Handling ambiguity via input-output kernel learning

Data ambiguities exist in many data mining and machine learning applications such as text categorization and image retrieval. For instance, it is generally beneficial to utilize the ambiguous unlabeled documents to learn a more robust classifier for text categorization under the semi-supervised lear...

Full description

Saved in:
Bibliographic Details
Main Authors: Xu, Xinxing, Tsang, Ivor Wai-Hung, Xu, Dong
Other Authors: School of Computer Engineering
Format: Conference or Workshop Item
Language:English
Published: 2013
Subjects:
Online Access:https://hdl.handle.net/10356/99740
http://hdl.handle.net/10220/13014
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-99740
record_format dspace
spelling sg-ntu-dr.10356-997402020-05-28T07:17:50Z Handling ambiguity via input-output kernel learning Xu, Xinxing Tsang, Ivor Wai-Hung Xu, Dong School of Computer Engineering IEEE International Conference on Data Mining (12th : 2012 : Brussels, Belgium) DRNTU::Engineering::Computer science and engineering Data ambiguities exist in many data mining and machine learning applications such as text categorization and image retrieval. For instance, it is generally beneficial to utilize the ambiguous unlabeled documents to learn a more robust classifier for text categorization under the semi-supervised learning setting. To handle general data ambiguities, we present a unified kernel learning framework named Input-Output Kernel Learning (IOKL). Based on our framework, we further propose a novel soft margin group sparse Multiple Kernel Learning (MKL) formulation by introducing a group kernel slack variable to each group of base input-output kernels. Moreover, an efficient block-wise coordinate descent algorithm with an analytical solution for the kernel combination coefficients is developed to solve the proposed formulation. We conduct comprehensive experiments on benchmark datasets for both semi-supervised learning and multiple instance learning tasks, and also apply our IOKL framework to a computer vision application called text-based image retrieval on the NUS-WIDE dataset. Promising results demonstrate the effectiveness of our proposed IOKL framework. 2013-08-05T06:45:10Z 2019-12-06T20:10:54Z 2013-08-05T06:45:10Z 2019-12-06T20:10:54Z 2012 2012 Conference Paper https://hdl.handle.net/10356/99740 http://hdl.handle.net/10220/13014 10.1109/ICDM.2012.105 en
institution Nanyang Technological University
building NTU Library
country Singapore
collection DR-NTU
language English
topic DRNTU::Engineering::Computer science and engineering
spellingShingle DRNTU::Engineering::Computer science and engineering
Xu, Xinxing
Tsang, Ivor Wai-Hung
Xu, Dong
Handling ambiguity via input-output kernel learning
description Data ambiguities exist in many data mining and machine learning applications such as text categorization and image retrieval. For instance, it is generally beneficial to utilize the ambiguous unlabeled documents to learn a more robust classifier for text categorization under the semi-supervised learning setting. To handle general data ambiguities, we present a unified kernel learning framework named Input-Output Kernel Learning (IOKL). Based on our framework, we further propose a novel soft margin group sparse Multiple Kernel Learning (MKL) formulation by introducing a group kernel slack variable to each group of base input-output kernels. Moreover, an efficient block-wise coordinate descent algorithm with an analytical solution for the kernel combination coefficients is developed to solve the proposed formulation. We conduct comprehensive experiments on benchmark datasets for both semi-supervised learning and multiple instance learning tasks, and also apply our IOKL framework to a computer vision application called text-based image retrieval on the NUS-WIDE dataset. Promising results demonstrate the effectiveness of our proposed IOKL framework.
author2 School of Computer Engineering
author_facet School of Computer Engineering
Xu, Xinxing
Tsang, Ivor Wai-Hung
Xu, Dong
format Conference or Workshop Item
author Xu, Xinxing
Tsang, Ivor Wai-Hung
Xu, Dong
author_sort Xu, Xinxing
title Handling ambiguity via input-output kernel learning
title_short Handling ambiguity via input-output kernel learning
title_full Handling ambiguity via input-output kernel learning
title_fullStr Handling ambiguity via input-output kernel learning
title_full_unstemmed Handling ambiguity via input-output kernel learning
title_sort handling ambiguity via input-output kernel learning
publishDate 2013
url https://hdl.handle.net/10356/99740
http://hdl.handle.net/10220/13014
_version_ 1681059127131897856