View Item 
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Statistics and Data Sciences
      • View Item
      •   IPB Repository
      • Dissertations and Theses
      • Undergraduate Theses
      • UT - Faculty of Mathematics and Natural Sciences
      • UT - Statistics and Data Sciences
      • View Item
      JavaScript is disabled for your browser. Some features of this site may not work without it.

      Penerapan Algoritma K-Means, K-Medoids, dan Fuzzy C-Means dalam Pemetaan Indikator Ketenagakerjaan di Jawa Tengah 2022

      Thumbnail
      View/Open
      Cover (467.2Kb)
      Fulltext (1.286Mb)
      Lampiran (95.75Kb)
      Date
      2025
      Author
      Maulany, Syifa
      Silvianti, Pika
      Kurnia, Anang
      Metadata
      Show full item record
      Abstract
      Ketenagakerjaan mencerminkan kondisi ekonomi dan sosial suatu wilayah melalui indikator seperti partisipasi angkatan kerja dan tingkat pengangguran. Namun, tantangan dalam sektor ketenagakerjaan seperti pengangguran dan kesenjangan pendidikan terus meningkat. Penelitian ini bertujuan untuk melakukan segmentasi kabupaten/kota di Provinsi Jawa Tengah tahun 2022 berdasarkan indikator ketenagakerjaan menggunakan algoritma K-Means, K-Medoids, dan Fuzzy C-Means, serta membandingkan keefektifan ketiga metode tersebut dan mengidentifikasi persebaran kabupaten/kota berdasarkan hasil penggerombolan. Tiga algoritma penggerombolan yang digunakan, yaitu K-Means dengan 6 gerombol, K-Medoids dengan 6 gerombol, dan FCM dengan 5 gerombol. Evaluasi metrik menunjukkan bahwa K-Means unggul dalam Calinski-Harabasz Index (9.400714) dan Davies-Bouldin Index (1.248888), menandakan pemisahan dan kekompakan gerombol yang baik. Sementara, FCM memiliki nilai Silhouette Index tertinggi (0.4636), menunjukkan keanggotaan gerombol yang jelas. K-Medoids menunjukkan performa terendah dengan nilai Silhouette Index (0.2012), meskipun mampu menangani pencilan. Dengan demikian, K-Means cocok digunakan pada data dengan struktur yang jelas dan berfokus pada pemisahan gerombol, sedangkan FCM dapat mengatasi kekompleksitasan dan ketidakpastian dengan membentuk gerombol yang lebih fleksibel (adaptif). Hasil penggerombolan mengidentifikasi wilayah di Jawa Tengah dengan indikator ketenagakerjaan yang tertinggal, seperti Kota Magelang, Kota Salatiga, Kota Surakarta, Kota Tegal, Banyumas, Tegal, Brebes, dan Cilacap. Seluruh komponen indikator ketenagakerjaan di daerah tersebut masih berada pada angka yang buruk.
       
      Employment reflects the economic and social conditions of a region through indicators such as labor force participation and unemployment rates. However, challenges in the employment sector, such as unemployment and educational disparities, continue to escalate. This study aims to segment the regencies/cities in Central Java Province in 2022 based on employment indicators using the K-Means, K-Medoids, and Fuzzy C-Means algorithms, compare the effectiveness of these three methods, and identify the distribution of regencies/cities based on the clustering results. The clustering algorithms applied include K-Means with six clusters, K-Medoids with six clusters, and FCM with five clusters. Evaluation metrics indicate that K-Means outperformed others in the Calinski-Harabasz Index (9.400714) and Davies-Bouldin Index (1.248888), demonstrating good cluster separation and compactness. Meanwhile, FCM achieved the highest Silhouette Index (0.4636), indicating clear cluster membership. K-Medoids showed the lowest performance with a Silhouette Index of 0.2012, although it effectively handled outliers. K-Means is suitable for datasets with clear structures and a focus on cluster separation, while FCM accommodates complexity and uncertainty by forming more flexible clusters. The clustering results identified underperforming regions in Central Java, such as Magelang City, Salatiga City, Surakarta City, Tegal City, Banyumas, Tegal, Brebes, and Cilacap, where employment indicators remain poor.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/160646
      Collections
      • UT - Statistics and Data Sciences [2260]

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository
        

       

      Browse

      All of IPB RepositoryCollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

      My Account

      Login

      Application

      google store

      Copyright © 2020 Library of IPB University
      All rights reserved
      Contact Us | Send Feedback
      Indonesia DSpace Group 
      IPB University Scientific Repository
      UIN Syarif Hidayatullah Institutional Repository
      Universitas Jember Digital Repository