Loading…

Discriminatively Enhanced Topic Models

This paper proposes a space-efficient, discriminatively enhanced topic model: a V structured topic model with an embedded log-linear component. The discriminative log-linear component reduces the number of parameters to be learnt while outperforming baseline generative models. At the same time, the...

Full description

Saved in:
Bibliographic Details
Main Authors: Chaturvedi, Snigdha, Daume, Hal, Moon, Taesun
Format: Conference Proceeding
Language:English
Subjects:
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
cited_by
cites
container_end_page 990
container_issue
container_start_page 985
container_title
container_volume
creator Chaturvedi, Snigdha
Daume, Hal
Moon, Taesun
description This paper proposes a space-efficient, discriminatively enhanced topic model: a V structured topic model with an embedded log-linear component. The discriminative log-linear component reduces the number of parameters to be learnt while outperforming baseline generative models. At the same time, the explanatory power of the generative component is not compromised. We establish its superiority over a purely generative model by applying it to two different ranking tasks: (a) In the first task, we look at the problem of proposing alternative citations given textual and bibliographic evidence. We solve it as a ranking problem in itself and as a platform for further qualitative analysis of convergence of scientific phenomenon. (b) In the second task we address the problem of ranking potential email recipients based on email content and sender information.
doi_str_mv 10.1109/ICDM.2013.107
format conference_proceeding
fullrecord <record><control><sourceid>proquest_CHZPO</sourceid><recordid>TN_cdi_ieee_primary_6729586</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><ieee_id>6729586</ieee_id><sourcerecordid>1669889711</sourcerecordid><originalsourceid>FETCH-LOGICAL-i208t-7f8d45da529eb2178c61e7d3c4febd90854b1a57841a178ba06888f0112cf6513</originalsourceid><addsrcrecordid>eNotjz1PAkEURUejiYiUVjZUxmbxvfl8UxpAJYHYYL2Z3X0bxyy7uAMm_HswWN3inpzcK8Q9wgQR_PNiOltNJKCaILgLMfKOwFlvDALhpRhI5XRGmuyVGKAxkGlH9kbcpvQNoKxVMBCPs5jKPm5iG3bxl5vDeN5-hbbkarzutrEcr7qKm3QnruvQJB7951B8vs7X0_ds-fG2mL4ssyiBdpmrqdKmCkZ6LiQ6Ki2yq1Spay4qD2R0gcE40hhObRHAElENiLKsrUE1FE9n77bvfvacdvnmtI-bJrTc7VOO1noi7_APfTijkZnz7elD6A-5ddIbsuoIuTJOTQ</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>conference_proceeding</recordtype><pqid>1669889711</pqid></control><display><type>conference_proceeding</type><title>Discriminatively Enhanced Topic Models</title><source>IEEE Xplore All Conference Series</source><creator>Chaturvedi, Snigdha ; Daume, Hal ; Moon, Taesun</creator><creatorcontrib>Chaturvedi, Snigdha ; Daume, Hal ; Moon, Taesun</creatorcontrib><description>This paper proposes a space-efficient, discriminatively enhanced topic model: a V structured topic model with an embedded log-linear component. The discriminative log-linear component reduces the number of parameters to be learnt while outperforming baseline generative models. At the same time, the explanatory power of the generative component is not compromised. We establish its superiority over a purely generative model by applying it to two different ranking tasks: (a) In the first task, we look at the problem of proposing alternative citations given textual and bibliographic evidence. We solve it as a ranking problem in itself and as a platform for further qualitative analysis of convergence of scientific phenomenon. (b) In the second task we address the problem of ranking potential email recipients based on email content and sender information.</description><identifier>ISSN: 1550-4786</identifier><identifier>EISSN: 2374-8486</identifier><identifier>EISBN: 9780769551081</identifier><identifier>EISBN: 0769551084</identifier><identifier>DOI: 10.1109/ICDM.2013.107</identifier><identifier>CODEN: IEEPAD</identifier><language>eng</language><publisher>IEEE</publisher><subject>Bibliographies ; Conferences ; Context modeling ; Convergence ; Data mining ; Electronic mail ; Email ; Hidden Markov models ; Hybrid power systems ; Log linear models ; Mathematical model ; Platforms ; Predictive models ; Probabilistic ranking ; Ranking ; Tasks ; Text Mining ; Topic Models ; Vectors</subject><ispartof>2013 IEEE 13th International Conference on Data Mining, 2013, p.985-990</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/6729586$$EHTML$$P50$$Gieee$$H</linktohtml><link.rule.ids>309,310,314,780,784,789,790,27923,27924,54554,54931</link.rule.ids><linktorsrc>$$Uhttps://ieeexplore.ieee.org/document/6729586$$EView_record_in_IEEE$$FView_record_in_$$GIEEE</linktorsrc></links><search><creatorcontrib>Chaturvedi, Snigdha</creatorcontrib><creatorcontrib>Daume, Hal</creatorcontrib><creatorcontrib>Moon, Taesun</creatorcontrib><title>Discriminatively Enhanced Topic Models</title><title>2013 IEEE 13th International Conference on Data Mining</title><addtitle>icdm</addtitle><description>This paper proposes a space-efficient, discriminatively enhanced topic model: a V structured topic model with an embedded log-linear component. The discriminative log-linear component reduces the number of parameters to be learnt while outperforming baseline generative models. At the same time, the explanatory power of the generative component is not compromised. We establish its superiority over a purely generative model by applying it to two different ranking tasks: (a) In the first task, we look at the problem of proposing alternative citations given textual and bibliographic evidence. We solve it as a ranking problem in itself and as a platform for further qualitative analysis of convergence of scientific phenomenon. (b) In the second task we address the problem of ranking potential email recipients based on email content and sender information.</description><subject>Bibliographies</subject><subject>Conferences</subject><subject>Context modeling</subject><subject>Convergence</subject><subject>Data mining</subject><subject>Electronic mail</subject><subject>Email</subject><subject>Hidden Markov models</subject><subject>Hybrid power systems</subject><subject>Log linear models</subject><subject>Mathematical model</subject><subject>Platforms</subject><subject>Predictive models</subject><subject>Probabilistic ranking</subject><subject>Ranking</subject><subject>Tasks</subject><subject>Text Mining</subject><subject>Topic Models</subject><subject>Vectors</subject><issn>1550-4786</issn><issn>2374-8486</issn><isbn>9780769551081</isbn><isbn>0769551084</isbn><fulltext>true</fulltext><rsrctype>conference_proceeding</rsrctype><creationdate>2013</creationdate><recordtype>conference_proceeding</recordtype><sourceid>6IE</sourceid><recordid>eNotjz1PAkEURUejiYiUVjZUxmbxvfl8UxpAJYHYYL2Z3X0bxyy7uAMm_HswWN3inpzcK8Q9wgQR_PNiOltNJKCaILgLMfKOwFlvDALhpRhI5XRGmuyVGKAxkGlH9kbcpvQNoKxVMBCPs5jKPm5iG3bxl5vDeN5-hbbkarzutrEcr7qKm3QnruvQJB7951B8vs7X0_ds-fG2mL4ssyiBdpmrqdKmCkZ6LiQ6Ki2yq1Spay4qD2R0gcE40hhObRHAElENiLKsrUE1FE9n77bvfvacdvnmtI-bJrTc7VOO1noi7_APfTijkZnz7elD6A-5ddIbsuoIuTJOTQ</recordid><startdate>20131201</startdate><enddate>20131201</enddate><creator>Chaturvedi, Snigdha</creator><creator>Daume, Hal</creator><creator>Moon, Taesun</creator><general>IEEE</general><scope>6IE</scope><scope>6IL</scope><scope>CBEJK</scope><scope>RIE</scope><scope>RIL</scope><scope>7SC</scope><scope>8FD</scope><scope>JQ2</scope><scope>L7M</scope><scope>L~C</scope><scope>L~D</scope></search><sort><creationdate>20131201</creationdate><title>Discriminatively Enhanced Topic Models</title><author>Chaturvedi, Snigdha ; Daume, Hal ; Moon, Taesun</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-i208t-7f8d45da529eb2178c61e7d3c4febd90854b1a57841a178ba06888f0112cf6513</frbrgroupid><rsrctype>conference_proceedings</rsrctype><prefilter>conference_proceedings</prefilter><language>eng</language><creationdate>2013</creationdate><topic>Bibliographies</topic><topic>Conferences</topic><topic>Context modeling</topic><topic>Convergence</topic><topic>Data mining</topic><topic>Electronic mail</topic><topic>Email</topic><topic>Hidden Markov models</topic><topic>Hybrid power systems</topic><topic>Log linear models</topic><topic>Mathematical model</topic><topic>Platforms</topic><topic>Predictive models</topic><topic>Probabilistic ranking</topic><topic>Ranking</topic><topic>Tasks</topic><topic>Text Mining</topic><topic>Topic Models</topic><topic>Vectors</topic><toplevel>online_resources</toplevel><creatorcontrib>Chaturvedi, Snigdha</creatorcontrib><creatorcontrib>Daume, Hal</creatorcontrib><creatorcontrib>Moon, Taesun</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>IEL</collection><collection>IEEE Proceedings Order Plans (POP All) 1998-Present</collection><collection>Computer and Information Systems Abstracts</collection><collection>Technology Research Database</collection><collection>ProQuest Computer Science Collection</collection><collection>Advanced Technologies Database with Aerospace</collection><collection>Computer and Information Systems Abstracts – Academic</collection><collection>Computer and Information Systems Abstracts Professional</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Chaturvedi, Snigdha</au><au>Daume, Hal</au><au>Moon, Taesun</au><format>book</format><genre>proceeding</genre><ristype>CONF</ristype><atitle>Discriminatively Enhanced Topic Models</atitle><btitle>2013 IEEE 13th International Conference on Data Mining</btitle><stitle>icdm</stitle><date>2013-12-01</date><risdate>2013</risdate><spage>985</spage><epage>990</epage><pages>985-990</pages><issn>1550-4786</issn><eissn>2374-8486</eissn><eisbn>9780769551081</eisbn><eisbn>0769551084</eisbn><coden>IEEPAD</coden><abstract>This paper proposes a space-efficient, discriminatively enhanced topic model: a V structured topic model with an embedded log-linear component. The discriminative log-linear component reduces the number of parameters to be learnt while outperforming baseline generative models. At the same time, the explanatory power of the generative component is not compromised. We establish its superiority over a purely generative model by applying it to two different ranking tasks: (a) In the first task, we look at the problem of proposing alternative citations given textual and bibliographic evidence. We solve it as a ranking problem in itself and as a platform for further qualitative analysis of convergence of scientific phenomenon. (b) In the second task we address the problem of ranking potential email recipients based on email content and sender information.</abstract><pub>IEEE</pub><doi>10.1109/ICDM.2013.107</doi><tpages>6</tpages></addata></record>
fulltext fulltext_linktorsrc
identifier ISSN: 1550-4786
ispartof 2013 IEEE 13th International Conference on Data Mining, 2013, p.985-990
issn 1550-4786
2374-8486
language eng
recordid cdi_ieee_primary_6729586
source IEEE Xplore All Conference Series
subjects Bibliographies
Conferences
Context modeling
Convergence
Data mining
Electronic mail
Email
Hidden Markov models
Hybrid power systems
Log linear models
Mathematical model
Platforms
Predictive models
Probabilistic ranking
Ranking
Tasks
Text Mining
Topic Models
Vectors
title Discriminatively Enhanced Topic Models
url http://sfxeu10.hosted.exlibrisgroup.com/loughborough?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-13T06%3A01%3A58IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest_CHZPO&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=proceeding&rft.atitle=Discriminatively%20Enhanced%20Topic%20Models&rft.btitle=2013%20IEEE%2013th%20International%20Conference%20on%20Data%20Mining&rft.au=Chaturvedi,%20Snigdha&rft.date=2013-12-01&rft.spage=985&rft.epage=990&rft.pages=985-990&rft.issn=1550-4786&rft.eissn=2374-8486&rft.coden=IEEPAD&rft_id=info:doi/10.1109/ICDM.2013.107&rft.eisbn=9780769551081&rft.eisbn_list=0769551084&rft_dat=%3Cproquest_CHZPO%3E1669889711%3C/proquest_CHZPO%3E%3Cgrp_id%3Ecdi_FETCH-LOGICAL-i208t-7f8d45da529eb2178c61e7d3c4febd90854b1a57841a178ba06888f0112cf6513%3C/grp_id%3E%3Coa%3E%3C/oa%3E%3Curl%3E%3C/url%3E&rft_id=info:oai/&rft_pqid=1669889711&rft_id=info:pmid/&rft_ieee_id=6729586&rfr_iscdi=true