Social network analysis can provide so much more than just targeted advertising, such as increasing customer loyalty and improving brand impression through better user understanding and engagement. In addition, through other in-depth analyses such as sentiment analysis and trend analysis, we can gain insight into the attitudes and/or behavioral patterns that users present. We can also identify different “types” of users (e.g., leaders vs. non-leaders), which helps target advertising further by tailoring marketing messages based on user influences.
I think the benefits apply to both small and large enterprises, but the strategies to deal with social network analysis may differ greatly based on available resources. Larger enterprises can take a more structured approach to social network analysis (e.g., by having an independent team that focuses on social media initiatives and strategies), while small enterprises may lack resources for intensive analysis. Therefore, you must design social media strategies carefully in order to optimize the potential benefits of social network analysis.
As platforms and their services evolve, what innovations in social network analysis do you foresee over the next 5 to 10 years?
I think one of the biggest innovations in social network analysis will emerge from the big data field – both the evolution, availability, and interconnectedness of large-scale datasets, as well as new methods and tools for data analytics. Although the scale of available data presents many challenges, it also opens exciting opportunities for even more in-depth knowledge mining and discovery, such as sentiment and trend analysis. Simultaneously, we will develop more integrated tools that support powerful platforms to search, monitor, analyze and visualize data.
How can corporate librarians integrate cutting-edge data research into their day-to-day work? What specific services do corporate librarians who lack a computer science background have to offer?
Fortunately, there are more and more data research tools that provide a complete, integrated solution for deep data analysis through user-friendly interfaces. I believe corporate librarians, or those who lack a computer science background, can master these tools. For example, I’m teaching a big data course this semester, and one of my students, who is working for a corporation, is learning how to use Splunk to analyze and visualize a large-scale dataset of consumer purchase history to gain deeper insights into location-based buying behaviors.
To enhance competitiveness, I think it is essential for corporate librarians to gain some skills and knowledge with large-scale data analysis and interpretation.
What specific topics will you be discussing at our upcoming Fall Seminar?
I will be discussing how librarians and information professionals can use information visualization to enhance our daily work. This includes analyzing, displaying, communicating and interpreting massive amounts of abstract data effectively and efficiently via visual representations. I will demonstrate how information visualization can be used to help libraries and librarians utilize abundant data resources (to which they now have more and more access) to provide better collection analysis, resource allocation, and user engagement.
Thank you so much for your time. We look forward to your presentation.