C-Support Vector Classifaction
Support vector machines (SVMs) are a set of supervised learning methods used for classification, regression and outliers detection.
The C-Support Vector Classifaction came out with the following results
The best performance was a linear classification that was 98% accurate
The worst performance was a sigmoid classification that was 47% accurate
Since we didn’t have a good idea of how the higher dimensional data might look, we chose to look through a variety of separation plane curves. We tried linear, a few polynomial fits (both even and odd function values), gaussian and sigmoid. As it turned out, just simple linear hyperplanes worked best with the same 98% accuracy and 3% false negative rates as most other approaches.
