Inayah, Riza (2014) Klasifikasi Rumah Tangga Miskin Di Kabupaten Jombang Berdasarkan Faktor-Faktor Yang Mempengaruhi Dengan Pendekatan CART - Classification Of Poor Households In Jombang District Based On Influences Factor Using CART (Classification And Regression Trees) Approach. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
1312105018-Undergraduate Thesis.pdf - Published Version Download (2MB) | Preview |
Abstract
Tingkat kemiskinan di Kabupaten Jombang dapat dikatakan
relatif sulit bergerak turun dimana masih ada kurang lebih 73.720
Kepala Keluarga (KK) miskin dari 344 ribu KK yang tersebar di 302
desa dan 4 kelurahan. Pada penelitian ini ingin ditentukan klasifikasi
status kemiskinan rumah tangga miskin di Kabupaten Jombang
berdasarkan faktor-faktor yang diduga dominan mempengaruhi hasil
klasifikasi dengan pendekatan CART (Classification and Regresion
Trees). Unit penelitian adalah rumah tangga miskin di kabupaten
Jombang sebanyak 73.720 rumah tangga. Variabel penelitian meliputi
satu variable respon yaitu status kemiskinan rumah tangga miskin
(Rumah Tangga Sangat Miskin atau RTSM, Rumah Tangga Miskin atau
RTM, dan Rumah Tangga Hampir Miskin atau RTHM) dan ada delapan
belas variabel prediktor. Penerapan pendekatan pohon klasifikasi
menunjukkan bahwa variabel terpenting atau paling dominan
berpengaruh dalam menentukan status kemiskinan suatu rumah tangga
miskin pada penelitian ini yaitu besarnya penghasilan rata-rata per
bulan (Rp) dengan skor tingkat kepentingan variabel sebesar 100.
Keakuratan klasifikasi yang dihasilkan pohon optimal untuk data
learning sebesar 40,986 persen sedangkan data testing sebesar 39,654
persen.
========================================================================================================================
Poverty levels in Jombang District is relatively difficult to move
down. There are about 73.720 Head of Households from 344 thousand
families who are spread out in 302 villages and 4 sub-district. This
research studied the poverty class levels of poor households in Jombang
District based on dominantly significant factors which affect the
classification using CART (Classification and Regression Trees)
approach. Analyzing unit in this research is poor households as many as
73.720 households. The response variable of this research is three
poverty class levels, they are very poor households (RTSM), poor
households (RTM), and almost poor households (RTHM). There are
eight teen predictor variables. The most important variable in optimal
classification tree for classifying poor household levels in Jombang
District is average income per month (Rp) with the score variable
interest rate of it is 100. Classification accuracy produced by the tree
optimal for learning data is 40.986 percent while 39.654 percent for
testing data
| Item Type: | Thesis (Undergraduate) |
|---|---|
| Additional Information: | RSSt 519.536 Ina k |
| Uncontrolled Keywords: | CART, Jombang District, Poor Households Classification, CART, Kabupaten Jombang, Klasifikasi Rumah Tangga Miskin |
| Subjects: | H Social Sciences > HA Statistics > HA29 Theory and method of social science statistics |
| Divisions: | Faculty of Mathematics and Science > Statistics |
| Depositing User: | ansi aflacha |
| Date Deposited: | 03 Dec 2019 08:10 |
| Last Modified: | 03 Dec 2019 08:10 |
| URI: | http://repository.its.ac.id/id/eprint/72171 |
Actions (login required)
 |
View Item |