Code and Dataset for Pattern Recognition Benchmarks
2016-12-12 19:21:33 | Datasets | Contributor(s): Jonas Hepp, Yellamraju Tarun, Mireille Boutin | doi:10.4231/R7G73BPN
This code computed a sequence of bounds for the error rate of a pattern recognition method. The bounds correspond to the error rate that one would expect to achieve by simply selecting features at random and thresholding the feature (TARP) approach.
https://purr.purdue.edu/publications/2030