Pengelompokan Penderita Tuberkulosis dalam Rumah Tangga di Surabaya dengan Metode CART (Classification and Regression Trees) Bagging

Wisaksono, Adheala Nathasya (2017) Pengelompokan Penderita Tuberkulosis dalam Rumah Tangga di Surabaya dengan Metode CART (Classification and Regression Trees) Bagging. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
1313100013-Undergraduate _Thesis.pdf
Restricted to Repository staff only

Download (5MB) | Request a copy

Abstract

Tuberkulosis di Indonesia merupakan penyebab kematian utama ketiga setelah penyakit jantung dan saluran pernafasan. Angka kejadian tuberkulosis di Jawa Timur mencapai 44.086 kasus pada tahun 2015 dengan Surabaya sebagai daerah dengan kasus tertinggi tuberkulosis di Jawa Timur. Setidaknya 10% kasus tuberkulosis terjadi di Surabaya dengan 2.330 kasus BTA Positif dan 2.409 kasus BTA Negatif. Tuberkulosis dikelompokkan menjadi dua kategori berdasarkan hasil pemeriksaaan dahak yaitu BTA Positif dan BTA Negatif di mana risiko penularan BTA Positif lebih tinggi daripada risiko penularan BTA Negatif. Metode yang sesuai dalam pengelompokan kasus tuberkulosis adalah CART (Classification and Regression Trees) dan bagging karena mampu menggambarkan karakteristik penderita BTA Positif dan BTA Negatif dalam rumah tangga. Hasil penelitian ini yaitu ketepatan klasifikasi analisis CART pada data testing sebesar 76% dan penggunaan bagging tidak memberikan peningkatan ketepatan klasifikasi. Sedangkan faktor-faktor yang berpengaruh terhadap kejadian tuberkulosis adalah penghasilan KRT per bulan, tempat pembuangan air limbah, tingkat pendidikan KRT, sumber air bersih yang digunakan dan status gizi. ================================================================= Tuberculosis in Indonesia is third main cause of death after heart disease and respiratory disorder. In East Java, there were 44.086 cases of tuberculosis in 2015 while 10% of this cases happened in Surabaya which made Surabaya as the biggest area with tuberculosis in East Java. About 2.330 cases of BTA Positive and 2.409 cases of BTA Negative happened in Surabaya by 2015. Tuberculosis is classified into two different categories based on the result of mucus inspection which is BTA Positive and BTA Negative where the transmission risk of BTA Positive is bigger than BTA Negative. The use of CART (Classification and Regression Trees) and bagging is expected to be able to define the characteristics of each category. The result of this research are the accuracy of CART in data testing is 76% and the use of bagging in CART doesn’t give the increase of the accuracy. While the influential factors are income of the household’s head, the disposal wastage place, education of the household’s head, source of clean water and nutrition status.

Item Type: Thesis (Undergraduate)
Additional Information: RSSt 519.536 Wis p
Uncontrolled Keywords: Bagging, CART, ketepatan klasifikasi, tuberkulosis, Accuracy, bagging
Subjects: Q Science > QA Mathematics > QA278.2 Regression Analysis
R Medicine > R Medicine (General)
Divisions: Faculty of Mathematics and Science > Statistics > (S1) Undergraduate Theses
Depositing User: Adheala Nathasya Wisaksono
Date Deposited: 14 Dec 2017 07:41
Last Modified: 14 Dec 2017 07:41
URI: http://repository.its.ac.id/id/eprint/47838

Actions (login required)

View Item View Item