ANALISIS KINERJA ALGORITMA FUZZY C-MEANS DAN K-MEANS PADA DATA KEMISKINAN

ANIQ NOVICIATIE ULFAH, NIM. 10650044 (2014) ANALISIS KINERJA ALGORITMA FUZZY C-MEANS DAN K-MEANS PADA DATA KEMISKINAN. Skripsi thesis, UIN SUNAN KALIJAGA.

Preview	Text (ANALISIS KINERJA ALGORITMA FUZZY C-MEANS DAN K-MEANS PADA DATA KEMISKINAN) BAB I, V, DAFTAR PUSTAKA.pdf Download (2MB) | Preview
	Text ( ANALISIS KINERJA ALGORITMA FUZZY C-MEANS DAN K-MEANS PADA DATA KEMISKINAN ) BAB II, III, IV.pdf Restricted to Registered users only Download (853kB)

Abstract

The Local Government of Gunungkidul creates some formulation in an effort to alleviate poverty. Indicators of poverty used are an integration of national standard indicators with consideration of locality in Gunungkidul. The Local Government of Gunungkidul uses a simple data classification, that is weighting indicators in the grouping of data poverty in data classification of, so it still frequently goes wrong and requires a long time. Therefore, this study aims to perform the data clustering of poverty to determine the appropriate algorithm between FCM algorithm and K - Means with the calculation of poverty indicators in Girijati Village, Purwosari , Gunungkidul , Yogyakarta. Data domain of this study is poverty data with 15 indicators that will produce 3 cluster. Before clustering the data, first performed pretreatment includes data cleaning and data transformation. Calculation of clustering, according 3 poverty criterias in Gunungkidul, done after the data is ready. The result of the calculation is then used to compare the FCM algorithm with K - Means. Based on the analysis of the clustering result of FCM algorithm and K- Means, it shows that time and iteration of FCM algorithm relatively much more than the K-Means algorithm. In addition, the FCM algorithm is more difficult applied to more varied data , unlike the K-Means algorithm that can be applied to data with less variation. The suitability of data between FCM algorithm and the calculation of poverty indicators in the Girijati village is 50 % and for the K - Means algorithm is 83.33 % . Therefore, K- Means algorithm is more appropriately used in data classification of poverty based on the three criteria of poverty, beside FCM algorithm. So, for poverty data domain , more appropriate algorithm used for poverty data domain are K-Means algorithm.

Share this knowledge with your friends :

Actions (login required)

View Item