MSR logo

pKNN+AL: Probabilistic Nearest Neighbor Classifier with Active Learning

About | Downloads | Documentation | Contact

Latest version: 1.0

About pKNN+AL

pKNN+AL is a Matlab implementation of Probabilistic Nearest Neighbor Classifier algorithm. pKNN+AL also provides routines for active selection of the examples to be labeled. pKNN uses kernel density estimation to specify probability of a point belonging to a particular class. Using the provided class information, the underlying kernel function can be learned using a modification of Bregman's cyclic projection algorithm. Active selection can be performed through a variety of heuristics that use the computed class probability distribution for each unlabeled point.

This software is currently being actively maintained; please check back often for updates. When using this code, please cite pKNN+AL and the relevant paper.

@inproceedings{jain-kapoor-2009,
  author    = {Prateek Jain and
               Ashish Kapoor}
  title     = {Active Learning for Large Multi-class Problems},
  booktitle = {CVPR},
  year      = {2009},
  address   = {Miami, Florida, USA}
  month     = {June}
}

@manual{pknn-al,
  title        = {Probabilistic Nearest Neighbor Classifier with Active Learning},
  author       = {Prateek Jain and Ashish Kapoor},
  organization = {Microsoft Research, Redmond},
  address      = {http://www.cs.utexas.edu/users/pjain/pknn/},
}

Downloads

Source code

Download the latest version (1.0, released 2009-8-05): pknn_al.zip


Documentation

README: How to install and use pKNN+AL.

Contact

If you have any questions, suggestions, or bug reports about this implementation, please contact Prateek Jain (pjain at cs dot utexas dot edu) and Ashish Kapoor (akapoor at microsoft dot com)