Previous and Current Research

Research Statement

As a computer scientist, my research interests involve the development of practical and intelligent tools to aid in the learning/educational process. In particular, I tend to focus on technological solutions to educational accessibility issues. Intelligent, capable students should not be excluded due to issues with reading, distractability, diminished hearing or sight. I look for technological means for including these students in "normal" educational settings.

It is important to note the two words that I use to describe the tools that I work on: practical and intelligent. By the latter, I look for appropriate applications of artificial intelligence methods, focusing usually on search methods and probabilistic activity recognition vis particle filtering. As for practical, the priority is to make tools that people want to and can use. Devices should not just showcase the latest technical breakthroughs. Understanding the application, the problem domain, and the usability issues are paramount.

For example, the inherent variety of how humans think and learn has led me to appreciate and strive for a variety of techniques. The educational and accessibility tools I study can vary in terms of technical underpinnings (from paper and pencil to the Tablet PC) as well as applications (from in-class activities to data analysis techniques and even daily life activities).

To conclude, the following topics are currently of high interest to me:

• Reading and reading disabilities (e.g. dyslexia)
• Classroom and educational accessibility issues
• The application of learning styles in the educational process
• The interrelations between social networks and the educational process
• Improving participation in collaborative learning exercise
• Developing tools to help understand student conceptual mappings of various topics

Educational Technology Research

Developing Tablet PC Tools for Individuals with Reading Disabilities
The recent development of the Tablet PC provides simultaneously a usable pen/paper interface and a computer. It is my aim to investigate how the tablet can be used to aid individuals with reading disabilities (dyslexia, dyscalculia, etc.), particular for learning activities. This project will involve the design of both reading and notetaking tools that will likely be usable for many users, regardless of reading ability.

Analyzing Card Sorts Through Edit Distance Clustering
This is a continuation of work, along with Richard Anderson and Ruth Anderson on the Bootstrapping Project. As part of our analysis of the card sorting data from this project, we have developed a method for determining the relative distance between card sorts using a defined edit metric. We have begun identifying clusters of closely-related sorts and examining the descriptions the sorters gave to their sorts.
Papers:

• Richard Anderson and Ruth Anderson and Katherine Deibel, "Analyzing Concept Groupings of Introductory Computer Programming Students," November, 2004. (pdf)

• The UW Card Sort Analyzer

Classroom Assessment Techniques for the CS Classroom
Classroom Assessment Techniques (CATs) are small, short assessments used to obtain various information within a classroom. CATs can be used to identify student understanding, determine class dynamics, promote interaction, etc. I am currently developing a suite of CATs to be used in introductory computer science programs, as well as determining what technology innovations (e.g. Steve Wolfman's Stuctured Interaction Presentations) can be applied to CATs.

• The CATS for Computer Science Library

Facilitating Group Work
From personal experience as both a student and an instructor (teaching assistant), the benefit of in-class group work for students is mixed. As part of my TA work, I became interested in using learning styles, particularly the Felder-Silverman styles, for forming groups that promote all members to participate. The initial work on this was part of normal class instruction, so I am looking to formally study the effects of the methods I have developed. This work ties strongly into my work with classroom assessment techniques.
Papers:

• Katherine Deibel, "Facilitating Group Discussions Through Learning Styles and Prior Student Knowledge," September, 2003. (pdf)
• Katherine Deibel, "Improving and Augmenting Classroom Group Work," SIGCSE Doctoral Consortium, March, 2004. (html)

Assisted Cognition Research

Contrail Filtering:
This project was the culmination of my Masters/quals work in the spring of 2003. In order to recognize human activities from a large plan library, a probabilistic inference engine requires not only efficiency and scalability, but also flexibility in terms of configurability. Contrail filtering extends the notion of particle filtering by incorporating memory management and allowing DBN structure to change when necessary.
Papers:

• Katherine Deibel, "Contrail Filtering: A Mechanism for Efficient and Robust Activity Recognition," UW-CSE-03-05, University of Washington, Department of Computer Science and Engineering, Seattle, WA, May 2003. (pdf)

• Katherine Deibel, "Contrail Filtering: A Mechanism for Structured Flexibility in Activity Recognition," May 2003. (ppt)