Supervised Cross-Modal Factor Analysis for Multiple Modal Data Classification

Handle URI:
http://hdl.handle.net/10754/609037
Title:
Supervised Cross-Modal Factor Analysis for Multiple Modal Data Classification
Authors:
Wang, Jingbin; Zhou, Yihua; Duan, Kanghong; Wang, Jim Jing-Yan; Bensmail, Halima
Abstract:
In this paper we study the problem of learning from multiple modal data for purpose of document classification. In this problem, each document is composed two different modals of data, i.e., An image and a text. Cross-modal factor analysis (CFA) has been proposed to project the two different modals of data to a shared data space, so that the classification of a image or a text can be performed directly in this space. A disadvantage of CFA is that it has ignored the supervision information. In this paper, we improve CFA by incorporating the supervision information to represent and classify both image and text modals of documents. We project both image and text data to a shared data space by factor analysis, and then train a class label predictor in the shared space to use the class label information. The factor analysis parameter and the predictor parameter are learned jointly by solving one single objective function. With this objective function, we minimize the distance between the projections of image and text of the same document, and the classification error of the projection measured by hinge loss function. The objective function is optimized by an alternate optimization strategy in an iterative algorithm. Experiments in two different multiple modal document data sets show the advantage of the proposed algorithm over other CFA methods.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Publisher:
Institute of Electrical and Electronics Engineers (IEEE)
Journal:
2015 IEEE International Conference on Systems, Man, and Cybernetics
Conference/Event name:
Systems, Man, and Cybernetics (SMC), 2015 IEEE International Conference on
Issue Date:
9-Oct-2015
DOI:
10.1109/SMC.2015.329
Type:
Conference Paper
Sponsors:
The research reported in this publication was supported by competitive research funding from King Abdullah University of Science and Technology (KAUST), Saudi Arabia.
Additional Links:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7379461
Appears in Collections:
Conference Papers; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorWang, Jingbinen
dc.contributor.authorZhou, Yihuaen
dc.contributor.authorDuan, Kanghongen
dc.contributor.authorWang, Jim Jing-Yanen
dc.contributor.authorBensmail, Halimaen
dc.date.accessioned2016-05-11T09:12:01Zen
dc.date.available2016-05-11T09:12:01Zen
dc.date.issued2015-10-09en
dc.identifier.doi10.1109/SMC.2015.329en
dc.identifier.urihttp://hdl.handle.net/10754/609037en
dc.description.abstractIn this paper we study the problem of learning from multiple modal data for purpose of document classification. In this problem, each document is composed two different modals of data, i.e., An image and a text. Cross-modal factor analysis (CFA) has been proposed to project the two different modals of data to a shared data space, so that the classification of a image or a text can be performed directly in this space. A disadvantage of CFA is that it has ignored the supervision information. In this paper, we improve CFA by incorporating the supervision information to represent and classify both image and text modals of documents. We project both image and text data to a shared data space by factor analysis, and then train a class label predictor in the shared space to use the class label information. The factor analysis parameter and the predictor parameter are learned jointly by solving one single objective function. With this objective function, we minimize the distance between the projections of image and text of the same document, and the classification error of the projection measured by hinge loss function. The objective function is optimized by an alternate optimization strategy in an iterative algorithm. Experiments in two different multiple modal document data sets show the advantage of the proposed algorithm over other CFA methods.en
dc.description.sponsorshipThe research reported in this publication was supported by competitive research funding from King Abdullah University of Science and Technology (KAUST), Saudi Arabia.en
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en
dc.relation.urlhttp://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7379461en
dc.rights(c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.en
dc.subjectCross-modal factor analysisen
dc.subjectMultiple modal learningen
dc.subjectSupervised learningen
dc.titleSupervised Cross-Modal Factor Analysis for Multiple Modal Data Classificationen
dc.typeConference Paperen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.identifier.journal2015 IEEE International Conference on Systems, Man, and Cyberneticsen
dc.conference.date9-12 Oct. 2015en
dc.conference.nameSystems, Man, and Cybernetics (SMC), 2015 IEEE International Conference onen
dc.conference.locationKowloonen
dc.eprint.versionPost-printen
dc.contributor.institutionNational Time Service Center Chinese Academy of Sciences, Xi’ an 710600 , China Graduate University of Chinese Academy of Sciences Beijing 100039, Chinaen
dc.contributor.institutionDepartment of Mechanical Engineering and Mechanics Lehigh University Bethlehem, PA 18015, USAen
dc.contributor.institutionNorth China Sea Marine Technical Support Center, State Oceanic Administration Qingdao 266033, Chinaen
dc.contributor.institutionQatar Computing Research Institute Doha 5825, Qataren
kaust.authorWang, Jim Jing-Yanen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.