On Estimation and Selection for Topic Models

On Estimation and Selection for Topic Models

This article describes posterior maximization for topic models, identifying computational and conceptual gains from inference under a non-standard parametrization. We then show that fitted parameters can be used as the basis for a novel approach to marginal likelihood estimation, via block-diagonal...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Taddy, Matthew A
Format: Artikel
Sprache:eng
Schlagworte:
Statistics - Applications
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Taddy, Matthew A
description This article describes posterior maximization for topic models, identifying computational and conceptual gains from inference under a non-standard parametrization. We then show that fitted parameters can be used as the basis for a novel approach to marginal likelihood estimation, via block-diagonal approximation to the information matrix,that facilitates choosing the number of latent topics. This likelihood-based model selection is complemented with a goodness-of-fit analysis built around estimated residual dispersion. Examples are provided to illustrate model selection as well as to compare our estimation against standard alternative techniques.
doi_str_mv 10.48550/arxiv.1109.4518
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_1109_4518</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>1109_4518</sourcerecordid><originalsourceid>FETCH-LOGICAL-a658-162d938204ca03039d0781d741c47d8f25810885a34e0661b3e9443c8ccbcb5a3</originalsourceid><addsrcrecordid>eNotzr0OgjAUBeAuDkbdnUwfQPBe2sJlNAZ_Eo2D7KS0JSFBIECMvr2iTifnDCcfY0sEX5JSsNHds3z4iBD7UiFN2fpa86Qfyrseyqbmurb85ipnvq1oOp42bWn4pbGu6udsUuiqd4t_zli6T9Ld0TtfD6fd9uzpUJGHYWBjQQFIo0GAiC1EhDaSaGRkqQgUIRApLaSDMMRcuFhKYciY3OSfecZWv9uvNmu7j657ZaM6G9XiDey3OnE</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>On Estimation and Selection for Topic Models</title><source>arXiv.org</source><creator>Taddy, Matthew A</creator><creatorcontrib>Taddy, Matthew A</creatorcontrib><description>This article describes posterior maximization for topic models, identifying computational and conceptual gains from inference under a non-standard parametrization. We then show that fitted parameters can be used as the basis for a novel approach to marginal likelihood estimation, via block-diagonal approximation to the information matrix,that facilitates choosing the number of latent topics. This likelihood-based model selection is complemented with a goodness-of-fit analysis built around estimated residual dispersion. Examples are provided to illustrate model selection as well as to compare our estimation against standard alternative techniques.</description><identifier>DOI: 10.48550/arxiv.1109.4518</identifier><language>eng</language><subject>Statistics - Applications</subject><creationdate>2011-09</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/1109.4518$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.1109.4518$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Taddy, Matthew A</creatorcontrib><title>On Estimation and Selection for Topic Models</title><description>This article describes posterior maximization for topic models, identifying computational and conceptual gains from inference under a non-standard parametrization. We then show that fitted parameters can be used as the basis for a novel approach to marginal likelihood estimation, via block-diagonal approximation to the information matrix,that facilitates choosing the number of latent topics. This likelihood-based model selection is complemented with a goodness-of-fit analysis built around estimated residual dispersion. Examples are provided to illustrate model selection as well as to compare our estimation against standard alternative techniques.</description><subject>Statistics - Applications</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2011</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotzr0OgjAUBeAuDkbdnUwfQPBe2sJlNAZ_Eo2D7KS0JSFBIECMvr2iTifnDCcfY0sEX5JSsNHds3z4iBD7UiFN2fpa86Qfyrseyqbmurb85ipnvq1oOp42bWn4pbGu6udsUuiqd4t_zli6T9Ld0TtfD6fd9uzpUJGHYWBjQQFIo0GAiC1EhDaSaGRkqQgUIRApLaSDMMRcuFhKYciY3OSfecZWv9uvNmu7j657ZaM6G9XiDey3OnE</recordid><startdate>20110921</startdate><enddate>20110921</enddate><creator>Taddy, Matthew A</creator><scope>EPD</scope><scope>GOX</scope></search><sort><creationdate>20110921</creationdate><title>On Estimation and Selection for Topic Models</title><author>Taddy, Matthew A</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a658-162d938204ca03039d0781d741c47d8f25810885a34e0661b3e9443c8ccbcb5a3</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2011</creationdate><topic>Statistics - Applications</topic><toplevel>online_resources</toplevel><creatorcontrib>Taddy, Matthew A</creatorcontrib><collection>arXiv Statistics</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Taddy, Matthew A</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>On Estimation and Selection for Topic Models</atitle><date>2011-09-21</date><risdate>2011</risdate><abstract>This article describes posterior maximization for topic models, identifying computational and conceptual gains from inference under a non-standard parametrization. We then show that fitted parameters can be used as the basis for a novel approach to marginal likelihood estimation, via block-diagonal approximation to the information matrix,that facilitates choosing the number of latent topics. This likelihood-based model selection is complemented with a goodness-of-fit analysis built around estimated residual dispersion. Examples are provided to illustrate model selection as well as to compare our estimation against standard alternative techniques.</abstract><doi>10.48550/arxiv.1109.4518</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.1109.4518
ispartof
issn
language eng
recordid cdi_arxiv_primary_1109_4518
source arXiv.org
subjects Statistics - Applications
title On Estimation and Selection for Topic Models
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-25T17%3A09%3A25IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=On%20Estimation%20and%20Selection%20for%20Topic%20Models&rft.au=Taddy,%20Matthew%20A&rft.date=2011-09-21&rft_id=info:doi/10.48550/arxiv.1109.4518&rft_dat=%3Carxiv_GOX%3E1109_4518%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true