{"id":"https://openalex.org/W4388093300","doi":"https://doi.org/10.48550/arxiv.2310.19263","title":"A Metadata-Driven Approach to Understand Graph Neural Networks","display_name":"A Metadata-Driven Approach to Understand Graph Neural Networks","publication_year":2023,"publication_date":"2023-10-30","ids":{"openalex":"https://openalex.org/W4388093300","doi":"https://doi.org/10.48550/arxiv.2310.19263"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2310.19263","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.19263","pdf_url":"https://arxiv.org/pdf/2310.19263","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2310.19263","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5111112819","display_name":"Ting Wei Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Li, Ting Wei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048955398","display_name":"Qiaozhu Mei","orcid":"https://orcid.org/0000-0002-8640-1942"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mei, Qiaozhu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5089063294","display_name":"Jiaqi Ma","orcid":"https://orcid.org/0000-0001-8292-5901"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ma, Jiaqi","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5111112819"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9983999729156494,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9983999729156494,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7850885391235352},{"id":"https://openalex.org/keywords/metadata","display_name":"Metadata","score":0.6663538813591003},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.557440459728241},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5361499786376953},{"id":"https://openalex.org/keywords/data-mining","display_name":"Data mining","score":0.5328220725059509},{"id":"https://openalex.org/keywords/heuristics","display_name":"Heuristics","score":0.5108006000518799},{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.4959293305873871},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.3570614457130432},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3552076816558838}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7850885391235352},{"id":"https://openalex.org/C93518851","wikidata":"https://www.wikidata.org/wiki/Q180160","display_name":"Metadata","level":2,"score":0.6663538813591003},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.557440459728241},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5361499786376953},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.5328220725059509},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.5108006000518799},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.4959293305873871},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3570614457130432},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3552076816558838},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.0},{"id":"https://openalex.org/C162853370","wikidata":"https://www.wikidata.org/wiki/Q39809","display_name":"Marketing","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2310.19263","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.19263","pdf_url":"https://arxiv.org/pdf/2310.19263","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"doi:10.48550/arxiv.2310.19263","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2310.19263","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2310.19263","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2310.19263","pdf_url":"https://arxiv.org/pdf/2310.19263","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4388093300.pdf","grobid_xml":"https://content.openalex.org/works/W4388093300.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4238897586","https://openalex.org/W435179959","https://openalex.org/W2619091065","https://openalex.org/W2059640416","https://openalex.org/W1490753184","https://openalex.org/W2284465472","https://openalex.org/W2291782699","https://openalex.org/W1993948687","https://openalex.org/W2000169967","https://openalex.org/W2112883198"],"abstract_inverted_index":{"Graph":[0],"Neural":[1],"Networks":[2],"(GNNs)":[3],"have":[4],"achieved":[5],"remarkable":[6],"success":[7],"in":[8,210],"various":[9],"applications,":[10],"but":[11],"their":[12],"performance":[13,106,142],"can":[14],"be":[15],"sensitive":[16],"to":[17,54,73,79,169],"specific":[18],"data":[19,81,115,130,213],"properties":[20,214],"of":[21,34,77,88,113,121,164],"the":[22,32,56,75,85,100,119,132,186,196,204],"graph":[23,52,80,89],"datasets":[24,154,180],"they":[25],"operate":[26],"on.":[27],"Current":[28],"literature":[29],"on":[30,99,127],"understanding":[31],"limitations":[33],"GNNs":[35,78],"has":[36],"primarily":[37],"employed":[38],"a":[39,70,94,111],"$\\textit{model-driven}$":[40],"approach":[41,72,207],"that":[42,153,203],"leverage":[43],"heuristics":[44],"and":[45,62,135,146,185,199],"domain":[46],"knowledge":[47],"from":[48,103],"network":[49],"science":[50],"or":[51],"theory":[53],"model":[55],"GNN":[57,105,141,171],"behaviors,":[58],"which":[59],"is":[60,208],"time-consuming":[61],"highly":[63],"subjective.":[64],"In":[65],"this":[66,138],"work,":[67],"we":[68,125],"propose":[69],"$\\textit{metadata-driven}$":[71],"analyze":[74],"sensitivity":[76],"properties,":[82],"motivated":[83],"by":[84],"increasing":[86],"availability":[87],"learning":[90],"benchmarks.":[91],"We":[92,173],"perform":[93],"multivariate":[95],"sparse":[96],"regression":[97],"analysis":[98,145,198],"metadata":[101],"derived":[102],"benchmarking":[104],"across":[107],"diverse":[108],"datasets,":[109],"yielding":[110],"set":[112],"salient":[114],"properties.":[116],"To":[117],"validate":[118],"effectiveness":[120],"our":[122,191],"data-driven":[123],"approach,":[124],"focus":[126],"one":[128],"identified":[129],"property,":[131],"degree":[133,158,183],"distribution,":[134],"investigate":[136],"how":[137],"property":[139],"influences":[140],"through":[143],"theoretical":[144,150,192,197],"controlled":[147,176,200],"experiments.":[148],"Our":[149],"findings":[151],"reveal":[152],"with":[155,181,190],"more":[156],"balanced":[157],"distribution":[159],"exhibit":[160],"better":[161,170],"linear":[162],"separability":[163],"node":[165],"representations,":[166],"thus":[167],"leading":[168],"performance.":[172],"also":[174],"conduct":[175],"experiments":[177,201],"using":[178],"synthetic":[179],"varying":[182],"distributions,":[184],"results":[187],"align":[188],"well":[189],"findings.":[193],"Collectively,":[194],"both":[195],"verify":[202],"proposed":[205],"metadata-driven":[206],"effective":[209],"identifying":[211],"critical":[212],"for":[215],"GNNs.":[216]},"counts_by_year":[],"updated_date":"2026-03-12T08:34:05.389933","created_date":"2025-10-10T00:00:00"}