IEEE Features CS Research
- By Dawn Marie Densmore
A research paper by computer science PhD student Narine Manukyan, and CEMS Professors Margaret J. Eppstein and Donna M. Rizzo, has been accepted for publication in the prestigious journal IEEE Transactions on Neural Networks and Machine Learning.
This article entitled, “Data-Driven Cluster Reinforcement and Visualization in Sparsely-Matched Self-Organizing Maps” proposes a new method using artificial neural networks (a form of computational intelligence inspired by the way the brain operates) for automatically detecting and visualizing clusters in high-dimensional data. As an example of its usefulness, the authors use the method to automatically determine the boundaries of a groundwater contaminant plume surrounding a leaking landfill using 209 microbial community profile variables measured at each of 22 monitoring wells.
The IEEE (Institute of Electrical and Electronics Engineers) is the world's largest technical professional society that promotes the development and application of electrotechnology and allied sciences for the benefit of humanity. This particular journal “is devoted to the science and technology of neural networks, which disclose significant technical knowledge, exploratory developments, and applications of neural networks from biology to software to hardware.”
“This method is really neat” says Prof. Eppstein, “because it lets you automatically detect and visualize hierarchical clusters of relationships in very complex data. We’re now starting to apply this method to other areas, such as mapping functional relationships in the brain. It seems ironic that we are using brain-inspired computing methodologies to try to understand how the brain works!”
“This research showcases some of the innovative ideas and technologies being explored by so many of our talented faculty and students at the University of Vermont,” says Bernard “Chip” Cole, Interim Dean of the UVM College of Engineering and Mathematical Sciences.
The journal web site is: http://ieee-cis.org/pubs/tnn/