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

      Perbandingan Kinerja Model Klasifikasi Support Vector Machine dan K-Nearest Neighbors dalam Mendiagnosis Anemia

      Thumbnail
      View/Open
      Cover (2.296Mb)
      Fulltext (3.366Mb)
      Lampiran (2.102Mb)
      Date
      2024
      Author
      Al-Fariz, Buya
      Julianto, Mochamad Tito
      Mangku, I Wayan
      Metadata
      Show full item record
      Abstract
      Anemia merupakan masalah kesehatan yang ditandai dengan penurunan kadar hemoglobin di bawah batas normal. Pendekatan yang lebih baik dalam diagnosis anemia diperlukan untuk memastikan penanganan yang tepat. Penelitian ini bertujuan membandingkan kinerja beberapa model klasifikasi pada machine learning untuk mendiagnosis anemia menggunakan data dengan variabel gender, hemoglobin, hematocrit, MCH, MCHC, MCV, RBC, dan variabel anemia. Empat model yang digunakan, yaitu support vector machine (SVM) kernel linear, support vector machine (SVM) kernel polynomial, support vector machine (SVM) kernel radial basis function, dan k-nearest neighbors (KNN). Data diolah menggunakan metode Mahalanobis distance untuk penanganan pencilan, min-max scaler untuk transformasi data, dan hyperparameters tuning untuk mengoptimalkan kinerja model. Berdasarkan hasil penelitian, SVM kernel polynomial menunjukkan kinerja terbaik dari segi accuracy, precision, recall, dan f1-score. Namun, dari sisi efisiensi, KNN menjadi yang paling unggul. Sementara itu, SVM kernel linear menunjukkan keseimbangan terbaik antara kinerja dan efisiensi sehingga dapat dipertimbangkan sebagai alternatif yang optimal dalam diagnosis anemia.
       
      Anemia is an health problem characterized by a decrease in hemoglobin levels below normal limits. A better approach in the diagnosis of anemia is needed to ensure proper treatment. This study aims to compare the performance of several classification models in machine learning to diagnose anemia using data with gender, hemoglobin, hematocrit, MCH, MCHC, MCV, RBC, and anemia variables. Four models are used, namely support vector machine (SVM) linear kernel, support vector machine (SVM) polynomial kernel, support vector machine (SVM) radial basis function kernel, and k-nearest neighbors (KNN). Data are processed using the Mahalanobis distance method for outlier handling, min-max scaler for data transformation, and hyperparameters tuning to optimize model performance. Based on the results, SVM kernel polynomial showed the best performance in terms of accuracy, precision, recall, and f1-score. However, in terms of efficiency, KNN is the most superior. Meanwhile, linear kernel SVM shows the best balance between performance and efficiency so that it can be considered as an optimal alternative in anemia diagnosis.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/159128
      Collections
      • UT - Mathematics [1487]

      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