Partially supervised Independent Factor Analysis using soft labels elicited from multiple experts: application to railway track circuit diagnosis
Using a statistical model in a diagnosis task generally requires a large amount of labeled data. When ground truth information is not available, too expensive or difficult to collect, one has to rely on expert knowledge. In this paper, it is proposed to use partial information from domain experts ex...
Gespeichert in:
| Veröffentlicht in: | Soft computing (Berlin, Germany) Germany), 2012-05, Vol.16 (5), p.741-754 |
|---|---|
| Hauptverfasser: | , , , , |
| Format: | Artikel |
| Sprache: | eng |
| Schlagworte: | |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| container_end_page | 754 |
|---|---|
| container_issue | 5 |
| container_start_page | 741 |
| container_title | Soft computing (Berlin, Germany) |
| container_volume | 16 |
| creator | Cherfi, Zohra L. Oukhellou, Latifa Côme, Etienne Denœux, Thierry Aknin, Patrice |
| description | Using a statistical model in a diagnosis task generally requires a large amount of labeled data. When ground truth information is not available, too expensive or difficult to collect, one has to rely on expert knowledge. In this paper, it is proposed to use partial information from domain experts expressed as belief functions. Expert opinions are combined in this framework and used with measurement data to estimate the parameters of a statistical model using a variant of the EM algorithm. The particular application investigated here concerns the diagnosis of railway track circuits. A noiseless Independent Factor Analysis model is postulated, assuming the observed variables extracted from railway track inspection signals to be generated by a linear mixture of independent latent variables linked to the system component states. Usually, learning with this statistical model is performed in an unsupervised way using unlabeled examples only. In this paper, it is proposed to handle this learning process in a soft-supervised way using imperfect information on the system component states. Fusing partially reliable information about cluster membership is shown to significantly improve classification results. |
| doi_str_mv | 10.1007/s00500-011-0766-4 |
| format | Article |
| fullrecord | <record><control><sourceid>proquest_hal_p</sourceid><recordid>TN_cdi_hal_primary_oai_HAL_hal_00688783v1</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2918050813</sourcerecordid><originalsourceid>FETCH-LOGICAL-c393t-3549d1f196116b30bd4b11ee7f5b11a648424b8f2388eef1764c6b75f0a1a13f3</originalsourceid><addsrcrecordid>eNp1kcFu1DAQhiMEEqXwANwsceIQ8MRO7HBbVZRWWgkOcLYmib24eOPgcUr3MXhjvATBicvMyPr-f-T5q-ol8DfAuXpLnLec1xyg5qrravmougApRK2k6h__nptadVI8rZ4R3XHegGrFRfXzE6bsMYQTo3Wx6d6TndjtPNnFljJndo1jjontZgwn8sRW8vOBUXSZBRxsIGaDH30uMpfikR3XkP0SLLMPxS_TO4bLUgjMPs4sR5bQhx94Yjnh-I2NPo2rz2zyeJhjWfC8euIwkH3xp19WX67ff766qfcfP9xe7fb1KHqRa9HKfgIHfQfQDYIPkxwArFWuLR07qWUjB-0aobW1Dsrfx25QreMICMKJy-r15vsVg1mSP2I6mYje3Oz25vzGeae10uIeCvtqY5cUv6-WsrmLayoXIdP0oMvpNYhCwUaNKRIl6_7aAjfnlMyWkikpmXNKRhZNs2mosPPBpn_O_xf9Ap2Mlww</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2918050813</pqid></control><display><type>article</type><title>Partially supervised Independent Factor Analysis using soft labels elicited from multiple experts: application to railway track circuit diagnosis</title><source>ProQuest Central UK/Ireland</source><source>SpringerLink Journals - AutoHoldings</source><source>ProQuest Central</source><creator>Cherfi, Zohra L. ; Oukhellou, Latifa ; Côme, Etienne ; Denœux, Thierry ; Aknin, Patrice</creator><creatorcontrib>Cherfi, Zohra L. ; Oukhellou, Latifa ; Côme, Etienne ; Denœux, Thierry ; Aknin, Patrice</creatorcontrib><description>Using a statistical model in a diagnosis task generally requires a large amount of labeled data. When ground truth information is not available, too expensive or difficult to collect, one has to rely on expert knowledge. In this paper, it is proposed to use partial information from domain experts expressed as belief functions. Expert opinions are combined in this framework and used with measurement data to estimate the parameters of a statistical model using a variant of the EM algorithm. The particular application investigated here concerns the diagnosis of railway track circuits. A noiseless Independent Factor Analysis model is postulated, assuming the observed variables extracted from railway track inspection signals to be generated by a linear mixture of independent latent variables linked to the system component states. Usually, learning with this statistical model is performed in an unsupervised way using unlabeled examples only. In this paper, it is proposed to handle this learning process in a soft-supervised way using imperfect information on the system component states. Fusing partially reliable information about cluster membership is shown to significantly improve classification results.</description><identifier>ISSN: 1432-7643</identifier><identifier>EISSN: 1433-7479</identifier><identifier>DOI: 10.1007/s00500-011-0766-4</identifier><language>eng</language><publisher>Berlin/Heidelberg: Springer-Verlag</publisher><subject>Algorithms ; Artificial Intelligence ; Computational Intelligence ; Computer Science ; Control ; Diagnosis ; Dogmatism ; Engineering ; Epistemology ; Factor analysis ; Fault diagnosis ; Focus ; Hypotheses ; Independent variables ; Labeling ; Learning ; Machine learning ; Mathematical Logic and Foundations ; Mechatronics ; Preventive maintenance ; Railway tracks ; Robotics ; Statistical models ; Track circuits ; Variables</subject><ispartof>Soft computing (Berlin, Germany), 2012-05, Vol.16 (5), p.741-754</ispartof><rights>Springer-Verlag 2011</rights><rights>Springer-Verlag 2011.</rights><rights>Distributed under a Creative Commons Attribution 4.0 International License</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><citedby>FETCH-LOGICAL-c393t-3549d1f196116b30bd4b11ee7f5b11a648424b8f2388eef1764c6b75f0a1a13f3</citedby><cites>FETCH-LOGICAL-c393t-3549d1f196116b30bd4b11ee7f5b11a648424b8f2388eef1764c6b75f0a1a13f3</cites><orcidid>0000-0002-0660-5436 ; 0000-0002-0459-6388 ; 0000-0002-5193-1732 ; 0000-0003-3557-0225</orcidid></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/s00500-011-0766-4$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://www.proquest.com/docview/2918050813?pq-origsite=primo$$EHTML$$P50$$Gproquest$$H</linktohtml><link.rule.ids>230,314,780,784,885,21387,27923,27924,33743,41487,42556,43804,51318,64384,64388,72240</link.rule.ids><backlink>$$Uhttps://hal.science/hal-00688783$$DView record in HAL$$Hfree_for_read</backlink></links><search><creatorcontrib>Cherfi, Zohra L.</creatorcontrib><creatorcontrib>Oukhellou, Latifa</creatorcontrib><creatorcontrib>Côme, Etienne</creatorcontrib><creatorcontrib>Denœux, Thierry</creatorcontrib><creatorcontrib>Aknin, Patrice</creatorcontrib><title>Partially supervised Independent Factor Analysis using soft labels elicited from multiple experts: application to railway track circuit diagnosis</title><title>Soft computing (Berlin, Germany)</title><addtitle>Soft Comput</addtitle><description>Using a statistical model in a diagnosis task generally requires a large amount of labeled data. When ground truth information is not available, too expensive or difficult to collect, one has to rely on expert knowledge. In this paper, it is proposed to use partial information from domain experts expressed as belief functions. Expert opinions are combined in this framework and used with measurement data to estimate the parameters of a statistical model using a variant of the EM algorithm. The particular application investigated here concerns the diagnosis of railway track circuits. A noiseless Independent Factor Analysis model is postulated, assuming the observed variables extracted from railway track inspection signals to be generated by a linear mixture of independent latent variables linked to the system component states. Usually, learning with this statistical model is performed in an unsupervised way using unlabeled examples only. In this paper, it is proposed to handle this learning process in a soft-supervised way using imperfect information on the system component states. Fusing partially reliable information about cluster membership is shown to significantly improve classification results.</description><subject>Algorithms</subject><subject>Artificial Intelligence</subject><subject>Computational Intelligence</subject><subject>Computer Science</subject><subject>Control</subject><subject>Diagnosis</subject><subject>Dogmatism</subject><subject>Engineering</subject><subject>Epistemology</subject><subject>Factor analysis</subject><subject>Fault diagnosis</subject><subject>Focus</subject><subject>Hypotheses</subject><subject>Independent variables</subject><subject>Labeling</subject><subject>Learning</subject><subject>Machine learning</subject><subject>Mathematical Logic and Foundations</subject><subject>Mechatronics</subject><subject>Preventive maintenance</subject><subject>Railway tracks</subject><subject>Robotics</subject><subject>Statistical models</subject><subject>Track circuits</subject><subject>Variables</subject><issn>1432-7643</issn><issn>1433-7479</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2012</creationdate><recordtype>article</recordtype><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><sourceid>GNUQQ</sourceid><recordid>eNp1kcFu1DAQhiMEEqXwANwsceIQ8MRO7HBbVZRWWgkOcLYmib24eOPgcUr3MXhjvATBicvMyPr-f-T5q-ol8DfAuXpLnLec1xyg5qrravmougApRK2k6h__nptadVI8rZ4R3XHegGrFRfXzE6bsMYQTo3Wx6d6TndjtPNnFljJndo1jjontZgwn8sRW8vOBUXSZBRxsIGaDH30uMpfikR3XkP0SLLMPxS_TO4bLUgjMPs4sR5bQhx94Yjnh-I2NPo2rz2zyeJhjWfC8euIwkH3xp19WX67ff766qfcfP9xe7fb1KHqRa9HKfgIHfQfQDYIPkxwArFWuLR07qWUjB-0aobW1Dsrfx25QreMICMKJy-r15vsVg1mSP2I6mYje3Oz25vzGeae10uIeCvtqY5cUv6-WsrmLayoXIdP0oMvpNYhCwUaNKRIl6_7aAjfnlMyWkikpmXNKRhZNs2mosPPBpn_O_xf9Ap2Mlww</recordid><startdate>20120501</startdate><enddate>20120501</enddate><creator>Cherfi, Zohra L.</creator><creator>Oukhellou, Latifa</creator><creator>Côme, Etienne</creator><creator>Denœux, Thierry</creator><creator>Aknin, Patrice</creator><general>Springer-Verlag</general><general>Springer Nature B.V</general><general>Springer Verlag</general><scope>AAYXX</scope><scope>CITATION</scope><scope>8FE</scope><scope>8FG</scope><scope>AFKRA</scope><scope>ARAPS</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>GNUQQ</scope><scope>HCIFZ</scope><scope>JQ2</scope><scope>K7-</scope><scope>P5Z</scope><scope>P62</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>1XC</scope><scope>VOOES</scope><orcidid>https://orcid.org/0000-0002-0660-5436</orcidid><orcidid>https://orcid.org/0000-0002-0459-6388</orcidid><orcidid>https://orcid.org/0000-0002-5193-1732</orcidid><orcidid>https://orcid.org/0000-0003-3557-0225</orcidid></search><sort><creationdate>20120501</creationdate><title>Partially supervised Independent Factor Analysis using soft labels elicited from multiple experts: application to railway track circuit diagnosis</title><author>Cherfi, Zohra L. ; Oukhellou, Latifa ; Côme, Etienne ; Denœux, Thierry ; Aknin, Patrice</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-c393t-3549d1f196116b30bd4b11ee7f5b11a648424b8f2388eef1764c6b75f0a1a13f3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2012</creationdate><topic>Algorithms</topic><topic>Artificial Intelligence</topic><topic>Computational Intelligence</topic><topic>Computer Science</topic><topic>Control</topic><topic>Diagnosis</topic><topic>Dogmatism</topic><topic>Engineering</topic><topic>Epistemology</topic><topic>Factor analysis</topic><topic>Fault diagnosis</topic><topic>Focus</topic><topic>Hypotheses</topic><topic>Independent variables</topic><topic>Labeling</topic><topic>Learning</topic><topic>Machine learning</topic><topic>Mathematical Logic and Foundations</topic><topic>Mechatronics</topic><topic>Preventive maintenance</topic><topic>Railway tracks</topic><topic>Robotics</topic><topic>Statistical models</topic><topic>Track circuits</topic><topic>Variables</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Cherfi, Zohra L.</creatorcontrib><creatorcontrib>Oukhellou, Latifa</creatorcontrib><creatorcontrib>Côme, Etienne</creatorcontrib><creatorcontrib>Denœux, Thierry</creatorcontrib><creatorcontrib>Aknin, Patrice</creatorcontrib><collection>CrossRef</collection><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>ProQuest Central UK/Ireland</collection><collection>Advanced Technologies & Aerospace Collection</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>ProQuest Central Student</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Computer Science Collection</collection><collection>Computer Science Database</collection><collection>Advanced Technologies & Aerospace Database</collection><collection>ProQuest Advanced Technologies & Aerospace Collection</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Hyper Article en Ligne (HAL)</collection><collection>Hyper Article en Ligne (HAL) (Open Access)</collection><jtitle>Soft computing (Berlin, Germany)</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Cherfi, Zohra L.</au><au>Oukhellou, Latifa</au><au>Côme, Etienne</au><au>Denœux, Thierry</au><au>Aknin, Patrice</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Partially supervised Independent Factor Analysis using soft labels elicited from multiple experts: application to railway track circuit diagnosis</atitle><jtitle>Soft computing (Berlin, Germany)</jtitle><stitle>Soft Comput</stitle><date>2012-05-01</date><risdate>2012</risdate><volume>16</volume><issue>5</issue><spage>741</spage><epage>754</epage><pages>741-754</pages><issn>1432-7643</issn><eissn>1433-7479</eissn><abstract>Using a statistical model in a diagnosis task generally requires a large amount of labeled data. When ground truth information is not available, too expensive or difficult to collect, one has to rely on expert knowledge. In this paper, it is proposed to use partial information from domain experts expressed as belief functions. Expert opinions are combined in this framework and used with measurement data to estimate the parameters of a statistical model using a variant of the EM algorithm. The particular application investigated here concerns the diagnosis of railway track circuits. A noiseless Independent Factor Analysis model is postulated, assuming the observed variables extracted from railway track inspection signals to be generated by a linear mixture of independent latent variables linked to the system component states. Usually, learning with this statistical model is performed in an unsupervised way using unlabeled examples only. In this paper, it is proposed to handle this learning process in a soft-supervised way using imperfect information on the system component states. Fusing partially reliable information about cluster membership is shown to significantly improve classification results.</abstract><cop>Berlin/Heidelberg</cop><pub>Springer-Verlag</pub><doi>10.1007/s00500-011-0766-4</doi><tpages>14</tpages><orcidid>https://orcid.org/0000-0002-0660-5436</orcidid><orcidid>https://orcid.org/0000-0002-0459-6388</orcidid><orcidid>https://orcid.org/0000-0002-5193-1732</orcidid><orcidid>https://orcid.org/0000-0003-3557-0225</orcidid><oa>free_for_read</oa></addata></record> |
| fulltext | fulltext |
| identifier | ISSN: 1432-7643 |
| ispartof | Soft computing (Berlin, Germany), 2012-05, Vol.16 (5), p.741-754 |
| issn | 1432-7643 1433-7479 |
| language | eng |
| recordid | cdi_hal_primary_oai_HAL_hal_00688783v1 |
| source | ProQuest Central UK/Ireland; SpringerLink Journals - AutoHoldings; ProQuest Central |
| subjects | Algorithms Artificial Intelligence Computational Intelligence Computer Science Control Diagnosis Dogmatism Engineering Epistemology Factor analysis Fault diagnosis Focus Hypotheses Independent variables Labeling Learning Machine learning Mathematical Logic and Foundations Mechatronics Preventive maintenance Railway tracks Robotics Statistical models Track circuits Variables |
| title | Partially supervised Independent Factor Analysis using soft labels elicited from multiple experts: application to railway track circuit diagnosis |
| url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-12T10%3A46%3A53IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_hal_p&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Partially%20supervised%20Independent%20Factor%20Analysis%20using%20soft%20labels%20elicited%20from%20multiple%20experts:%20application%20to%20railway%20track%20circuit%20diagnosis&rft.jtitle=Soft%20computing%20(Berlin,%20Germany)&rft.au=Cherfi,%20Zohra%20L.&rft.date=2012-05-01&rft.volume=16&rft.issue=5&rft.spage=741&rft.epage=754&rft.pages=741-754&rft.issn=1432-7643&rft.eissn=1433-7479&rft_id=info:doi/10.1007/s00500-011-0766-4&rft_dat=%3Cproquest_hal_p%3E2918050813%3C/proquest_hal_p%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2918050813&rft_id=info:pmid/&rfr_iscdi=true |