jjtan / outlierdetection Goto Github PK

Name: JJ Tan
Comp 135 - Final Project: Outlier detection
December 20, 2012

//////// Relevant Files ////////
General Documentation: README
Code: arff.py space.py kMeans.py custom.py
Custom Method: custom.txt SyntheticTests.png
Custom Method Test: synthetic1.arff synthetic2.arff data_gen.py
Results: k-means-dataset1.dat k-means-dataset2.dat cod-dataset1.dat
         cod-dataset2.dat

//////// Running the code ////////

Using Python 2.7.3

Part 1:
python kMeans.py Dataset1-outlier.arff Dataset2-outlier.arff

Part 2:
python custom.py Dataset1-outlier.arff Dataset2-outlier.arff
(Note: The order of the arff files is important. I have altered the parameters
to account for the larger set in the 2nd file.)

*For explanation of the custom algorithm see custom.txt