PROXIMITY-BASED GRAPH EMBEDDINGS FOR MULTI-LABEL CLASSIFICATION

Tingting Mu, Sophia Ananiadou

2010

Abstract

In many real applications of text mining, information retrieval and natural language processing, large-scale features are frequently used, which often make the employed machine learning algorithms intractable, leading to the well-known problem “curse of dimensionality”. Aiming at not only removing the redundant information from the original features but also improving their discriminating ability, we present a novel approach on supervised generation of low-dimensional, proximity-based, graph embeddings to facilitate multi-label classification. The optimal embeddings are computed from a supervised adjacency graph, called multi-label graph, which simultaneously preserves proximity structures between samples constructed based on feature and multi-label class information. We propose different ways to obtain this multi-label graph, by either working in a binary label space or a projected real label space. To reduce the training cost in the dimensionality reduction procedure caused by large-scale features, a smaller set of relation features between each sample and a set of representative prototypes are employed. The effectiveness of our proposed method is demonstrated with two document collections for text categorization based on the “bag of words” model.

Paper Citation

in Harvard Style

Mu T. and Ananiadou S. (2010). PROXIMITY-BASED GRAPH EMBEDDINGS FOR MULTI-LABEL CLASSIFICATION . In Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2010) ISBN 978-989-8425-28-7, pages 74-84. DOI: 10.5220/0003092200740084

in Bibtex Style

@conference{kdir10,
author={Tingting Mu and Sophia Ananiadou},
title={PROXIMITY-BASED GRAPH EMBEDDINGS FOR MULTI-LABEL CLASSIFICATION},
booktitle={Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2010)},
year={2010},
pages={74-84},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003092200740084},
isbn={978-989-8425-28-7},
}

in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Knowledge Discovery and Information Retrieval - Volume 1: KDIR, (IC3K 2010)
TI - PROXIMITY-BASED GRAPH EMBEDDINGS FOR MULTI-LABEL CLASSIFICATION
SN - 978-989-8425-28-7
AU - Mu T.
AU - Ananiadou S.
PY - 2010
SP - 74
EP - 84
DO - 10.5220/0003092200740084