Matches in SemOpenAlex for { <https://semopenalex.org/work/W3144257051> ?p ?o ?g. }
- W3144257051 abstract "Much data with graph structures satisfy the principle of homophily, meaning that connected nodes tend to be similar with respect to a specific attribute. As such, ubiquitous datasets for graph machine learning tasks have generally been highly homophilous, rewarding methods that leverage homophily as an inductive bias. Recent work has pointed out this particular focus, as new non-homophilous datasets have been introduced and graph representation learning models better suited for low-homophily settings have been developed. However, these datasets are small and poorly suited to truly testing the effectiveness of new methods in non-homophilous settings. We present a series of improved graph datasets with node label relationships that do not satisfy the homophily principle. Along with this, we introduce a new measure of the presence or absence of homophily that is better suited than existing measures in different regimes. We benchmark a range of simple methods and graph neural networks across our proposed datasets, drawing new insights for further research. Data and codes can be found at this https URL." @default.
- W3144257051 created "2021-04-13" @default.
- W3144257051 creator A5037779507 @default.
- W3144257051 creator A5067605731 @default.
- W3144257051 creator A5068905954 @default.
- W3144257051 creator A5079620830 @default.
- W3144257051 date "2021-04-03" @default.
- W3144257051 modified "2023-09-27" @default.
- W3144257051 title "New Benchmarks for Learning on Non-Homophilous Graphs." @default.
- W3144257051 cites W1522301498 @default.
- W3144257051 cites W1836465849 @default.
- W3144257051 cites W1904715515 @default.
- W3144257051 cites W200376986 @default.
- W3144257051 cites W2008620264 @default.
- W3144257051 cites W2033193852 @default.
- W3144257051 cites W2064058256 @default.
- W3144257051 cites W2095705004 @default.
- W3144257051 cites W2103133870 @default.
- W3144257051 cites W2111708605 @default.
- W3144257051 cites W2126209209 @default.
- W3144257051 cites W2130354913 @default.
- W3144257051 cites W2148123869 @default.
- W3144257051 cites W2150256170 @default.
- W3144257051 cites W2154455818 @default.
- W3144257051 cites W2176412452 @default.
- W3144257051 cites W2381162149 @default.
- W3144257051 cites W2519887557 @default.
- W3144257051 cites W2557283755 @default.
- W3144257051 cites W2559839022 @default.
- W3144257051 cites W2624431344 @default.
- W3144257051 cites W2755088640 @default.
- W3144257051 cites W2790076745 @default.
- W3144257051 cites W2805516822 @default.
- W3144257051 cites W2883384506 @default.
- W3144257051 cites W2884040606 @default.
- W3144257051 cites W2900470550 @default.
- W3144257051 cites W2900763475 @default.
- W3144257051 cites W2908510526 @default.
- W3144257051 cites W2913305520 @default.
- W3144257051 cites W2942681667 @default.
- W3144257051 cites W2945848398 @default.
- W3144257051 cites W2962751458 @default.
- W3144257051 cites W2962897394 @default.
- W3144257051 cites W2963858333 @default.
- W3144257051 cites W2963893572 @default.
- W3144257051 cites W2964114465 @default.
- W3144257051 cites W2964124573 @default.
- W3144257051 cites W2970843311 @default.
- W3144257051 cites W2975966238 @default.
- W3144257051 cites W2995345478 @default.
- W3144257051 cites W2995509042 @default.
- W3144257051 cites W3007332492 @default.
- W3144257051 cites W3030661718 @default.
- W3144257051 cites W3034492151 @default.
- W3144257051 cites W3087257704 @default.
- W3144257051 cites W3090999459 @default.
- W3144257051 cites W3094504436 @default.
- W3144257051 cites W3120233338 @default.
- W3144257051 cites W3120587221 @default.
- W3144257051 cites W3122063025 @default.
- W3144257051 cites W3123408649 @default.
- W3144257051 cites W3133222630 @default.
- W3144257051 cites W893486657 @default.
- W3144257051 cites W3021975806 @default.
- W3144257051 hasPublicationYear "2021" @default.
- W3144257051 type Work @default.
- W3144257051 sameAs 3144257051 @default.
- W3144257051 citedByCount "7" @default.
- W3144257051 countsByYear W31442570512021 @default.
- W3144257051 countsByYear W31442570512022 @default.
- W3144257051 crossrefType "posted-content" @default.
- W3144257051 hasAuthorship W3144257051A5037779507 @default.
- W3144257051 hasAuthorship W3144257051A5067605731 @default.
- W3144257051 hasAuthorship W3144257051A5068905954 @default.
- W3144257051 hasAuthorship W3144257051A5079620830 @default.
- W3144257051 hasConcept C114614502 @default.
- W3144257051 hasConcept C119857082 @default.
- W3144257051 hasConcept C132525143 @default.
- W3144257051 hasConcept C153083717 @default.
- W3144257051 hasConcept C154945302 @default.
- W3144257051 hasConcept C2522767166 @default.
- W3144257051 hasConcept C2779812341 @default.
- W3144257051 hasConcept C33923547 @default.
- W3144257051 hasConcept C41008148 @default.
- W3144257051 hasConcept C80444323 @default.
- W3144257051 hasConceptScore W3144257051C114614502 @default.
- W3144257051 hasConceptScore W3144257051C119857082 @default.
- W3144257051 hasConceptScore W3144257051C132525143 @default.
- W3144257051 hasConceptScore W3144257051C153083717 @default.
- W3144257051 hasConceptScore W3144257051C154945302 @default.
- W3144257051 hasConceptScore W3144257051C2522767166 @default.
- W3144257051 hasConceptScore W3144257051C2779812341 @default.
- W3144257051 hasConceptScore W3144257051C33923547 @default.
- W3144257051 hasConceptScore W3144257051C41008148 @default.
- W3144257051 hasConceptScore W3144257051C80444323 @default.
- W3144257051 hasLocation W31442570511 @default.
- W3144257051 hasOpenAccess W3144257051 @default.
- W3144257051 hasPrimaryLocation W31442570511 @default.
- W3144257051 hasRelatedWork W2070265204 @default.
- W3144257051 hasRelatedWork W2125869687 @default.