English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

This paper proposes an inexpensive way to learn an effective dissimilarityfunction to be used for $k$-nearest neighbor ($k$-NN) classification. UnlikeMahalanobis metric learning methods that map both query (unlabeled) objects andlabeled objects to new coordinates by a single transformation, our methodlearns a transformation of labeled objects to new points in the feature spacewhereas query objects are kept in their original coordinates. This method hasseveral advantages over existing distance metric learning methods: (i) Inexperiments with large document and image datasets, it achieves $k$-NNclassification accuracy better than or at least comparable to thestate-of-the-art metric learning methods. (ii) The transformation can belearned efficiently by solving a standard ridge regression problem. Fordocument and image datasets, training is often more than two orders ofmagnitude faster than the fastest metric learning methods tested. This speed-upis also due to the fact that the proposed method eliminates the optimizationover "negative" object pairs, i.e., objects whose class labels are different.(iii) The formulation has a theoretical justification in terms of reducinghubness in data.

A Fast and Easy Regression Technique for k-NN Classification Without Using Negative Pairs