Ordinal Regression via Binary Preference vs Simple Regression: Statistical and Experimental Perspectives

Ordinal regression with anchored reference samples (ORARS) has been proposed for predicting the subjective Mean Opinion Score (MOS) of input stimuli automatically. The ORARS addresses the MOS prediction problem by pairing a test sample with each of the pre-scored anchored reference samples. A traine...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Su, Bin, Mao, Shaoguang, Soong, Frank, Wu, Zhiyong
Format: Artikel
Sprache:eng
Schlagworte:
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 Su, Bin
Mao, Shaoguang
Soong, Frank
Wu, Zhiyong
description Ordinal regression with anchored reference samples (ORARS) has been proposed for predicting the subjective Mean Opinion Score (MOS) of input stimuli automatically. The ORARS addresses the MOS prediction problem by pairing a test sample with each of the pre-scored anchored reference samples. A trained binary classifier is then used to predict which sample, test or anchor, is better statistically. Posteriors of the binary preference decision are then used to predict the MOS of the test sample. In this paper, rigorous framework, analysis, and experiments to demonstrate that ORARS are advantageous over simple regressions are presented. The contributions of this work are: 1) Show that traditional regression can be reformulated into multiple preference tests to yield a better performance, which is confirmed with simulations experimentally; 2) Generalize ORARS to other regression problems and verify its effectiveness; 3) Provide some prerequisite conditions which can insure proper application of ORARS.
doi_str_mv 10.48550/arxiv.2207.02454
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2207_02454</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2207_02454</sourcerecordid><originalsourceid>FETCH-LOGICAL-a674-fc4a6ca7e526cb7e096839742329621dcb3b16cd3ee26c33e57c948c3b0816403</originalsourceid><addsrcrecordid>eNpNj8tOwzAURL1hgQofwAr_QILjZ8IOqvKQKrWi3UfOzU1rKTWRbUXl7zGFBauRRqOjOYTcVayUtVLswYazm0vOmSkZl0pek-Mm9M7bkX7gIWCM7tPT2Vn6nMvwRbcBBwzoAekc6c6dphH_TR_pLtnkYnKQEdb3dHWeMLgT-pSLLYY4ISQ3Y7whV4MdI97-5YLsX1b75Vux3ry-L5_WhdVGFgNIq8EaVFxDZ5A1uhaNkVzwRvOqh050lYZeIOaBEKgMNLIG0bG60pKJBbn_xV5U2yl_yR7tj3J7URbfSHBSxw</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Ordinal Regression via Binary Preference vs Simple Regression: Statistical and Experimental Perspectives</title><source>arXiv.org</source><creator>Su, Bin ; Mao, Shaoguang ; Soong, Frank ; Wu, Zhiyong</creator><creatorcontrib>Su, Bin ; Mao, Shaoguang ; Soong, Frank ; Wu, Zhiyong</creatorcontrib><description>Ordinal regression with anchored reference samples (ORARS) has been proposed for predicting the subjective Mean Opinion Score (MOS) of input stimuli automatically. The ORARS addresses the MOS prediction problem by pairing a test sample with each of the pre-scored anchored reference samples. A trained binary classifier is then used to predict which sample, test or anchor, is better statistically. Posteriors of the binary preference decision are then used to predict the MOS of the test sample. In this paper, rigorous framework, analysis, and experiments to demonstrate that ORARS are advantageous over simple regressions are presented. The contributions of this work are: 1) Show that traditional regression can be reformulated into multiple preference tests to yield a better performance, which is confirmed with simulations experimentally; 2) Generalize ORARS to other regression problems and verify its effectiveness; 3) Provide some prerequisite conditions which can insure proper application of ORARS.</description><identifier>DOI: 10.48550/arxiv.2207.02454</identifier><language>eng</language><subject>Computer Science - Learning</subject><creationdate>2022-07</creationdate><rights>http://creativecommons.org/licenses/by/4.0</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/2207.02454$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2207.02454$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Su, Bin</creatorcontrib><creatorcontrib>Mao, Shaoguang</creatorcontrib><creatorcontrib>Soong, Frank</creatorcontrib><creatorcontrib>Wu, Zhiyong</creatorcontrib><title>Ordinal Regression via Binary Preference vs Simple Regression: Statistical and Experimental Perspectives</title><description>Ordinal regression with anchored reference samples (ORARS) has been proposed for predicting the subjective Mean Opinion Score (MOS) of input stimuli automatically. The ORARS addresses the MOS prediction problem by pairing a test sample with each of the pre-scored anchored reference samples. A trained binary classifier is then used to predict which sample, test or anchor, is better statistically. Posteriors of the binary preference decision are then used to predict the MOS of the test sample. In this paper, rigorous framework, analysis, and experiments to demonstrate that ORARS are advantageous over simple regressions are presented. The contributions of this work are: 1) Show that traditional regression can be reformulated into multiple preference tests to yield a better performance, which is confirmed with simulations experimentally; 2) Generalize ORARS to other regression problems and verify its effectiveness; 3) Provide some prerequisite conditions which can insure proper application of ORARS.</description><subject>Computer Science - Learning</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNpNj8tOwzAURL1hgQofwAr_QILjZ8IOqvKQKrWi3UfOzU1rKTWRbUXl7zGFBauRRqOjOYTcVayUtVLswYazm0vOmSkZl0pek-Mm9M7bkX7gIWCM7tPT2Vn6nMvwRbcBBwzoAekc6c6dphH_TR_pLtnkYnKQEdb3dHWeMLgT-pSLLYY4ISQ3Y7whV4MdI97-5YLsX1b75Vux3ry-L5_WhdVGFgNIq8EaVFxDZ5A1uhaNkVzwRvOqh050lYZeIOaBEKgMNLIG0bG60pKJBbn_xV5U2yl_yR7tj3J7URbfSHBSxw</recordid><startdate>20220706</startdate><enddate>20220706</enddate><creator>Su, Bin</creator><creator>Mao, Shaoguang</creator><creator>Soong, Frank</creator><creator>Wu, Zhiyong</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20220706</creationdate><title>Ordinal Regression via Binary Preference vs Simple Regression: Statistical and Experimental Perspectives</title><author>Su, Bin ; Mao, Shaoguang ; Soong, Frank ; Wu, Zhiyong</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a674-fc4a6ca7e526cb7e096839742329621dcb3b16cd3ee26c33e57c948c3b0816403</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Computer Science - Learning</topic><toplevel>online_resources</toplevel><creatorcontrib>Su, Bin</creatorcontrib><creatorcontrib>Mao, Shaoguang</creatorcontrib><creatorcontrib>Soong, Frank</creatorcontrib><creatorcontrib>Wu, Zhiyong</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Su, Bin</au><au>Mao, Shaoguang</au><au>Soong, Frank</au><au>Wu, Zhiyong</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Ordinal Regression via Binary Preference vs Simple Regression: Statistical and Experimental Perspectives</atitle><date>2022-07-06</date><risdate>2022</risdate><abstract>Ordinal regression with anchored reference samples (ORARS) has been proposed for predicting the subjective Mean Opinion Score (MOS) of input stimuli automatically. The ORARS addresses the MOS prediction problem by pairing a test sample with each of the pre-scored anchored reference samples. A trained binary classifier is then used to predict which sample, test or anchor, is better statistically. Posteriors of the binary preference decision are then used to predict the MOS of the test sample. In this paper, rigorous framework, analysis, and experiments to demonstrate that ORARS are advantageous over simple regressions are presented. The contributions of this work are: 1) Show that traditional regression can be reformulated into multiple preference tests to yield a better performance, which is confirmed with simulations experimentally; 2) Generalize ORARS to other regression problems and verify its effectiveness; 3) Provide some prerequisite conditions which can insure proper application of ORARS.</abstract><doi>10.48550/arxiv.2207.02454</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2207.02454
ispartof
issn
language eng
recordid cdi_arxiv_primary_2207_02454
source arXiv.org
subjects Computer Science - Learning
title Ordinal Regression via Binary Preference vs Simple Regression: Statistical and Experimental Perspectives
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-02-02T13%3A25%3A36IST&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=Ordinal%20Regression%20via%20Binary%20Preference%20vs%20Simple%20Regression:%20Statistical%20and%20Experimental%20Perspectives&rft.au=Su,%20Bin&rft.date=2022-07-06&rft_id=info:doi/10.48550/arxiv.2207.02454&rft_dat=%3Carxiv_GOX%3E2207_02454%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