UPT. PERPUSTAKAAN
Institut Teknologi Sepuluh Nopember Surabaya
Kampus ITS Sukolilo - Surabaya 60111
Phone
:
031-5921733 , 5923623
Fax
:
031-5937774
E-mail
:
libits@its.ac.id
Website
:
http://library.its.ac.id
Support (Customer Service) :
timit_perpus@its.ac.id
Welcome..guys!
Have a problem with your access?
Please, contact our technical support below:
LIVE SUPPORT
Davi Wahyuni
Tondo Indra Nyata
Anis Wulandari
Ansi Aflacha
ITS » Master Theses » Statistika - S2 Posted by fandikaaqsa@its.ac.id at 26/02/2016 14:33:48 • 1978 Views
A HYBRID MODEL FUZZY C-MEANS FCM AND MULTIVARIATE ADAPTIVE REGRESSION SPLINE MARS ON THE CASE OF POOR HOUSEHOLDS IN JOMBANG DISTRICT
Author : SUKMAWATY, YUANA ( 1312201014 )
ABSTRAK
Kemiskinan merupakan salah satu permasalahan utama yang menjadi
pusat perhatian pemerintah dalam proses pembangunan nasional. Namun kriteria
kemiskinan dan cara pandang yang berbeda-beda akan menimbulkan penafsiran
yang berbeda-beda pula mengenai jumlah penduduk miskin kriteria penduduk
miskin dan tingkat penanganan terhadap masalah kemiskinan sehingga
diperlukan suatu proses penggalian informasi tersembunyi dalam data tersebut
yang dikenal dengan data mining. Metode yang dikenal dalam data mining
diantaranya adalah metode pengelompokkan dengan pendekatan clustering dan
klasifikasi. Namun saat kondisi keanggotaan suatu data untuk dikelompokkan
tidak jelas batasannya himpunan fuzzy digunakan untuk mengantisipasi hal
tersebut. Konsep himpunan fuzzy ini kemudian yang mendasari berkembangnya
metode fuzzy clustering dimana salah satu pendekatan metode ini adalah Fuzzy CMeans
FCM. Adapun untuk kasus rumah tangga miskin Kabupaten Jombang yang
terdiri dari banyak variabel prediktor pendekatan regresi nonparametrik dapat
digunakan Multivariate Adaptive Regression Splines MARS untuk memperoleh
model hubungan variabel prediktor terhadap respon dan besarnya ketepatan
klasifikasi yang dihasilkan dari pemodelan rumah tangga miskinnya. Penggabungan
metode FCM dan MARS ini menghasilkan cluster terbaik sebanyak 3 cluster yang
kemudian dimodelkan dengan MARS. Kemudian hasil ketepatan klasifikasinya
dibandingkan dengan model MARS dengan respon berupa status rumah tangga
miskin ketetapan Badan Pusat Statistik BPS dan diperoleh bahwa nilai
sensitivity specificity dan accuracy untuk ketepatan klasifikasi MARS respon
hasil pembentukan cluster dengan FCM lebih baik dibandingkan hasil klasifikasi
model MARS dengan respon status rumah tangga miskin ketetapan BPS.
ABSTRACT
Poverty is one of the main problems that becomes the focus in the process
of national development. However criteria poverty and different view will cause
different interpretation about poor population poverty criteria and handling of
poverty problem that needed poverty data analyzed well so we need a process of
extracting information hidden in the data known as data mining. Method in data
mining is a method of grouping them with clustering and classification
approaches. However as a condition of membership of the data to be grouped is
not clearly defined fuzzy set is used to anticipate it. The concept of fuzzy set is
then that underlie the development of fuzzy clustering method where one
approach is the method of Fuzzy C-Means FCM. As for the case of poor
households Jombang consisting of many predictor variables nonparametric
regression approach can be used Multivariate Adaptive Regression Splines
MARS to obtain a model of the relationship of the predictor variables and the
response magnitude of the resulting classification accuracy of the modeling of
poor households. A hybrid Method of MARS and FCM can be produce better
clusters as much as 3 clusters than modeled with MARS only. Then the results of
the classification accuracy compared with MARS models with a response status
of poor households Badan Pusat Statistik BPS statutes and found that the value
of sensitivity specificity and accuracy of classification MARS response to the
formation of clusters with FCM results better than classification results compared
with the MARS models response status of poor households BPS statutes.
Source Master Thesis of Statistics, RTSt 519.536 Suk h, 2014
Coverage ITS Community
Rights Copyright @2016 by ITS Library. This publication is protected by copyright and per obtained from the ITS Library prior to any prohibited reproduction, storage in a re transmission in any form or by any means, electronic, mechanical, photocopying, reco For information regarding permission(s), write to ITS Library