Ph.D. Defense of Dissertation Announcement
Title: Integration of Computational Methods and Visual Analytics for Large-scale High-dimensional Data

Jaegul Choo
Computational Science and Engineering Ph.D. Student
School of Computational Science and Engineering
College of Computing
Georgia Institute of Technology

Friday, January 4, 2013
2:00PM
Location: KACB 1212 (NEW LOCATION)

Committee:

• Prof. Haesun Park, School of Computational Science and Engineering, Advisor
• Prof. Alexander Gray, School of Computational Science and Engineering
• Prof. Guy Lebanon, School of Computational Science and Engineering
• Prof. John Stasko, School of Interactive Computing
• Dr. Pak Chung Wong, Pacific Northwest National Laboratory

Abstract:
Large-scale high-dimensional data analysis is becoming important in many areas with an increasing amount of collected data. One approach to analyze this data is by using fully computational methods, and the other is by leveraging humans' capability via interactive visualization. While the former can deal with large-scale data, it lacks deep data understanding, which may be critical to the analysis tasks. On the other hand, the latter can give the insight about the data, but it suffers from large-scale of the data. Even with a clear motivation to integrate these two approaches, little progress has been made. To tackle this problem, I claim that computational methods have to be re-designed both theoretically and algorithmically and the visual analytics system has to expose these computational methods to users so that they can choose the proper algorithms and settings.

To achieve an appropriate integration between computational methods and visual analytics, the thesis focuses on essential computational methods for visualization such as dimension reduction and clustering, and it presents fundamental improvements of computational methods as well as visual analytic systems involving such improved methods.

For the improvements of computational methods, the contributions of the thesis include:
1. Two-stage dimension reduction framework that better handles significant information loss in visualization of high-dimensional data.
2. Efficient parametric updating of computational methods for fast and smooth user interactions.
3. Iteration-wise integration framework of computational methods in real-time visual analytics.

On the other hand, the latter parts of the thesis focus on the development of visual analytics systems involving the presented improvements, such as 4. Testbed: an interactive visual testbed system for various dimension reduction and clustering methods, 5. iVisClassifier: an interactive visual classification system using supervised dimension reduction, and 6. VisIRR: an interactive visual information retrieval and recommender system for large-scale document data.

During the defense presentation, I will put more emphasis on my recent research, such as 3, 4, and 6 than the rest.