Browsing by Author "Kwon, Kyeongan 1979-"
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Item Interfacing Information in User Studies with Mixed Methods(2014-05) Kwon, Kyeongan 1979-; Pavlidis, Ioannis T.; Chen, Guoning; Tolar, TammyUser studies have been growing larger in terms of size and duration. This is an exciting development because user studies are realistically performed beyond the lab environments. Yet, these studies are challenging because researchers have to organize and analyze large amounts of data. Proper statistical analysis is critical and has received a lot of attention. One aspect that has been neglected is a visual interface to the study's results. Such an interface can support qualitative understanding, conveying at a glance studies of sympathetic responses. A case in point is the student exam study that we discuss in this research. One challenge that such a research faces is effective visualization of the study data (e.g., physiological data) that otherwise requires technical expertise to comprehend. Another challenge is the multidimensionality of the study data. Static snapshots (e.g., performance data) and dynamic evolution (e.g., physiological data) have to be visualized at a glance. Moreover, the visualization scheme should take into account spatiotemporal aspect of the examination by presenting study results from multiple subjects over a period. In this research we propose a set of designing principles for effective visualization of the study results. The designing principles are evaluated on the student exam study which aims is to understand students' stress patterns while taking course exams. In particular, a visualization interface is developed as per the designing principles to comprehend the exam study results at a glance. The interface represents voluminous data in a way that the users with little domain specific knowledge can be able to derive meaningful conclusions. It effectively displays the students' course performance data and their stress profile in one glance. The interface also allows inter-subject and intra-subject comparisons in a few mouse clicks or finger taps.Item Scalable Data Visualization Methods for Academic Careers(2016-08) Kwon, Kyeongan 1979-; Pavlidis, Ioannis T.; Deng, Zhigang; Chen, Guoning; Uzzi, BrianIn this dissertation, I have developed scalable data visualization methods to depict a scholar's accomplishments at a glance. The evaluation of scholarly achievements in academia is largely based on the researcher's publication record. This record is communicated in exhaustive detail in the researcher's curriculum vitae (CV) or in summary via her/his h-index. The h-index, although a convenient abstraction, does not consider neither the time of the publication nor the impact factor (IF) of the journal where it appeared. I present a novel method that visually complements the h-index, revealing at a glance the nature of a researcher's scholastic record. This method (which includes the visualizations Scholar Plot and Academic Garden) is particularly appropriate for web interfaces, as it produces information that is compact and simple, yet highly illuminating. Scholar Plot uses Google Scholar, Impact Factor, and NSF/NIH/NASA funding data to create a temporal representation of a researcher's publication/funding record that blends publication prestige with paper popularity and funding information. Scholar Plot affords an insightful appraisal of academics at one's fingertips. Academic Garden applies to individual academics, departments, colleges, and any other academic group thereof, such as a research lab or a project team. Academic Garden uses the flower metaphor to visually articulate performance of academic entities. The width of the flower's stem is commensurate to the academic funding the entity received (`juice conduit'). The height of the flower's stem is commensurate to the impact of the entity's intellectual products (`visibility'). The diameter of the flower's disc is commensurate to the prestige of the venues where these products appeared (`fancy factor'). Scholar Plot and Academic Garden bring clarity, transparency, and fairness in hiring, promotion, tenure, and funding decisions. For the validation of the Academic Garden, I ran data analysis using Endowed Chaired Faculty, a prestigious honor in the United States, for the top 10 universities according to the US News Report 2015. The analysis demonstrated that chaired faculty can be predicted using the 3 merit criteria of citations, impact factor, and funding.