Header menu link for other important links
X
A data mining approach to choosing categorical attributes for ranked lists
, Sebastian Michel
Published in OpenProceedings.org
2016
Volume: 2016-March
   
Pages: 664 - 665
Abstract
This work proposes and evaluates a novel approach to determine interesting category for ranked lists using $\nu$-SVM. We identify three characteristics (features), entropy, unlikability and peculiarity and show how to train a classifier on these features using a set of Wikipedia tables. The learned model is evaluated by relevance assessments obtained through a user study, reflecting the correctness of our approach.
About the journal
JournalAdvances in Database Technology - EDBT
PublisherOpenProceedings.org
ISSN23672005
Open AccessNo