/outlier-detect

Code that implements the novel outlier detection algorithms from my Ph.D. dissertation.

Primary LanguagePythonOtherNOASSERTION

outlier-detect (http://github.com/benb111/outlier-detect)
Ben Birnbaum (benjamin.birnbaum@gmail.com)
Licensed under Apache (see LICENSE)

Python implementions of the Multinomial Model Algorithm (MMA) and the s-Value
Algorithm (SVA), as described in

B. Birnbaum, B. DeRenzi, A. D. Flaxman, and N. Lesh.  Automated quality control
for mobile data collection. In DEV ’12, pages 1:1–1:10, 2012.

B. Birnbaum. Algorithmic approaches to detecting interviewer fabrication in
surveys.  Ph.D. Dissertation, Univeristy of Washington, Department of Computer
Science and Engineering, 2012.

(See http://bbirnbaum.com/pubs.html for PDF versions of
these papers.)

To use, copy outlierdetect.py into the directory with your code, and import
outlierdetect.  See the module documentation in outlierdetect.py and example.py
for more details.

Switch to the 'display' branch on git to use experimental visualization code in
outlierdetect.