| 
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • Stop wasting time looking for files and revisions! Dokkio, a new product from the PBworks team, integrates and organizes your Drive, Dropbox, Box, Slack and Gmail files. Sign up for free.

View
 

Week04

Page history last edited by mike@mbowles.com 8 years, 11 months ago

 

Here are some references for lecture #4. 

 

There is considerable documentation for the "mvoutlier" package.  The programs were developed in conjunction with a study of soil concentrations of various minerals in the soil of Finland and Norway.  There's a slide set of the results along with output plots from R Filzmoser.pdf   .  There is a text document that gives English language descriptions of the elements of the mvoutlier package MVoutliers.pdf , and there's the standard R-documentation which includes the data sets and example code to generate the graphs in the preceding two pdf's mvoutlier-r-Documentation.pdf .

 

The original paper deriving the technique of using SVM for one-class classification is a microsoft technical report.  tr-99-87.pdf

 

Here are slides for Introduction to Data Mining by Tan et.al. Chapter 10 :chap10_anomaly_detection(1).pdf

 

Here's the package documentation for the extremevalues package.  extremevalues.pdf

 

R-examples:

 

mvOutlier:  mvOutlier.R

One Class SVM: oneClassSVM.R

extreme values: extremeValues.R

convex hull: convexhull.R

Comments (0)

You don't have permission to comment on this page.