A Harol

StatisticallearningalgorithmsoftenrelyontheEuclideandistance.Inpractice,non-Euclideanornon-metricdissimilaritymeasuresmayarisewhencontours,spectraorshapesarecomparedbyeditdistancesorasaconsequenceofrobustobjectmatching[1,2].Itisanopenissuewhethersuchmeasuresareadvantageousforstat ...

Augmented embedding of dissimilarity data into (pseudo-) Euclidean spaces

Conference paper (2006) - A Harol (author), E.M. Pekalska (author), S Verzakov (author), Bob Duin (author), R.P.W. Duin (author)

Augmented embedding of dissimilarity data into (pseudo-)Euclidean spaces

Conference paper (2006) - A Harol (author), E.M. Pekalska (author), S Verzakov (author), Bob Duin (author), R.P.W. Duin (author)

Pairwiseproximitiesdescribethepropertiesofobjectsintermsoftheirsimilarities.Byusingdi¿erentdistance-basedfunctionsonemayencodedi¿erentcharacteristicsofagivenproblem.However,tousetheframeworkofstatisticalpatternrecognitionsomevectorrepresentationshouldbeconstructed.Oneofthesimples ...

Pairwise selection of features and prototypes

Conference paper (2005) - E.M. Pekalska (author), A Harol (author), Bob Duin (author), R.P.W. Duin (author)

A new model for prototype selection on distance data

Conference paper (2005) - A Harol (author), Bob Duin (author), R.P.W. Duin (author), Marcel Reinders (author), M.J.T. Reinders (author), Marcel J.T. Reinders (author), M. J.T. Reinders (author)

On euclidean correction for non-euclidean dissimilarities

Pairwise feature evaluation for constructing reduced representations

Non-Euclidean or non-metric measures can be informative

Augmented embedding of dissimilarity data into (pseudo-) Euclidean spaces

Augmented embedding of dissimilarity data into (pseudo-)Euclidean spaces

Pairwise selection of features and prototypes

A new model for prototype selection on distance data