TildeCRF: Conditional Random Fields for Logical Sequences

Conditional Random Fields (CRFs) provide a powerful instrument for labeling sequences. So far, however, CRFs have only been considered for labeling sequences over flat alphabets. In this paper, we describe TildeCRF, the first method for training CRFs on logical sequences, i.e., sequences over an alp...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Hauptverfasser:	Gutmann, Bernd, Kersting, Kristian
Format:	Buchkapitel
Sprache:	eng
Schlagworte:	Applied sciences Artificial intelligence Computer science control theory systems Conditional Random Field Exact sciences and technology Ground Atom Hide Markov Model Logical Sequence Regression Tree
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

container_end_page	185
container_issue
container_start_page	174
container_title
container_volume
creator	Gutmann, Bernd Kersting, Kristian
description	Conditional Random Fields (CRFs) provide a powerful instrument for labeling sequences. So far, however, CRFs have only been considered for labeling sequences over flat alphabets. In this paper, we describe TildeCRF, the first method for training CRFs on logical sequences, i.e., sequences over an alphabet of logical atoms. TildeCRF’s key idea is to use relational regression trees in Dietterich et al.’s gradient tree boosting approach. Thus, the CRF potential functions are represented as weighted sums of relational regression trees. Experiments show a significant improvement over established results achieved with hidden Markov models and Fisher kernels for logical sequences.
doi_str_mv	10.1007/11871842_20
format	Book Chapter
fullrecord	<record><control><sourceid>pascalfrancis_sprin</sourceid><recordid>TN_cdi_pascalfrancis_primary_19910363</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>19910363</sourcerecordid><originalsourceid>FETCH-LOGICAL-p284t-17ba6ef131e741241fc9a3f5a67f5822aa57ab80772308ff7d15b58e86a405363</originalsourceid><addsrcrecordid>eNpNkDtPwzAUhc1Loiqd-ANZGBgC9_odNhS1gFQJqZTZchK7MqRxicvAv8eoSHCXM3yfjq4OIZcINwigbhG1Qs2poXBEZpXSTHDgEoQUx2SCErFkjFcnf0wwJfQpmQADWlaKs3MyS-kN8rFcxumEVOvQd65eLe6KOg5d2Ic42L5Y2aGL22IRXN-lwsexWMZNaDN5cR-fbmhduiBn3vbJzX5zSl4X83X9WC6fH57q-2W5o5rvS1SNlc4jQ6c4Uo6-rSzzwkrlhabUWqFso0EpykB7rzoUjdBOS8tBMMmm5OrQu7MpP-BHO7Qhmd0Ytnb8MlhVCFnL3vXBSxkNGzeaJsb3ZBDMz3zm33zsG3K3Wdc</addsrcrecordid><sourcetype>Index Database</sourcetype><iscdi>true</iscdi><recordtype>book_chapter</recordtype></control><display><type>book_chapter</type><title>TildeCRF: Conditional Random Fields for Logical Sequences</title><source>Springer Books</source><creator>Gutmann, Bernd ; Kersting, Kristian</creator><contributor>Fürnkranz, Johannes ; Spiliopoulou, Myra ; Scheffer, Tobias</contributor><creatorcontrib>Gutmann, Bernd ; Kersting, Kristian ; Fürnkranz, Johannes ; Spiliopoulou, Myra ; Scheffer, Tobias</creatorcontrib><description>Conditional Random Fields (CRFs) provide a powerful instrument for labeling sequences. So far, however, CRFs have only been considered for labeling sequences over flat alphabets. In this paper, we describe TildeCRF, the first method for training CRFs on logical sequences, i.e., sequences over an alphabet of logical atoms. TildeCRF’s key idea is to use relational regression trees in Dietterich et al.’s gradient tree boosting approach. Thus, the CRF potential functions are represented as weighted sums of relational regression trees. Experiments show a significant improvement over established results achieved with hidden Markov models and Fisher kernels for logical sequences.</description><identifier>ISSN: 0302-9743</identifier><identifier>ISBN: 9783540453758</identifier><identifier>ISBN: 354045375X</identifier><identifier>EISSN: 1611-3349</identifier><identifier>EISBN: 9783540460565</identifier><identifier>EISBN: 354046056X</identifier><identifier>DOI: 10.1007/11871842_20</identifier><language>eng</language><publisher>Berlin, Heidelberg: Springer Berlin Heidelberg</publisher><subject>Applied sciences ; Artificial intelligence ; Computer science; control theory; systems ; Conditional Random Field ; Exact sciences and technology ; Ground Atom ; Hide Markov Model ; Logical Sequence ; Regression Tree</subject><ispartof>Lecture notes in computer science, 2006, p.174-185</ispartof><rights>Springer-Verlag Berlin Heidelberg 2006</rights><rights>2008 INIST-CNRS</rights><lds50>peer_reviewed</lds50><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed><relation>Lecture Notes in Computer Science</relation></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://link.springer.com/content/pdf/10.1007/11871842_20$$EPDF$$P50$$Gspringer$$H</linktopdf><linktohtml>$$Uhttps://link.springer.com/10.1007/11871842_20$$EHTML$$P50$$Gspringer$$H</linktohtml><link.rule.ids>310,311,780,781,785,790,791,794,27930,38260,41447,42516</link.rule.ids><backlink>$$Uhttp://pascal-francis.inist.fr/vibad/index.php?action=getRecordDetail&idt=19910363$$DView record in Pascal Francis$$Hfree_for_read</backlink></links><search><contributor>Fürnkranz, Johannes</contributor><contributor>Spiliopoulou, Myra</contributor><contributor>Scheffer, Tobias</contributor><creatorcontrib>Gutmann, Bernd</creatorcontrib><creatorcontrib>Kersting, Kristian</creatorcontrib><title>TildeCRF: Conditional Random Fields for Logical Sequences</title><title>Lecture notes in computer science</title><description>Conditional Random Fields (CRFs) provide a powerful instrument for labeling sequences. So far, however, CRFs have only been considered for labeling sequences over flat alphabets. In this paper, we describe TildeCRF, the first method for training CRFs on logical sequences, i.e., sequences over an alphabet of logical atoms. TildeCRF’s key idea is to use relational regression trees in Dietterich et al.’s gradient tree boosting approach. Thus, the CRF potential functions are represented as weighted sums of relational regression trees. Experiments show a significant improvement over established results achieved with hidden Markov models and Fisher kernels for logical sequences.</description><subject>Applied sciences</subject><subject>Artificial intelligence</subject><subject>Computer science; control theory; systems</subject><subject>Conditional Random Field</subject><subject>Exact sciences and technology</subject><subject>Ground Atom</subject><subject>Hide Markov Model</subject><subject>Logical Sequence</subject><subject>Regression Tree</subject><issn>0302-9743</issn><issn>1611-3349</issn><isbn>9783540453758</isbn><isbn>354045375X</isbn><isbn>9783540460565</isbn><isbn>354046056X</isbn><fulltext>true</fulltext><rsrctype>book_chapter</rsrctype><creationdate>2006</creationdate><recordtype>book_chapter</recordtype><recordid>eNpNkDtPwzAUhc1Loiqd-ANZGBgC9_odNhS1gFQJqZTZchK7MqRxicvAv8eoSHCXM3yfjq4OIZcINwigbhG1Qs2poXBEZpXSTHDgEoQUx2SCErFkjFcnf0wwJfQpmQADWlaKs3MyS-kN8rFcxumEVOvQd65eLe6KOg5d2Ic42L5Y2aGL22IRXN-lwsexWMZNaDN5cR-fbmhduiBn3vbJzX5zSl4X83X9WC6fH57q-2W5o5rvS1SNlc4jQ6c4Uo6-rSzzwkrlhabUWqFso0EpykB7rzoUjdBOS8tBMMmm5OrQu7MpP-BHO7Qhmd0Ytnb8MlhVCFnL3vXBSxkNGzeaJsb3ZBDMz3zm33zsG3K3Wdc</recordid><startdate>20060101</startdate><enddate>20060101</enddate><creator>Gutmann, Bernd</creator><creator>Kersting, Kristian</creator><general>Springer Berlin Heidelberg</general><general>Springer</general><scope>IQODW</scope></search><sort><creationdate>20060101</creationdate><title>TildeCRF: Conditional Random Fields for Logical Sequences</title><author>Gutmann, Bernd ; Kersting, Kristian</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-p284t-17ba6ef131e741241fc9a3f5a67f5822aa57ab80772308ff7d15b58e86a405363</frbrgroupid><rsrctype>book_chapters</rsrctype><prefilter>book_chapters</prefilter><language>eng</language><creationdate>2006</creationdate><topic>Applied sciences</topic><topic>Artificial intelligence</topic><topic>Computer science; control theory; systems</topic><topic>Conditional Random Field</topic><topic>Exact sciences and technology</topic><topic>Ground Atom</topic><topic>Hide Markov Model</topic><topic>Logical Sequence</topic><topic>Regression Tree</topic><toplevel>peer_reviewed</toplevel><toplevel>online_resources</toplevel><creatorcontrib>Gutmann, Bernd</creatorcontrib><creatorcontrib>Kersting, Kristian</creatorcontrib><collection>Pascal-Francis</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Gutmann, Bernd</au><au>Kersting, Kristian</au><au>Fürnkranz, Johannes</au><au>Spiliopoulou, Myra</au><au>Scheffer, Tobias</au><format>book</format><genre>bookitem</genre><ristype>CHAP</ristype><atitle>TildeCRF: Conditional Random Fields for Logical Sequences</atitle><btitle>Lecture notes in computer science</btitle><seriestitle>Lecture Notes in Computer Science</seriestitle><date>2006-01-01</date><risdate>2006</risdate><spage>174</spage><epage>185</epage><pages>174-185</pages><issn>0302-9743</issn><eissn>1611-3349</eissn><isbn>9783540453758</isbn><isbn>354045375X</isbn><eisbn>9783540460565</eisbn><eisbn>354046056X</eisbn><abstract>Conditional Random Fields (CRFs) provide a powerful instrument for labeling sequences. So far, however, CRFs have only been considered for labeling sequences over flat alphabets. In this paper, we describe TildeCRF, the first method for training CRFs on logical sequences, i.e., sequences over an alphabet of logical atoms. TildeCRF’s key idea is to use relational regression trees in Dietterich et al.’s gradient tree boosting approach. Thus, the CRF potential functions are represented as weighted sums of relational regression trees. Experiments show a significant improvement over established results achieved with hidden Markov models and Fisher kernels for logical sequences.</abstract><cop>Berlin, Heidelberg</cop><pub>Springer Berlin Heidelberg</pub><doi>10.1007/11871842_20</doi><tpages>12</tpages><oa>free_for_read</oa></addata></record>
fulltext	fulltext
identifier	ISSN: 0302-9743
ispartof	Lecture notes in computer science, 2006, p.174-185
issn	0302-9743 1611-3349
language	eng
recordid	cdi_pascalfrancis_primary_19910363
source	Springer Books
subjects	Applied sciences Artificial intelligence Computer science control theory systems Conditional Random Field Exact sciences and technology Ground Atom Hide Markov Model Logical Sequence Regression Tree
title	TildeCRF: Conditional Random Fields for Logical Sequences
url	https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-12T18%3A38%3A09IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-pascalfrancis_sprin&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=bookitem&rft.atitle=TildeCRF:%20Conditional%20Random%20Fields%20for%20Logical%20Sequences&rft.btitle=Lecture%20notes%20in%20computer%20science&rft.au=Gutmann,%20Bernd&rft.date=2006-01-01&rft.spage=174&rft.epage=185&rft.pages=174-185&rft.issn=0302-9743&rft.eissn=1611-3349&rft.isbn=9783540453758&rft.isbn_list=354045375X&rft_id=info:doi/10.1007/11871842_20&rft_dat=%3Cpascalfrancis_sprin%3E19910363%3C/pascalfrancis_sprin%3E%3Curl%3E%3C/url%3E&rft.eisbn=9783540460565&rft.eisbn_list=354046056X&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true