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 Model Regresi Logistik Biner dengan dan Tanpa SMOTE pada Ketepatan Waktu Kelulusan Mahasiswa Program Sarjana IPB

      Thumbnail
      View/Open
      Cover (249.4Kb)
      Fulltext (2.376Mb)
      Lampiran (199.1Kb)
      Date
      2024
      Author
      Adinda, Dhea Puspita
      Anisa, Rahma
      Syafitri, Utami Dyah
      Metadata
      Show full item record
      Abstract
      IPB memiliki tujuan menghasilkan lulusan unggul. Hal ini dapat dicapai dengan menghasilkan mahasiswa lulus tepat waktu. Mahasiswa program sarjana dianggap lulus tepat waktu jika menyelesaikan studi dalam waktu maksimal empat tahun. Oleh karena itu, penting dilakukan penelitian untuk mengetahui pengaruh faktor-faktor terhadap ketepatan waktu kelulusan mahasiswa. Salah satu analisis yang digunakan untuk melakukan analisis terhadap respon yang biner (lulus tepat waktu/tidak) adalah analisis regresi logistik biner. Data yang digunakan adalah data wisudawan program sarjana IPB bulan Agustus dan September 2023. Proporsi lulusan tidak tepat waktu sebesar 30% menunjukkan data relatif tidak seimbang, sehingga diterapkan metode SMOTE. Penelitian ini bertujuan untuk membandingkan performa model dengan dan tanpa SMOTE serta mengetahui besar pengaruh faktor-faktor terhadap ketepatan waktu kelulusan mahasiswa. Hasil yang diperoleh menunjukkan model regresi logistik biner tanpa SMOTE lebih baik dibandingkan model dengan SMOTE berdasarkan nilai sensitivitas model tanpa SMOTE yang lebih besar dibandingkan model dengan SMOTE yaitu sebesar 90,7%. Hal ini disebabkan ketidakseimbangan data didominasi oleh mahasiswa lulus tepat waktu, sehingga model tanpa SMOTE mampu menggambarkan mahasiswa lulus tepat waktu dengan baik. Model menunjukkan bahwa perbedaan fakultas, IPK, predikat kelulusan, beasiswa, jenis kelamin, jalur masuk, SKS, UKT, dan keikutsertaan organisasi memiliki pengaruh yang berbeda terhadap kecenderungan mahasiwa lulus tepat waktu.
       
      IPB aims to produce outstanding graduates by ensuring students graduate on time. Undergraduate students are considered to graduate on time if they complete their program within four years. Therefore, it is important to research factors influencing the punctuality of student graduation. Binary logistic regression analysis is used to analyze the binary response. This research used IPB undergraduate program graduates dataset in August and September 2023. The percentage of graduates who did not graduate on time is 30%, implying relatively unbalanced class in the dataset, thus the SMOTE method was applied. This research aims to compare the performance of models with and without SMOTE and to determine the factors influencing the punctuality of student graduation. The results showed that the binary logistic regression model without SMOTE was better than the model with SMOTE. Based on the sensitivity, which was higher at 90,7% compared to the model with SMOTE. Due to the imbalance in data dominated by students graduating on time, the model without SMOTE accurately represents students graduating on time. The model showed that differences in faculty, GPA, graduation predicate, scholarship, gender, admission path, cumulative credit, tuition fee, and organizational participation have different effects on the timeliness of student graduation.
       
      URI
      http://repository.ipb.ac.id/handle/123456789/154707
      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