Senior Honors Thesis
A Multifaceted Interactive Analysis of the Behavior of a Single Neuron Model
Elizabeth A. Gifford
Brandeis University, Department of Computer Science
Senior Honors Thesis, May 2006
Introduction
In the course of my research, I developed several techniques to explore the contents of a large database of the various activity patterns of a single neuron model. I used feature extraction and k-means clustering to assign attributes to neuron models that could vary as the modeľs inputs vary and developed a visualization software tool to facilitate the exploration of that relationship.
Other Contents
- Biological Background
- Creating the Data Set
- Computational Goals
- Software Design
- Compression
- Human Classification Tool
- Feature Extraction
- Two-Dimensional Slice Viewer
- Mining Activity Patterns Using K-Means Clustering
- Conclusion
- Future Work