Perbandingan 3 Metode Data Mining Dalam Pencarian Pengetahuan Kualitas Pelayanan Kesehatan Bagi Pasien BPJS

Zakkiyah, Tiwik Hanif and , Yusuf Sulistyo Nugroho, S.T., M.Eng. (2016) Perbandingan 3 Metode Data Mining Dalam Pencarian Pengetahuan Kualitas Pelayanan Kesehatan Bagi Pasien BPJS. Skripsi thesis, Universitas Muhammadiyah Surakarta.

	PDF (Naskah Publikasi) naskah publikasi.pdf Download (2MB)
	PDF (Pernyataan Publikasi Ilmiah) pernyataan naskah publikasi.pdf Restricted to Repository staff only Download (234kB) | Request a copy

Abstract

“BPJS Kesehatan” is a social health assurance organizer that is formed by government. The aim of its establishment is in order to give health assurance to citizen. Most of poor people be ruled out by health instance, the high cost of treatment is being the reason. “BPJS Kesehatan” appears along with the hope to become problem solver of health service in Indonesia. Knowing the benefit of BPJS towards health sevice, hopfully the better health service can be obtained by people who have been registered in BPJS program in the entire of health instance. “BPJS Kesehatan” appears along with the hope to become problem solver of health service in Indonesia, such as there is a big gap in health service that is received between those who have good financial with those who have not. Hopefully, people who have been registered in BPJS can receive the equal health service in entire health instance without differences. Based on that case, data mining technique will be used to ge information about the health service quality for BPJS patients in Surakarta. The methods that will be used are Naive Bayes, Decision Tree algoritma Index Gini, dan Rule Induction. Attributees that are used consist of BPJS Users, Category Of Patients, Physical Substantiation, Reliability, Perceptive, Guarantee, and Attention. In this research, RapidMiner 5 is used to analyze the data. The result of this research shows that prediction result of naïve bayes has accuracy percentage 47,89%, precision 46,89%, recall 43,10%. Index gini result shows accuracy percentage 52%, precision 51,58%, recall 49,49%. Rule induction result shows accuracy percentage 53%, precision 52,90%, recall 46,13%. Rule induction method is better to use in this research, viewed from accuracy percentage and precision. It is because the percentage is higher than other method. However, if it is viewed from recall percentage, decision tree method is better than other method. Health service quality for BPJS users in Surakarta is not good. This result can be seen from 301 respondents of BPJS users, 145 of BPJS users state that health service quality is “Good” and “Not Good” statement is stated by 156 BPJS users. While from 299 respondents of non BPJS users, there are 152 respondents state that health service quality is “Good” and 147 respondents state “Not Good”.

Item Type:	Karya ilmiah (Skripsi)
Uncontrolled Keywords:	BPJS, Data Mining, Decision Tree Algoritma Index Gini, Naive Bayes, Rule Induction
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering
Divisions:	Fakultas Ilmu Komunikasi dan Informatika > Teknik Informatika
Depositing User:	TIWIK HANIF ZAKKIYAH
Date Deposited:	02 May 2016 06:33
Last Modified:	02 May 2016 06:33
URI:	http://eprints.ums.ac.id/id/eprint/43224

Actions (login required)

View Item