PENERAPAN ALGORITMA SVM BERBASIS PSO UNTUK TINGKAT PELAYANAN MARKETING TERHADAP LOYALITAS PELANGGAN KARTU KREDIT

  • Elin Panca Saputra (1*) Manajemen Informatika AMIK BSI Jakarta

  • (*) Corresponding Author
Keywords: evaluation of the service, particle swarm optimization, support vector machine, the selection of attributes.

Abstract

This research will be used method of support vector machine and will do the selection of attributes by using particle swarm optimization to determine the level of service. After the test results obtained are support vector machine produces an accuracy value of 92.25%, 95.98% and a precision value AUC value of 0.976% then be selected attributes using particle swarm optimization attributes, amounting to 8 predictor variables selected two attributes used. The results showed higher accuracy value that is equal to 93.75%, 93.91% and a precision value AUC value of 0.973%. Thus increasing the accuracy of 1.5%, and increased the AUC of 0.006. With accuracy and AUC values, the algorithm of support vector machines based on particle swarm optimization in the category of classification is very good.

Published
2015-09-15
How to Cite
Saputra, E. (2015). PENERAPAN ALGORITMA SVM BERBASIS PSO UNTUK TINGKAT PELAYANAN MARKETING TERHADAP LOYALITAS PELANGGAN KARTU KREDIT. Jurnal Techno Nusa Mandiri, 12(2), 109-118. https://doi.org/10.33480/techno.v12i2.440
Article Metrics

Abstract viewed = 24 times
PDF downloaded = 39 times