LES DERNIÈRES INFORMATIONS
Message d'état
PURL test ID: finland
Nearest Neighbors Graph Construction: Peer Sampling to the Rescue
Titre | Nearest Neighbors Graph Construction: Peer Sampling to the Rescue |
Publication Type | Conference Paper |
Year of Publication | 2016 |
Authors | Benkaouz, Y, Erradi, M, Kermarrec, A-M |
Editor | Abdulla, PA, DelporteGallet, C |
Conference Name | Networked Systems, NETYS 2016 |
ISBN Number | 978-3-319-46140-3; 978-3-319-46139-7 |
Abstract | In this paper, we propose an efficient KNN service, called KPS (KNN-Peer-Sampling). The KPS service can be used in various contexts e.g. recommendation systems, information retrieval and data mining. KPS borrows concepts from P2P gossip-based clustering protocols to provide a localized and efficient KNN computation in large-scale systems. KPS is a sampling-based iterative approach, combining randomness, to provide serendipity and avoid local minimum, and clustering, to ensure fast convergence. We compare KPS against the state of the art KNN centralized computation algorithm NNDescent, on multiple datasets. The experiments confirm the efficiency of KPS over NNDescent: KPS improves significantly on the computational cost while converging quickly to a close to optimal KNN graph. For instance, the cost, expressed in number of pairwise similarity computations, is reduced by approximate to 23% and approximate to 49% to construct high quality KNN graphs for Jester and MovieLens datasets, respectively. In addition, the randomized nature of KPS ensures eventual convergence, not always achieved with NNDescent.
|
DOI | 10.1007/978-3-319-46140-3\_4 |
Contactez-nous
ENSIAS
Avenue Mohammed Ben Abdallah Regragui, Madinat Al Irfane, BP 713, Agdal Rabat, Maroc
Télécopie : (+212) 5 37 68 60 78
Secrétariat de direction : 06 61 48 10 97
Secrétariat général : 06 61 34 09 27
Service des affaires financières : 06 61 44 76 79
Service des affaires estudiantines : 06 62 77 10 17 / n.mhirich@um5s.net.ma
Résidences : 06 61 82 89 77
Contacts
Education - This is a contributing Drupal Theme
Design by
WeebPal.