User-Based Active Learning

Active learning has been proven a reliable strategy to reduce manual efforts in training data labeling. Such strategies incorporate the user as oracle: the classifier selects the most appropriate example and the user provides the label. While this approach is tailored towards the classifier, more in...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Seifert, C, Granitzer, M
Format: Tagungsbericht
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page 425
container_issue
container_start_page 418
container_title
container_volume
creator Seifert, C
Granitzer, M
description Active learning has been proven a reliable strategy to reduce manual efforts in training data labeling. Such strategies incorporate the user as oracle: the classifier selects the most appropriate example and the user provides the label. While this approach is tailored towards the classifier, more intelligent input from the user may be beneficial. For instance, given only one example at a time users are hardly able to determine whether this example is an outlier or not. In this paper we propose user-based visually-supported active learning strategies that allow the user to do both, selecting and labeling examples given a trained classifier. While labeling is straightforward, selection takes place using a interactive visualization of the classifier's a-posteriori output probabilities. By simulating different user selection strategies we show, that user-based active learning outperforms uncertainty based sampling methods and yields a more robust approach on different data sets. The obtained results point towards the potential of combining active learning strategies with results from the field of information visualization.
doi_str_mv 10.1109/ICDMW.2010.181
format Conference Proceeding
fullrecord <record><control><sourceid>ieee_6IE</sourceid><recordid>TN_cdi_ieee_primary_5693328</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>5693328</ieee_id><sourcerecordid>5693328</sourcerecordid><originalsourceid>FETCH-LOGICAL-i90t-d1d5e6b72ec9cda81dbf2842051d53d4971c00a9be7770532ff99f28960546043</originalsourceid><addsrcrecordid>eNo9jUtLA0EQhMcXGOJePeglf2CTnp7pme1jXB8JrHiJeAyzO70yokF2g-C_d3xgQdEUX9Gl1LmGudbAi3V9ff80R_jOlT5QBfsKvGOySN4fqgkaTyUj8dEP0xat5Ww8_mcGT1Uxji-QRei9p4m6eBxlKK_CKHG27PbpQ2aNhGGXds9n6qQPr6MUf3eqNrc3m3pVNg9363rZlIlhX0YdSVzrUTruYqh0bHusLAJlYKJlrzuAwK3kQSCDfc-cG-yArANrpury920Ske37kN7C8Lklx8ZgZb4ABec-yw</addsrcrecordid><sourcetype>Publisher</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype></control><display><type>conference_proceeding</type><title>User-Based Active Learning</title><source>IEEE Electronic Library (IEL) Conference Proceedings</source><creator>Seifert, C ; Granitzer, M</creator><creatorcontrib>Seifert, C ; Granitzer, M</creatorcontrib><description>Active learning has been proven a reliable strategy to reduce manual efforts in training data labeling. Such strategies incorporate the user as oracle: the classifier selects the most appropriate example and the user provides the label. While this approach is tailored towards the classifier, more intelligent input from the user may be beneficial. For instance, given only one example at a time users are hardly able to determine whether this example is an outlier or not. In this paper we propose user-based visually-supported active learning strategies that allow the user to do both, selecting and labeling examples given a trained classifier. While labeling is straightforward, selection takes place using a interactive visualization of the classifier's a-posteriori output probabilities. By simulating different user selection strategies we show, that user-based active learning outperforms uncertainty based sampling methods and yields a more robust approach on different data sets. The obtained results point towards the potential of combining active learning strategies with results from the field of information visualization.</description><identifier>ISSN: 2375-9232</identifier><identifier>ISBN: 9781424492442</identifier><identifier>ISBN: 1424492440</identifier><identifier>EISSN: 2375-9259</identifier><identifier>EISBN: 9780769542577</identifier><identifier>EISBN: 0769542573</identifier><identifier>DOI: 10.1109/ICDMW.2010.181</identifier><language>eng</language><publisher>IEEE</publisher><subject>active learning ; Data models ; Data visualization ; Entropy ; information visualization ; Labeling ; Training ; Uncertainty ; user behavior ; Visualization</subject><ispartof>2010 IEEE International Conference on Data Mining Workshops, 2010, p.418-425</ispartof><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktohtml>$$Uhttps://ieeexplore.ieee.org/document/5693328$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>310,311,781,785,790,791,2059,27930,54925</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/5693328$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Seifert, C</creatorcontrib><creatorcontrib>Granitzer, M</creatorcontrib><title>User-Based Active Learning</title><title>2010 IEEE International Conference on Data Mining Workshops</title><addtitle>icdmw</addtitle><description>Active learning has been proven a reliable strategy to reduce manual efforts in training data labeling. Such strategies incorporate the user as oracle: the classifier selects the most appropriate example and the user provides the label. While this approach is tailored towards the classifier, more intelligent input from the user may be beneficial. For instance, given only one example at a time users are hardly able to determine whether this example is an outlier or not. In this paper we propose user-based visually-supported active learning strategies that allow the user to do both, selecting and labeling examples given a trained classifier. While labeling is straightforward, selection takes place using a interactive visualization of the classifier's a-posteriori output probabilities. By simulating different user selection strategies we show, that user-based active learning outperforms uncertainty based sampling methods and yields a more robust approach on different data sets. The obtained results point towards the potential of combining active learning strategies with results from the field of information visualization.</description><subject>active learning</subject><subject>Data models</subject><subject>Data visualization</subject><subject>Entropy</subject><subject>information visualization</subject><subject>Labeling</subject><subject>Training</subject><subject>Uncertainty</subject><subject>user behavior</subject><subject>Visualization</subject><issn>2375-9232</issn><issn>2375-9259</issn><isbn>9781424492442</isbn><isbn>1424492440</isbn><isbn>9780769542577</isbn><isbn>0769542573</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2010</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><sourceid>RIE</sourceid><recordid>eNo9jUtLA0EQhMcXGOJePeglf2CTnp7pme1jXB8JrHiJeAyzO70yokF2g-C_d3xgQdEUX9Gl1LmGudbAi3V9ff80R_jOlT5QBfsKvGOySN4fqgkaTyUj8dEP0xat5Ww8_mcGT1Uxji-QRei9p4m6eBxlKK_CKHG27PbpQ2aNhGGXds9n6qQPr6MUf3eqNrc3m3pVNg9363rZlIlhX0YdSVzrUTruYqh0bHusLAJlYKJlrzuAwK3kQSCDfc-cG-yArANrpury920Ske37kN7C8Lklx8ZgZb4ABec-yw</recordid><startdate>201012</startdate><enddate>201012</enddate><creator>Seifert, C</creator><creator>Granitzer, M</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope></search><sort><creationdate>201012</creationdate><title>User-Based Active Learning</title><author>Seifert, C ; Granitzer, M</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i90t-d1d5e6b72ec9cda81dbf2842051d53d4971c00a9be7770532ff99f28960546043</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2010</creationdate><topic>active learning</topic><topic>Data models</topic><topic>Data visualization</topic><topic>Entropy</topic><topic>information visualization</topic><topic>Labeling</topic><topic>Training</topic><topic>Uncertainty</topic><topic>user behavior</topic><topic>Visualization</topic><toplevel>online_resources</toplevel><creatorcontrib>Seifert, C</creatorcontrib><creatorcontrib>Granitzer, M</creatorcontrib><collection>IEEE Electronic Library (IEL) Conference Proceedings</collection><collection>IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume</collection><collection>IEEE Xplore All Conference Proceedings</collection><collection>IEEE Electronic Library Online</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Seifert, C</au><au>Granitzer, M</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>User-Based Active Learning</atitle><btitle>2010 IEEE International Conference on Data Mining Workshops</btitle><stitle>icdmw</stitle><date>2010-12</date><risdate>2010</risdate><spage>418</spage><epage>425</epage><pages>418-425</pages><issn>2375-9232</issn><eissn>2375-9259</eissn><isbn>9781424492442</isbn><isbn>1424492440</isbn><eisbn>9780769542577</eisbn><eisbn>0769542573</eisbn><abstract>Active learning has been proven a reliable strategy to reduce manual efforts in training data labeling. Such strategies incorporate the user as oracle: the classifier selects the most appropriate example and the user provides the label. While this approach is tailored towards the classifier, more intelligent input from the user may be beneficial. For instance, given only one example at a time users are hardly able to determine whether this example is an outlier or not. In this paper we propose user-based visually-supported active learning strategies that allow the user to do both, selecting and labeling examples given a trained classifier. While labeling is straightforward, selection takes place using a interactive visualization of the classifier's a-posteriori output probabilities. By simulating different user selection strategies we show, that user-based active learning outperforms uncertainty based sampling methods and yields a more robust approach on different data sets. The obtained results point towards the potential of combining active learning strategies with results from the field of information visualization.</abstract><pub>IEEE</pub><doi>10.1109/ICDMW.2010.181</doi><tpages>8</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 2375-9232
ispartof 2010 IEEE International Conference on Data Mining Workshops, 2010, p.418-425
issn 2375-9232
2375-9259
language eng
recordid cdi_ieee_primary_5693328
source IEEE Electronic Library (IEL) Conference Proceedings
subjects active learning
Data models
Data visualization
Entropy
information visualization
Labeling
Training
Uncertainty
user behavior
Visualization
title User-Based Active Learning
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-10T18%3A04%3A33IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-ieee_6IE&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=User-Based%20Active%20Learning&rft.btitle=2010%20IEEE%20International%20Conference%20on%20Data%20Mining%20Workshops&rft.au=Seifert,%20C&rft.date=2010-12&rft.spage=418&rft.epage=425&rft.pages=418-425&rft.issn=2375-9232&rft.eissn=2375-9259&rft.isbn=9781424492442&rft.isbn_list=1424492440&rft_id=info:doi/10.1109/ICDMW.2010.181&rft_dat=%3Cieee_6IE%3E5693328%3C/ieee_6IE%3E%3Curl%3E%3C/url%3E&rft.eisbn=9780769542577&rft.eisbn_list=0769542573&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rft_ieee_id=5693328&rfr_iscdi=true