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 Naïve Bayes dalam Prediksi Kekambuhan Kanker Tiroid Berdiferensiasi Baik

      Thumbnail
      View/Open
      Cover (267.8Kb)
      Fulltext (597.0Kb)
      Lampiran (97.60Kb)
      Date
      2025
      Author
      Pahlevi, Muhammad Dylan
      Alamudi, Aam
      Sumertajaya, I Made
      Metadata
      Show full item record
      Abstract
      Kanker tiroid biasanya bisa disembuhkan secara total, namun kekambuhan dapat terjadi setelah beberapa tahun. Untuk mengurangi waktu prediksi kekambuhan kanker tiroid, dapat digunakan algoritma machine learning seperti Naïve Bayes. Penelitian ini bertujuan untuk menerapkan algoritma Naïve Bayes dalam memprediksi kekambuhan kanker tiroid berdiferensiasi baik. Penelitian ini juga bertujuan untuk mengetahui nilai kepentingan tiap peubah dan membandingkan algoritma Naïve Bayes dengan algoritma machine learning lain yang dilakukan Borzooei et al. (2023). Dataset dibagi menjadi tiga jenis, yaitu dataset tanpa peubah Risk, dataset yang hanya memiliki peubah Risk dan Recurred, dan dataset lengkap. Hasil penelitian menunjukkan bahwa Naïve Bayes menunjukkan kinerja yang baik dalam memprediksi kekambuhan kanker tiroid. Dataset 3 menghasilkan model terbaik karena memiliki nilai accuracy sebesar 0,97, recall sebesar 0,926, precision sebesar 0,962, specificity sebesar 0,986, dan AUC sebesar 0,991. Selain itu, Model 3 bagus dalam menangani data tidak seimbang karena memiliki nilai balanced accuracy sebesar 0,956 dan f1-score macro sebesar 0,962. Melalui metode permutation importance, peubah yang paling penting adalah Risk. Peubah penting lainnya adalah Response, T, dan N. Ketika dibandingkan dengan algoritma lain, Naïve Bayes dapat menjadi metode alternatif yang cepat dan efisien karena memiliki nilai metrik yang tidak berbeda jauh dibandingkan dengan algoritma lain.
       
      Thyroid cancer is usually curable, but recurrence can occur after several years. To reduce the time for predicting thyroid cancer recurrence, machine learning algorithms like Naïve Bayes can be used. This study aims to apply the Naïve Bayes algorithm to predict differentiated thyroid cancer recurrence. It also seeks to determine the importance of each feature and compare Naïve Bayes with other machine learning algorithms used by Borzooei et al. (2023). The dataset was divided into three types: a dataset without the Risk, a dataset containing only the Risk and Recurred, and a complete dataset. The results show that Naïve Bayes performs well in predicting thyroid cancer recurrence. Dataset 3 produced the best model, with an accuracy of 0.97, recall of 0.926, precision of 0.962, specificity of 0.986, and AUC of 0.991. Additionally, Model 3 is effective at handling imbalanced data with a balanced accuracy of 0.956 and a macro F1-score of 0.962. Using permutation importance, the most important feature is Risk, followed by Response, T, and N. Compared to other algorithms, Naïve Bayes is a fast and efficient alternative with similar performance metrics.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/160650
      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