PENERAPAN K-MEANS CLUSTERING UNTUK PENGELOMPOKAN CLUSTER PASIEN BERDASAR DATA PASIEN SIMPUS

Nadhij Hakiman Alim, NIM: 17106050026 (2021) PENERAPAN K-MEANS CLUSTERING UNTUK PENGELOMPOKAN CLUSTER PASIEN BERDASAR DATA PASIEN SIMPUS. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (PENERAPAN K-MEANS CLUSTERING UNTUK PENGELOMPOKAN CLUSTER PASIEN BERDASAR DATA PASIEN SIMPUS) 17106050026_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (PENERAPAN K-MEANS CLUSTERING UNTUK PENGELOMPOKAN CLUSTER PASIEN BERDASAR DATA PASIEN SIMPUS) 17106050026_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (4MB)

Abstract

During the pandemic, the government continued to encourage improvements in the health sector through information technology. At the puskesmas level, this can be seen by adopting an information system at the puskesmas level or better known as SIMPUS. This information system is a place where large amounts of health data are collected. However, not everything went smoothly, the sudden adoption of technology, with huge amounts of data constantly accumulating, caused many unaccustomed employees to become overwhelmed. Therefore, it is necessary to apply algorithms in structuring, observing and analyzing data. This study aims to directly apply the K-Means clustering algorithm to the data in the Puskesmas Information System. The K-Means algorithm was chosen because it is the most widely used clustering analysis process algorithm. The selection of this algorithm is expected to be the beginning of the adoption of other algorithms. Implementation is done by creating a program/system that can perform K-Means Clustering analysis complete with an easy-to-use interface. In this study, it was concluded that the application of the K-Means Clustering Algorithm in a program was successfully carried out with good results. The trial conducted found the optimal number of clusters = 10 which has a DBI score: 0.714042, it was found that the file has data with the percentage of patient stature: group 1 = 47%, group 2 = 9%. group 3 = 4%, group 4 = 2%, group 5 = 3%, group 6 = 4%, lean height group 7 = 9%, group 8 = 8%, group 9 = 8%, group 10 = 6% .. The results of the average speed of the average execution time are considered satisfactory, with the average execution time = 9,773 and the ready speed = 2,846 data per second. Clustering process speed performance is very satisfactory with the average speed of the four Scaling methods is 1.460324 seconds. Each of the existing scaling methods has a different speed. the first order is the Standard Scaling method (1.441934 seconds), the second order is Strong Scaling (1.448861 seconds), the third order is NO-Scaling (without scaling) (1.450657 seconds) and the last order is MinMax Scaling (1.499845 second).

Share this knowledge with your friends :

Actions (login required)

View Item