cot4501fa10_project

cot4501fa10_project - Numerical Analysis Class Project...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Numerical Analysis Class Project Notes Fall 2010 November 4, 2010 The class project involves fitting least-squares classifiers to a political survey assembled by the class using responses obtained at http://www.computingpolitics.com . The machine learning algorithm designed by us is only intended to separate self-identified Republicans from Democrats (setting aside self-identified Independents). Training the classifier Start by assigning Republicans a label of +1 and Democrats a label of -1. We now describe an extremely simple machine learning algorithmbased on linear least-squaresaimed at separating the two classes. Furthermore, arrange the answers to every question on a -1 to 1 scale using { 1 , . 5 , , . 5 , 1 } as the values to the possible answers. It makes sense to re-arrange the answers to coincide with a standard left-right political axis (regardless of how outmoded the spectrum concept is at the present time)....
View Full Document

This note was uploaded on 01/29/2012 for the course COT 4510 taught by Professor Rangarajan during the Fall '10 term at University of Florida.

Page1 / 2

cot4501fa10_project - Numerical Analysis Class Project...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online