Perbandingan Kinerja Metode Fuzzy K-Means Clustering Dan Fuzzy Gustafson-Kessel Clustering Berdasarkan Realisasi Pajak Daerah Kota Surabaya

Rahadian, Yanuar Rafi (2018) Perbandingan Kinerja Metode Fuzzy K-Means Clustering Dan Fuzzy Gustafson-Kessel Clustering Berdasarkan Realisasi Pajak Daerah Kota Surabaya. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
1313100091-Undergraduate Thesis.pdf - Accepted Version

Download (1MB) | Preview

Abstract

Kota Surabaya merupakan kabupaten/kota dengan pendapatan asli daerah tertinggi di Indonesia. Di sisi lain, Surabaya memiliki masalah ketimpangan wilayah yang tinggi berdasarkan volume PDRB per kecamatan. Penelitian ini bertujuan mengelompokkan realisasi pajak daerah tingkat kecamatan di Kota Surabaya. Studi simulasi pengelompokan menunjukkan bahwa akurasi metode Fuzzy Gustafson-Kessel (FGK) cenderung lebih tinggi dibandingkan metode Fuzzy K-Means (FKM). Digunakan 7 variabel pajak daerah di Kota Surabaya, yaitu pajak air tanah, pajak hiburan, pajak penerangan jalan, pajak hotel, pajak restoran, pajak parkir serta pajak bumi dan bangunan. Pereduksian variabel menggunakan analisis faktor menghasilkan 2 komponen utama yang menjelas-kan varians total sebesar 73%. Jumlah optimum klaster yang dihasilkan metode FGK adalah 4 sedangkan metode FKM adalah 2 klaster. Perbandingan antara hasil optimum metode FGK dan metode FKM menghasilkan metode FGK yang lebih akurat. Jumlah pajak kecamatan tertinggi ditunjukkan oleh rata-rata tertinggi pada klaster yaitu kecamatan Dukuh Pakis, Gayungan, Genteng, Mulyorejo, Sambikerep, Tegalsari dan Wonokromo. ========================================================================================================= Surabaya city is the highest regency income in Indonesia. On the other hand, Surabaya city has a high regional inequality problem based on GDP total between sub-districts. The objective of this research is to classify the realization of regional taxes in each sub-district in Surabaya. Simulation study shows that Fuzzy Gustafson-Kessel (FGK) method more accurate than Fuzzy K-Means (FKM) method. This research uses 7 variables of regional tax in Surabaya. Factor analysis is used to reduce 7 regional tax variables into 2 main components that can explain the total variance of 73%. FGK method reach 4 clusters as optimum number of clusters, meanwhile 2 clusters for FKM method. Comparing the two results, FGK method is more accurate than FKM method based on icd-rate criteria. The highest tax rate indicated by sub-districts of the highest average cluster that is Dukuh Pakis, Gayungan, Genteng, Mulyorejo, Sambikerep, Tegalsari and Wonokromo.

Item Type: Thesis (Undergraduate)
Additional Information: RSSt 519.535 Rah p-1 3100018074715
Uncontrolled Keywords: Realisasi pajak daerah; analisis faktor; fuzzy k-means; fuzzy Gustafson-Kessel; Realization of regional tax; factor analysis; fuzzy kmeans.
Subjects: Q Science > QA Mathematics > QA9.64 Fuzzy logic
Q Science > QA Mathematics > QA248_Fuzzy Sets
Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics)
Divisions: Faculty of Mathematics, Computation, and Data Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: Yanuar Rafi Rahadian
Date Deposited: 14 Apr 2020 08:56
Last Modified: 14 Apr 2020 08:56
URI: https://repository.its.ac.id/id/eprint/50716

Actions (login required)

View Item View Item