Ana gezinime atla Aramaya atla Ana içeriğe atla

Yüksek-derece çizge yapilari ile maksimum klik sayma problemine yönelik buluşsal yöntemleri öǧrenme

Araştırma sonucu: Kitap/Rapor/Konferans Bildirisinde BölümKonferans katkısıbilirkişi

1 Alıntı (Scopus)

Özet

Recently, various NP-hard combinatorial optimization problems have been solved by learned heuristics using complex learning models. In particular, node classification in graphs has been a helpful method towards finding the decision boundary to distinguish nodes in an optimal set from the rest. In this work, we investigate the role of local graphlet counts surrounding a node as graph features in the node classification towards solving the maximum clique enumeration problem. Graphlets are small induced subgraphs, whose local and global frequencies have been important features in the analysis of networks. We use a learning framework to identify the nodes that belong to some maximum clique of the network. Consequently, this idea is used in a pruning process to reduce the runtime of the maximum clique enumeration problem. Besides the high accuracy of the results, the performance of this framework is shown to be scalable and robust. The method presented here is applicable to networks from all sizes and can be used in estimating the solution of other graph search problems with high complexity.

Tercüme edilen katkı başlığıLearning heuristics for the maximum clique enumeration problem using higher-order graph structures
Orijinal dilTürkçe
Ana bilgisayar yayını başlığıSIU 2021 - 29th IEEE Conference on Signal Processing and Communications Applications, Proceedings
YayınlayanInstitute of Electrical and Electronics Engineers Inc.
ISBN (Elektronik)9781665436496
DOI'lar
Yayın durumuYayınlandı - 9 Haz 2021
Etkinlik29th IEEE Conference on Signal Processing and Communications Applications, SIU 2021 - Virtual, Istanbul, !!Turkey
Süre: 9 Haz 202111 Haz 2021

Yayın serisi

AdıSIU 2021 - 29th IEEE Conference on Signal Processing and Communications Applications, Proceedings

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???29th IEEE Conference on Signal Processing and Communications Applications, SIU 2021
Ülke/Bölge!!Turkey
ŞehirVirtual, Istanbul
Periyot9/06/2111/06/21

Keywords

  • Graphlet
  • Machine learning model
  • Maximum clique enumeration problem
  • Node classification

Parmak izi

Yüksek-derece çizge yapilari ile maksimum klik sayma problemine yönelik buluşsal yöntemleri öǧrenme' araştırma başlıklarına git. Birlikte benzersiz bir parmak izi oluştururlar.

Bundan alıntı yap