Matthew Ward's MQP Topics

1. Multivariate Data Visualization: Several projects are available that deal with enhancement and application of a WPI-generated software package (XmdvTool) for visualizing high-dimensional data sets. Knowledge of Graphics and HCI is essential, and knowing something about OpenGL programming would be very beneficial. Some ideas for potential enhancements to the system include adding animation capabilities, extending some of the display techniques to 3D, and processing streaming or dynamically changing data. We also have a number of database-related projects associated with this effort, co-advised with Prof. Rundensteiner. These include issues such as caching and prefetching algorithms, indexing structures for hierarchical data, and using XML for capturing and presenting data semantics. Projects are also available that involve application of the package to new domains; the projects would benefit from having at least one team member whose major includes the application area. Knowledge of C++ and Eclipse is required - experience with Qt would be a plus.

2. Simulation and Visualization of the Early Biological Development: Computer simulations of biological processes allow us to combine large amounts of different types of information in a way that is meaningful and useful to biologists. Much genetic and cellular information is known about early development in C. elegans, an important model organism. A previous MQP has begun to assemble this information into a visual simulation of development of the animal, starting from a single cell. This project would build on the previous one, making genetic and cellular interactions and visualization more realistic. The simulation will be a useful tool for both research and teaching purposes. A team consisting of both CS and BBT students would be ideal (co-advised with Prof. Ryder).

3. Software Visualization: Prof. Heineman and I are interested in developing methods for visually exploring the structure and behavior of large software systems to aid in debugging, monitoring, testing, and comprehension. Java skills are a must, as well as graphics and HCI.

4. Visual Data Mining: Prof. Ruiz and I are interested in integrating analytic and visual data mining technologies to help identify structure in large databases of information. Application areas could include medical informatics, homeland security, bioinformatics, network security, or even e-mail processing. Knowledge of data mining and/or graphics is essential.

5. Visualization Across the Curriculum : Most, if not all, educational activities can benefit from visual representations. The question is how one could expand on its use. Projects of this type can include developing visualization-centered learning modules for given courses, designing curriculum components to help students design better visualizations, and creating tools to facilitate the creation of visualizations by students and faculty. Clearly, knowledge of HCI and graphics is essential. Some projects of this sort could be combinations of IQPs and MQPs.