Abstract : In the context of web-scale taxonomies such as Directory Mozilla(www.dmoz.org), previous works have shown the existence of power law distribution in the size of the categories for every level in the taxonomy. In this work, we analyse how such high-level semantics can be leveraged to evaluate accuracy of hierarchical classifiers which automati- cally assign the unseen documents to leaf-level categories. The proposed method offers computational advantages over k-fold cross-validation.
Type de document :
Communication dans un congrès
The Semantic Web: ESWC 2013 Satellite Events, May 2013, France. pp.310-311, 2013
https://hal.archives-ouvertes.fr/hal-00944212
Contributeur : Rohit Babbar
<>
Soumis le : lundi 10 février 2014 - 12:50:32
Dernière modification le : jeudi 11 octobre 2018 - 08:48:04
Massih-Reza Amini, Rohit Babbar, Eric Gaussier, Ioannis Partalas. Comparative Classifier Evaluation for Web-scale Taxonomies using Power Law. The Semantic Web: ESWC 2013 Satellite Events, May 2013, France. pp.310-311, 2013. 〈hal-00944212〉
