Open AccessPENERAPAN ALGORITMA GLOWWORM SWARM OPTIMIZATION PADA MODEL GEOGRAPHICALLY WEIGHTED REGRESSION DENGAN KERNEL ADAPTIF
Author(s) -
I GEDE HARDI KARMANA,
LUH PUTU IDA HARINI,
KETUT JAYANEGARA,
I PUTU EKA NILA KENCANA
Publication year - 2020
Publication title -
e-jurnal matematika
Language(s) - English
Resource type - Journals
ISSN - 2303-1751
DOI - 10.24843/mtk.2020.v09.i01.p282
Subject(s) - kernel (algebra) , computer science , bandwidth (computing) , artificial intelligence , kernel regression , k nearest neighbors algorithm , particle swarm optimization , geographically weighted regression , algorithm , regression , mathematics , pattern recognition (psychology) , mathematical optimization , statistics , combinatorics , computer network
This study aimed to apply glowworm swarm optimization (GSO) algorithm as an alternate way to obtain optimal bandwidth in geographically weighted regression (GWR) model with adaptive kernel function. The result showed that GSO was able to obtain optimal bandwidth with lower cross validation (CV) value than the traditional way that was using k-nearest neighbor (KNN) algorithm. Unfortunately, the running time of GSO was far slower than KNN was.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom