FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3213091306> ?p ?o ?g. }

• W3213091306 endingPage "359" @default.
• W3213091306 startingPage "342" @default.
• W3213091306 abstract "Link prediction aims to predict missing (or future) links in complex networks. There exist an increasing number of link prediction methods, and each method has its own advantages and disadvantages. However, the problem of how to effectively and efficiently combine different methods is far from understood. Here, we address the problem of how to optimally combine different methods. Mathematically, we present a comprehensive analysis on the mechanism of the hybrid method, and then propose an explainable metric (PNR, Precision-Noise-Ratio) to characterize the prediction accuracy of uncertain links. Algorithmically, we provide a scalable and parameter-free algorithm to optimally combine different methods. Besides, we analytically prove the optimum of the proposed method based on posterior bayesian estimation. Empirically, we compare our method with the state of the art methods in different real datasets. The results not only demonstrate the effectiveness of the proposed method, but also explain why the proposed method could increase the prediction accuracy." @default.
• W3213091306 created "2021-11-22" @default.
• W3213091306 creator A5023686661 @default.
• W3213091306 creator A5026719663 @default.
• W3213091306 creator A5091221734 @default.
• W3213091306 date "2022-01-01" @default.
• W3213091306 modified "2023-10-16" @default.
• W3213091306 title "PNR: How to optimally combine different link prediction approaches?" @default.
• W3213091306 cites W1971040550 @default.
• W3213091306 cites W1979104937 @default.
• W3213091306 cites W2085837260 @default.
• W3213091306 cites W2158275940 @default.
• W3213091306 cites W2186071116 @default.
• W3213091306 cites W2417010772 @default.
• W3213091306 cites W2650688300 @default.
• W3213091306 cites W2799228184 @default.
• W3213091306 cites W2805766156 @default.
• W3213091306 cites W2914866826 @default.
• W3213091306 cites W2914912562 @default.
• W3213091306 cites W2950742983 @default.
• W3213091306 cites W2966694634 @default.
• W3213091306 cites W3124650867 @default.
• W3213091306 cites W843551813 @default.
• W3213091306 doi "https://doi.org/10.1016/j.ins.2021.10.061" @default.
• W3213091306 hasPublicationYear "2022" @default.
• W3213091306 type Work @default.
• W3213091306 sameAs 3213091306 @default.
• W3213091306 citedByCount "2" @default.
• W3213091306 countsByYear W32130913062022 @default.
• W3213091306 countsByYear W32130913062023 @default.
• W3213091306 crossrefType "journal-article" @default.
• W3213091306 hasAuthorship W3213091306A5023686661 @default.
• W3213091306 hasAuthorship W3213091306A5026719663 @default.
• W3213091306 hasAuthorship W3213091306A5091221734 @default.
• W3213091306 hasConcept C107673813 @default.
• W3213091306 hasConcept C11413529 @default.
• W3213091306 hasConcept C119857082 @default.
• W3213091306 hasConcept C124101348 @default.
• W3213091306 hasConcept C126255220 @default.
• W3213091306 hasConcept C154945302 @default.
• W3213091306 hasConcept C162324750 @default.
• W3213091306 hasConcept C176217482 @default.
• W3213091306 hasConcept C21547014 @default.
• W3213091306 hasConcept C2778753846 @default.
• W3213091306 hasConcept C31258907 @default.
• W3213091306 hasConcept C33724603 @default.
• W3213091306 hasConcept C33923547 @default.
• W3213091306 hasConcept C41008148 @default.
• W3213091306 hasConcept C48044578 @default.
• W3213091306 hasConcept C77088390 @default.
• W3213091306 hasConceptScore W3213091306C107673813 @default.
• W3213091306 hasConceptScore W3213091306C11413529 @default.
• W3213091306 hasConceptScore W3213091306C119857082 @default.
• W3213091306 hasConceptScore W3213091306C124101348 @default.
• W3213091306 hasConceptScore W3213091306C126255220 @default.
• W3213091306 hasConceptScore W3213091306C154945302 @default.
• W3213091306 hasConceptScore W3213091306C162324750 @default.
• W3213091306 hasConceptScore W3213091306C176217482 @default.
• W3213091306 hasConceptScore W3213091306C21547014 @default.
• W3213091306 hasConceptScore W3213091306C2778753846 @default.
• W3213091306 hasConceptScore W3213091306C31258907 @default.
• W3213091306 hasConceptScore W3213091306C33724603 @default.
• W3213091306 hasConceptScore W3213091306C33923547 @default.
• W3213091306 hasConceptScore W3213091306C41008148 @default.
• W3213091306 hasConceptScore W3213091306C48044578 @default.
• W3213091306 hasConceptScore W3213091306C77088390 @default.
• W3213091306 hasFunder F4320317337 @default.
• W3213091306 hasFunder F4320321001 @default.
• W3213091306 hasFunder F4320321921 @default.
• W3213091306 hasFunder F4320329236 @default.
• W3213091306 hasLocation W32130913061 @default.
• W3213091306 hasOpenAccess W3213091306 @default.
• W3213091306 hasPrimaryLocation W32130913061 @default.
• W3213091306 hasRelatedWork W1956930971 @default.
• W3213091306 hasRelatedWork W2115000855 @default.
• W3213091306 hasRelatedWork W2368984408 @default.
• W3213091306 hasRelatedWork W2902946190 @default.
• W3213091306 hasRelatedWork W2963058055 @default.
• W3213091306 hasRelatedWork W3026534141 @default.
• W3213091306 hasRelatedWork W3154094704 @default.
• W3213091306 hasRelatedWork W4200196661 @default.
• W3213091306 hasRelatedWork W4210723310 @default.
• W3213091306 hasRelatedWork W4287777341 @default.
• W3213091306 hasVolume "584" @default.
• W3213091306 isParatext "false" @default.
• W3213091306 isRetracted "false" @default.
• W3213091306 magId "3213091306" @default.
• W3213091306 workType "article" @default.