Data mining merupakan pemprosesan sebuah informasi dari suatu database yang dapat digunakan untuk berbagai kebutuhan sektor swasta. Salah satu metode dalam data mining, yaitu Clustering yang bertujuan untuk menemukan pengelompokan dari serangkaian pola, titik, objek maupun dokumen. Algoritma K-Means clustering merupakan algoritma yang berperan penting dalam bidang data mining serta sederhana untuk diimplementasikan dan dijalankan. Selain itu, terdapat pengembangan varian dari metode K-Means Clustering yaitu K-Medoids yang muncul sebagai penanggulangan kelemahan Algoritma K-Means yang sensitif terhadap outlier dikarenakan sebuah objek dengan sebuah nilai yang besar mungkin secara substansial menyimpang dari distribusi data. Penelitian ini membandingkan kedua algoritma pada suatu dataset. Adapun data yang digunakan merupakan data transaksi bongkar muat selama tahun 2017 pada PT Pelabuhan Indonesia I Cabang Dumai berdasarkan atribut agen, keterangan barang, jenis, dan jumlah ton. Dari percobaan yang dilakukan, diperoleh hasil pengolahan KMeans hanya membutuhkan waktu rata-rata 1 detik sedangkan pengolahan data pada K-Medoids membutuhkan waktu rata-rata 1 menit 38 detik pada RapidMiner. Nilai DBI pada K-Means lebih rendah dibandingkan KMedoids yaitu masing-masing 0.112 dan 0.119. Perolehan cluster dominan, menunjukkan bahwa agen Buana Listya Tama TBK, PT mendominasi diikuti agen Samudera Sarana Karunia, PT.
Abstract: Presidential Instruction No. 7 of 2014 mandates PIP to the Ministry of Education and Culture to summarize Indonesia Smart Card (KIP) and spread PIP funds to students that cannot afford to pay education. However, Indonesia Corruption Watch (2018) explained that the data used for the Smart Indonesia Program (PIP) was still inaccurate because almost half of the poor people with a percentage of 42.9% were not registered as participants in the Smart Indonesia Program (PIP). According to ICW, this is due to the data used for the process of determining the candidates for the Smart Indonesia Program recipients of the funds are still inaccurate and harming others who supposed to get funds. One method that usually used as a decision-making technique in the research is the Multi-Objective Optimization Ratio Analysis (MOORA) method which is a multi-criteria decision-making that has five main steps as a technique and it can be used to rank prospective PIP fund recipients based on the highest to the lowest preference values. The results of this studyindicate that the first rank with the highest value was 0.0539 and the last rank with the lowest value was 0.0211 so it used to ease the stakeholders to determine the amount of KIP recipients based on the preference values. This method can be applied for stakeholders needed in compared to monotonous data processing using estimates.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.