No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
ANALISIS CLUSTER DENGAN METODE SOM UNTUK PENGELOMPOKAN PROVINSI DI INDONESIA BERDASARKAN INDIKATOR KRIMINALITAS
Abstract
Kriminalitas atau kejahatan adalah suatu perbuatan yang dapat mengakibatkan masalah bagi masyarakat. Salah satu teknik analisis untuk mengatasi masalah ini adalah analisis cluster. Hasil analisis ini dapat dijadikan evaluasi bagi pemerintah dalam memperhatikan kriminalitas yang terjadi. Penelitian ini bertujuan untuk mendeskripsikan hasil analisis cluster dan karakteristik dari hasil analisis cluster dengan metode Self-Organizing Maps (SOM) untuk pengelompokan provinsi di Indonesia berdasarkan indikator kriminalitas nasional. Penelitian ini menggunakan data indikator kriminalitas nasional tahun 2021 yang diperoleh dari laman resmi Badan Pusat Statistik Indonesia. Data yang digunakan terdiri dari 34 provinsi dan 7 variabel. Hasil penelitian menunjukkan bahwa 5 cluster adalah jumlah cluster optimal dengan nilai validasi DBI 0,3793. Setiap cluster memiliki karakteristiknya tersendiri. Cluster ke-1 memiliki rata-rata jumlah kejahatan ringan, sedang, berat, dan jumlah kejahatan yang diselesaikan paling rendah, serta selang waktu terjadinya kejahatan paling lama. Cluster ke-2 memiliki rata-rata jumlah kejahatan sedang dan berat paling tinggi, serta selang waktu terjadinya kejahatan paling cepat. Cluster ke-3 memiliki rata-rata jumlah kejahatan ringan paling tinggi. Cluster ke-4 memiliki rata-rata persentase penyelesaian kejahatan dan risiko penduduk terkena kejahatan paling rendah. Cluster ke-5 memiliki rata-rata jumlah kejahatan yang diselesaikan, persentase penyelesaian kejahatan, dan risiko penduduk terkena kejahatan paling tinggi.Kata kunci: Indikator Kriminalitas Nasional, analisis cluster, Self-Organizing Maps, Davies Bouldin Index
ResearchGate has not been able to resolve any citations for this publication.
Lung Cancer is a disease in which there are cells that grow in the lungs by a collection of carcionogens uncontrollably. Lung Cancer can be treated with surgery, chemotherapy and radiotherapy. Early treatment that needs to be done to reduce the mortality rate in patients with lung cancer after performing thoracic surgery, by collecting data from each patients regarding this information causes a new problem, including the data obtained including high-dimensional data and has many attributes so that it can produce less accurate information. So it is necessary to calculate data mining clustering. In general, the methods for performing clustering are grouped into four parts, namely partitioning, hierarchial, grid-based and model-based. This study used the k-medoids algorithm because it is able to handle data sensitive to outliers and has high accuracy and efficiency in processing large numbers objects. The results of the k-medoids calculation were evaluated using the euclidean distance Davies Bouldin Index which resulted in a DBI value of 0.93543 indicating that the k-medoids algoritm achieves good grouping because the final result of the calculation is less than 0. From the results of the evaluation using DBI it shows that the k-medoids algorithm has an average accumulation average at the time of execution is quite fast and the cluster quality is good.
p class="Abstrak">Penggunaan internet dimasyarakat global terus tumbuh, tak hanya terjadi pada masyarakat dewasa melainkan juga pada anak-anak. Internet tidak hanya berdampak pada hal positif melainkan juga pada hal negatif. Di Ternate penggunaan internet terus tumbuh hal ini karena semakin mudah dalam mengakses internet. Namun laporan secara ilmiah mengenai penggunaan internet di Kota Ternate belum ada. untuk itu, bagaimana mengetahui penggunaan internet dikalangan anak SD di kota Ternate. Penelitian itu bertujuan untuk mencari tahu penggunaan internet di Kota Ternate dengan cara survey secara langsung kepada kalangan anak SD di kota Ternate. Selain itu, data-data dari hasil survey kemudian di cluster dengan menggunakan algoritma k-means clustering . kemudian dilakukan validasi clustering dengan davies bouldin index . Hasil dari penelitian ini dari 933 responden diperoleh 51,45 % siswa SD di kota Ternate aktif di jejaring sosial dengan 53,70% di whatsapp , 40,30% di instagram dan 27,80% di facebook. Untuk aktivitas ketika membuka youtube terdapat 61,60% sering menonton video di youtube dengan 61,60% video karton, komedi 49,80% dan konten edukasi 28,40%. Sedangkan untuk game online, yang aktif dalam bermain game online yaitu 49,41%. Untuk penerapan algoritma clustering k-means pada 32 sekolah SD di Kota Ternate diperoleh cluster terbaik saat pembagian 4 cluster, hal ini berdasarkan nilai davies bouldin index yang diperoleh sebesar 0,773 lebih kecil dibandingkan dengan pembagian cluster lainnya.
Abstract
The use of the internet in the global community continues to grow, not only in adults but also in children. The internet does not only have positive effects but also negative things. In Ternate the use of the internet continues to grow because it is easier to access the internet. However, scientific reports regarding the use of the internet in the city of Ternate do not yet exist. for that, how to find out the use of the internet among elementary school children in the city of Ternate. The research aims to find out the use of the internet in the city of Ternate by means of a direct survey among elementary school children in the city of Ternate. In addition, the data from the survey results are then clustered using the k-means clustering algorithm. Then the clustering validation was performed with the bouldin index davies. The results of this study of 933 respondents obtained 51.45% of elementary school students in Ternate were active in social networks with 53.70% on whatsapp, 40.30% on Instagram and 27.80% on Facebook. For activities when opening YouTube there are 61.60% often watching videos on YouTube with 61.60% cardboard videos, comedy 49.80% and educational content 28.40%. As for online games, those active in playing online games are 49.41%. For the application of the k-means clustering algorithm in 32 elementary schools in Ternate, the best cluster was obtained when the division of 4 clusters, this was based on the bouldin index davies value obtained by 0.773 smaller than the other cluster divisions. </p
First part in automatic video analisys is moving object detection. An accurate moving object detection is needed indeed to next step process of automatic video analisys like tracking object detected adn then analyze of detected object. Background Subtraction is a common approach in moving object detection. The common problems in background subtraction are illumination changes, object shadow, and dynamic background like waving tree. Self organizing Maps algorithm apllied in background Subtraction to handles these common problems. Median filtering and morphological operation added after background Subtraction procces in conjunction to increase and produce accurate moving object detection. Apllied SOM, median filtering, and morphological operation in background subtraction increasing object detection accuraccy with value of MSE in 1463,73 and PSNR in 17,035 compare with alpha based background Subtraction where 4268,50 for MSE and 12,018 for PSNR.
Self-organizing map (SOM) is one of the most popular neural network methods for cluster analysis. Clustering methods using SOM usually are two-stage procedures: first original data are projected onto a set of prototypes on an ordered grid by SOM, and these prototypes can be seen as proto-clusters which will be grouped in the second stage to obtain finally clustering results. Many methods have been proposed to cluster the proto-clusters, among which the prototypes are considered as isolated vectors, without relationship of SOM; others are based on the U-matrix, which represents the local distance structure coming from the topology of SOM. In this paper, we propose a novel method more related to SOM for the purpose to cluster the proto-clusters. In the second stage we use the grid information alternatively, regarding it as a graph partitioned by graph cut algorithm well-known as Normalized cut. We apply this method on image processing and seismic data analysis and obtain reasonable results.
K-Means Untuk Klasifikasi Pelanggan Perusahaan
- M A Hendrawan
- R A Hafidz
P., M. A., Hendrawan, R. A., & Hafidz, I. (2014). K-Means Untuk Klasifikasi Pelanggan
Perusahaan. Seminar Nasional Sistem Informasi Indonesia (SESINDO), September, 419-425.