@techreport{TR-IC-07-13,
  number = {IC-07-13},
  author = {J.~P.~Papa and A. X. Falcão and
            P.~A.~V. Miranda and C.~T.~N.~Suzuki and
            N.~D.~A.~Mascarenhas},
  title = {A New Pattern Classifier based on Optimum Path Forest},
  month = {May},
  year = {2007}, 
  institution = {Institute of Computing, University of Campinas},
  note = {In English, 12 pages.
    \par\selectlanguage{english}\textbf{Abstract}
       We introduce a supervised pattern classifier based on
       \emph{optimum path forest}. The samples in a training set are
       nodes of a complete graph, whose arcs are weighted by the
       distances between sample feature vectors. The training builds a
       classifier from key samples (prototypes) of all classes, where
       each prototype defines an optimum path tree whose nodes are its
       strongest connected samples.  The optimum paths are also
       considered to label unseen test samples with the classes of
       their strongest connected prototypes. We show how to find
       prototypes with none classification errors in the training set
       and propose a learning algorithm to improve accuracy over an
       evaluation set without overfitting the test set.  The method is
       robust to outliers, handles non-separable classes, and can
       outperform artificial neural networks and support vector
       machines.
  }
}