Clustering historical wildfire location data.
Project for CS 210 students, University of Oregon. Instructions for students in docs/HOWTO.md.
Learning objectives:
- Successive approximation as a fundamental algorithmic technique
- k-means clustering as an example of successive approximation
- Parallel array structures (lists with matching indexes)
- Incremental construction of an application with complex data structures, with testing on small example data sets
This project incorporates a fork of John Zelle's graphics.py
module, which carries a GPL license, so this project is necessarily
also covered by GPL. I will substitute CC-by-SA when and if I
produce a "cleanroom" implementation of the needed functionality
from graphics.py.
See data/README.md for some notes on substituting different data sets and basemaps. This project is used at University of Oregon in a CS-1 class (CS 210 at UO, equivalent to CS 161 at other Oregon colleges and universities), approximately four weeks into the academic term.