Marginal Nodes Matter: Towards Structure Fairness in Graphs
In social network, a person located at the periphery region (marginal node) is likely to be treated unfairly when compared with the persons at the center. While existing fairness works on graphs mainly focus on protecting sensitive attributes (e.g., age and gender), the fairness incurred by the grap...
Gespeichert in:
Veröffentlicht in: | SIGKDD explorations 2024-03, Vol.25 (2), p.4-13 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
container_end_page | 13 |
---|---|
container_issue | 2 |
container_start_page | 4 |
container_title | SIGKDD explorations |
container_volume | 25 |
creator | Han, Xiaotian Zhou, Kaixiong Wang, Ting-Hsiang Li, Jundong Wang, Fei Zou, Na |
description | In social network, a person located at the periphery region (marginal node) is likely to be treated unfairly when compared with the persons at the center. While existing fairness works on graphs mainly focus on protecting sensitive attributes (e.g., age and gender), the fairness incurred by the graph structure should also be given attention. On the other hand, the information aggregation mechanism of graph neural networks amplifies such structure unfairness, as marginal nodes are often far away from other nodes. In this paper, we focus on novel fairness incurred by the graph structure on graph neural networks, named structure fairness. Specifically, we first analyzed multiple graphs and observed that marginal nodes in graphs have a worse performance of downstream tasks than others in graph neural networks. Motivated by the observation, we propose Structural Fair Graph Neural Network (SFairGNN), which combines neighborhood expansion based structure debiasing with hop-aware attentive information aggregation to achieve structure fairness. Our experiments show SFairGNN can significantly improve structure fairness while maintaining overall performance in the downstream tasks. |
doi_str_mv | 10.1145/3655103.3655105 |
format | Article |
fullrecord | <record><control><sourceid>acm_cross</sourceid><recordid>TN_cdi_crossref_primary_10_1145_3655103_3655105</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>3655105</sourcerecordid><originalsourceid>FETCH-LOGICAL-a1005-c7d9fdce0a7103660b44b4c626e3d905e1c5bb5e6b4f9ba0bf15530d643416e23</originalsourceid><addsrcrecordid>eNo9j09Lw0AQxRdRsFbPgqf9AmlnsjtboycptgqtHqznMPsnGmnTspsifnsriT39Hsx7j3lCXCOMEDWNlSFCUKOOdCIGWCjMAEmdHrWmc3GR0hdAfotGDcT9kuNH3fBavmx9SHLJbRvinVxtvzn6JN_auHftPgY54zo2ISVZN3IeefeZLsVZxesUrnoOxfvscTV9yhav8-fpwyJjBKDMTXxReReAJ4f_jAGrtdXO5CYoXwAFdGQtBWN1VVgGWyGRAm-00mhCroZi3PW6uE0phqrcxXrD8adEKP-2l_32nnRI3HQJdpuj-f_4C-qDUzA</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Marginal Nodes Matter: Towards Structure Fairness in Graphs</title><source>Access via ACM Digital Library</source><source>EZB-FREE-00999 freely available EZB journals</source><creator>Han, Xiaotian ; Zhou, Kaixiong ; Wang, Ting-Hsiang ; Li, Jundong ; Wang, Fei ; Zou, Na</creator><creatorcontrib>Han, Xiaotian ; Zhou, Kaixiong ; Wang, Ting-Hsiang ; Li, Jundong ; Wang, Fei ; Zou, Na</creatorcontrib><description>In social network, a person located at the periphery region (marginal node) is likely to be treated unfairly when compared with the persons at the center. While existing fairness works on graphs mainly focus on protecting sensitive attributes (e.g., age and gender), the fairness incurred by the graph structure should also be given attention. On the other hand, the information aggregation mechanism of graph neural networks amplifies such structure unfairness, as marginal nodes are often far away from other nodes. In this paper, we focus on novel fairness incurred by the graph structure on graph neural networks, named structure fairness. Specifically, we first analyzed multiple graphs and observed that marginal nodes in graphs have a worse performance of downstream tasks than others in graph neural networks. Motivated by the observation, we propose Structural Fair Graph Neural Network (SFairGNN), which combines neighborhood expansion based structure debiasing with hop-aware attentive information aggregation to achieve structure fairness. Our experiments show SFairGNN can significantly improve structure fairness while maintaining overall performance in the downstream tasks.</description><identifier>ISSN: 1931-0145</identifier><identifier>EISSN: 1931-0153</identifier><identifier>DOI: 10.1145/3655103.3655105</identifier><language>eng</language><publisher>New York, NY, USA: ACM</publisher><subject>Computing methodologies ; Computing methodologies / Machine learning ; Computing methodologies / Machine learning / Machine learning approaches ; Computing methodologies / Machine learning / Machine learning approaches / Neural networks ; Information systems ; Information systems / Information systems applications</subject><ispartof>SIGKDD explorations, 2024-03, Vol.25 (2), p.4-13</ispartof><rights>Copyright is held by the owner/author(s)</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><linktopdf>$$Uhttps://dl.acm.org/doi/pdf/10.1145/3655103.3655105$$EPDF$$P50$$Gacm$$H</linktopdf><link.rule.ids>314,780,784,2282,27924,27925,40196,76228</link.rule.ids></links><search><creatorcontrib>Han, Xiaotian</creatorcontrib><creatorcontrib>Zhou, Kaixiong</creatorcontrib><creatorcontrib>Wang, Ting-Hsiang</creatorcontrib><creatorcontrib>Li, Jundong</creatorcontrib><creatorcontrib>Wang, Fei</creatorcontrib><creatorcontrib>Zou, Na</creatorcontrib><title>Marginal Nodes Matter: Towards Structure Fairness in Graphs</title><title>SIGKDD explorations</title><addtitle>ACM SIGKDD</addtitle><description>In social network, a person located at the periphery region (marginal node) is likely to be treated unfairly when compared with the persons at the center. While existing fairness works on graphs mainly focus on protecting sensitive attributes (e.g., age and gender), the fairness incurred by the graph structure should also be given attention. On the other hand, the information aggregation mechanism of graph neural networks amplifies such structure unfairness, as marginal nodes are often far away from other nodes. In this paper, we focus on novel fairness incurred by the graph structure on graph neural networks, named structure fairness. Specifically, we first analyzed multiple graphs and observed that marginal nodes in graphs have a worse performance of downstream tasks than others in graph neural networks. Motivated by the observation, we propose Structural Fair Graph Neural Network (SFairGNN), which combines neighborhood expansion based structure debiasing with hop-aware attentive information aggregation to achieve structure fairness. Our experiments show SFairGNN can significantly improve structure fairness while maintaining overall performance in the downstream tasks.</description><subject>Computing methodologies</subject><subject>Computing methodologies / Machine learning</subject><subject>Computing methodologies / Machine learning / Machine learning approaches</subject><subject>Computing methodologies / Machine learning / Machine learning approaches / Neural networks</subject><subject>Information systems</subject><subject>Information systems / Information systems applications</subject><issn>1931-0145</issn><issn>1931-0153</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2024</creationdate><recordtype>article</recordtype><recordid>eNo9j09Lw0AQxRdRsFbPgqf9AmlnsjtboycptgqtHqznMPsnGmnTspsifnsriT39Hsx7j3lCXCOMEDWNlSFCUKOOdCIGWCjMAEmdHrWmc3GR0hdAfotGDcT9kuNH3fBavmx9SHLJbRvinVxtvzn6JN_auHftPgY54zo2ISVZN3IeefeZLsVZxesUrnoOxfvscTV9yhav8-fpwyJjBKDMTXxReReAJ4f_jAGrtdXO5CYoXwAFdGQtBWN1VVgGWyGRAm-00mhCroZi3PW6uE0phqrcxXrD8adEKP-2l_32nnRI3HQJdpuj-f_4C-qDUzA</recordid><startdate>20240328</startdate><enddate>20240328</enddate><creator>Han, Xiaotian</creator><creator>Zhou, Kaixiong</creator><creator>Wang, Ting-Hsiang</creator><creator>Li, Jundong</creator><creator>Wang, Fei</creator><creator>Zou, Na</creator><general>ACM</general><scope>AAYXX</scope><scope>CITATION</scope></search><sort><creationdate>20240328</creationdate><title>Marginal Nodes Matter: Towards Structure Fairness in Graphs</title><author>Han, Xiaotian ; Zhou, Kaixiong ; Wang, Ting-Hsiang ; Li, Jundong ; Wang, Fei ; Zou, Na</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a1005-c7d9fdce0a7103660b44b4c626e3d905e1c5bb5e6b4f9ba0bf15530d643416e23</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2024</creationdate><topic>Computing methodologies</topic><topic>Computing methodologies / Machine learning</topic><topic>Computing methodologies / Machine learning / Machine learning approaches</topic><topic>Computing methodologies / Machine learning / Machine learning approaches / Neural networks</topic><topic>Information systems</topic><topic>Information systems / Information systems applications</topic><toplevel>online_resources</toplevel><creatorcontrib>Han, Xiaotian</creatorcontrib><creatorcontrib>Zhou, Kaixiong</creatorcontrib><creatorcontrib>Wang, Ting-Hsiang</creatorcontrib><creatorcontrib>Li, Jundong</creatorcontrib><creatorcontrib>Wang, Fei</creatorcontrib><creatorcontrib>Zou, Na</creatorcontrib><collection>CrossRef</collection><jtitle>SIGKDD explorations</jtitle></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Han, Xiaotian</au><au>Zhou, Kaixiong</au><au>Wang, Ting-Hsiang</au><au>Li, Jundong</au><au>Wang, Fei</au><au>Zou, Na</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Marginal Nodes Matter: Towards Structure Fairness in Graphs</atitle><jtitle>SIGKDD explorations</jtitle><stitle>ACM SIGKDD</stitle><date>2024-03-28</date><risdate>2024</risdate><volume>25</volume><issue>2</issue><spage>4</spage><epage>13</epage><pages>4-13</pages><issn>1931-0145</issn><eissn>1931-0153</eissn><abstract>In social network, a person located at the periphery region (marginal node) is likely to be treated unfairly when compared with the persons at the center. While existing fairness works on graphs mainly focus on protecting sensitive attributes (e.g., age and gender), the fairness incurred by the graph structure should also be given attention. On the other hand, the information aggregation mechanism of graph neural networks amplifies such structure unfairness, as marginal nodes are often far away from other nodes. In this paper, we focus on novel fairness incurred by the graph structure on graph neural networks, named structure fairness. Specifically, we first analyzed multiple graphs and observed that marginal nodes in graphs have a worse performance of downstream tasks than others in graph neural networks. Motivated by the observation, we propose Structural Fair Graph Neural Network (SFairGNN), which combines neighborhood expansion based structure debiasing with hop-aware attentive information aggregation to achieve structure fairness. Our experiments show SFairGNN can significantly improve structure fairness while maintaining overall performance in the downstream tasks.</abstract><cop>New York, NY, USA</cop><pub>ACM</pub><doi>10.1145/3655103.3655105</doi><tpages>10</tpages><oa>free_for_read</oa></addata></record> |
fulltext | fulltext |
identifier | ISSN: 1931-0145 |
ispartof | SIGKDD explorations, 2024-03, Vol.25 (2), p.4-13 |
issn | 1931-0145 1931-0153 |
language | eng |
recordid | cdi_crossref_primary_10_1145_3655103_3655105 |
source | Access via ACM Digital Library; EZB-FREE-00999 freely available EZB journals |
subjects | Computing methodologies Computing methodologies / Machine learning Computing methodologies / Machine learning / Machine learning approaches Computing methodologies / Machine learning / Machine learning approaches / Neural networks Information systems Information systems / Information systems applications |
title | Marginal Nodes Matter: Towards Structure Fairness in Graphs |
url | https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2024-12-25T01%3A34%3A10IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-acm_cross&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Marginal%20Nodes%20Matter:%20Towards%20Structure%20Fairness%20in%20Graphs&rft.jtitle=SIGKDD%20explorations&rft.au=Han,%20Xiaotian&rft.date=2024-03-28&rft.volume=25&rft.issue=2&rft.spage=4&rft.epage=13&rft.pages=4-13&rft.issn=1931-0145&rft.eissn=1931-0153&rft_id=info:doi/10.1145/3655103.3655105&rft_dat=%3Cacm_cross%3E3655105%3C/acm_cross%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 |