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.

      Perbandingan Pengelompokan Wanita Anemia dan Non-Anemia dengan Diskriminan Logistik dan Clustering Non-Hierarki Menggunakan Data Campuran

      Thumbnail
      View/Open
      Cover (480.9Kb)
      Fullteks (1.901Mb)
      Lampiran (133.2Kb)
      Date
      2022-08
      Author
      Husna, Amalia Nailul
      Aidi, Muhammad Nur
      Anisa, Rahma
      Julianti, Elisa Diana
      Metadata
      Show full item record
      Abstract
      Anemia merupakan kondisi ketika kadar Hemoglobin (Hb) kurang dari normal, yaitu kurang dari 12 g/dL untuk wanita usia subur. Analisis untuk mendeteksi ketergantungan peubah pendukung risiko anemia penting dilakukan untuk membedakan status wanita anemia dan non-anemia dengan menggunakan metode pengelompokan. Metode pengelompokan yang digunakan pada penelitian ini yaitu diskriminan logistik dan analisis cluster k-prototype, dengan alasan peubah yang digunakan bertipe numerik dan kategorik. Diskriminan logistik mengelompokkan amatan berdasarkan peubah respon dan menghasilkan suatu model, serta peubah yang memengaruhi pengelompokan dapat diketahui dari koefisien model yang signifikan. Analisis cluster k-prototype mengelompokkan amatan berdasarkan ukuran kemiripan, peubah yang memengaruhi pengelompokan dapat diketahui menggunakan uji rata-rata dan uji proporsi. Pada penelitian ini dilakukan pengelompokan wanita anemia dan non-anemia menggunakan diskriminan logistik yang merupakan pengelompokan terbimbing dan analisis cluster k-prototype yang merupakan pengelompokan tidak terbimbing dengan tujuan melihat perbedaan kedua metode tersebut berdasarkan peubah yang memengaruhi pengelompokan dan keanggotaan hasil pengelompokan. Peubah yang memengaruhi pengelompokan menggunakan diskriminan logistik dan analisis cluster k-prototype diperoleh hasil yang berbeda pada peubah penyakit pneumonia. Keanggotaan pengelompokan mengunakan diskriminan logistik dan analisis cluster k-prototype menghasilkan keanggotaan yang berbeda. Diskriminan logistik menghasilkan lebih banyak amatan benar negatif, sedangkan analisis cluster k-prototype menghasilkan amatan benar positif lebih banyak.
       
      Anemia is a condition when the hemoglobin (Hb) level is less than normal, which is less than 12 g/dL for women of reproductive age. It is important to conduct an analysis for detecting the dependence of supporting variables for anemia risk in order to distinguish the status of anemic and non-anemic women by using the grouping method. The grouping method used in this research was logistic discriminant and k-prototype cluster analysis, to facilitate numerical and categorical variables used in this research. The logistic discriminant groups the observations based on the response variables and obtain a model, and the variables that affect the grouping can identified from the significant coefficients. The k-prototype cluster analysis groups observations based on similarity distance. The variables that affect the clustering can be identified using the average test and the proportion test. In this study, anemia and non-anemia women were grouped using a logistic discriminant which is supervised learning and a k-prototype cluster analysis which is unsupervised learning with the aim of seeing the differences between the two methods based on the variables that affect the grouping and membership of the grouping results. Variables that affect grouping using logistic discriminant and k-prototypes cluster analysis had the different results on pneumonia variable. The membership of grouping using logistic discriminant and k-prototype cluster analysis had different result. The logistic discriminant obtained more true negative observations, while the k-prototype cluster analysis obtained more true positive observations.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/113229
      Collections
      • UT - Statistics and Data Sciences [1212]

      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