Abstract : This paper proposes a novel approach to compute K-Nearest Neighbors (KNN) algorithm on a large set of users by lever-aging disk and memory efficiently on a commodity PC. The system is designed to minimize random accesses to disk as well as the amount of data loaded/unloaded from/to disk so as to better utilize the computational power, thus improving the algorithmic efficiency.
https://hal.inria.fr/hal-01095557 Contributor : Javier OlivaresConnect in order to contact the contributor Submitted on : Thursday, December 18, 2014 - 1:48:40 PM Last modification on : Thursday, January 20, 2022 - 4:20:00 PM Long-term archiving on: : Monday, March 23, 2015 - 1:20:38 PM
Nitin Chiluka, Anne-Marie Kermarrec, Javier Olivares. Scaling KNN Computation over Large Graphs on a PC. Middleware 2014, Dec 2014, Bourdeaux, France. ⟨10.1145/2678508.2678513⟩. ⟨hal-01095557⟩