Home // UBICOMM 2011, The Fifth International Conference on Mobile Ubiquitous Computing, Systems, Services and Technologies // View article

An Extension of RankBoost for semi-supervised Learning of Ranking Functions

Authors:
Faïza Dammak
Hager Kammoun
Abdelmajid Ben Hamadou

Keywords: learning to rank; ranking functions; semi-supervised learning; RankBoost algorithm.

Abstract:
The purpose of this paper was a semi-supervised learning method of alternatives ranking functions. This method extends the supervised RankBoost algorithm to combines labeled and unlabeled data. RankBoost is a supervised boosting algorithm adapted to the ranking of instances. Previous work on ranking algorithms has focused on supervised learning (i.e. only labeled data is available for training) or semi-supervised learning of instances. We are interested in semi-supervised learning, which has as objective to learn in the presence of a small quantity of labeled data, simultaneously a great quantity of unlabeled data, to generate a ranking method of alternatives. The goal is to understand how combining labeled and unlabeled data may change the ranking behavior, and how RankBoost can with its character inductive improve ranking performance.

Pages: 49 to 54

Copyright: Copyright (c) The Government of Tunisia, 2011. Used by permission to IARIA.

Publication date: November 20, 2011

Published in: conference

ISSN: 2308-4278

ISBN: 978-1-61208-171-7

Location: Lisbon, Portugal

Dates: from November 20, 2011 to November 25, 2011