Fahdina, Audrey (2022) Klasifikasi Status Keterangan Polis Asuransi Jiwa Syariah Menggunakan Regresi logistik Biner, Naive Bayes, dan Support Vector Machine (Studi Kasus : PT. Asuransi Syariah Keluarga Indonesia). Other thesis, Institut Teknologi Sepuluh Nopember.
Text
06311840000057-Undergraduate_Thesis.pdf - Accepted Version Download (2MB) |
Abstract
Segala ketentuan pada asuransi tertulis dalam polis asuransi, salah satunya adalah jatuh tempo pembayaran biaya kontribusi, dimana apabila biaya kontribusi tidak dibayarkan hingga jatuh tempo yang telah ditentukan maka status keterangan polis asuransi akan berubah dari polis aktif menjadi polis tidak aktif (lapse), hal tersebut merupakan salah satu penyebab perubahan status keterangan polis yang berdampak dapat merugikan kedua belah pihak. Oleh karena itu akan dilakukan penelitian berupa klasifikasi status keterangan polis asuransi jiwa syariah di PT Asuransi Syariah Keluarga Indonesia guna menjaga kejelasan dari permasalahan status keterangan polis asuransi nasabahnya. Analisis tersebut menggunakan Regresi Logistik Biner, Naïve Bayes dan Support Vector Machine (SVM), dengan menggunakan data yang didapatkan dari PT Asuransi Syariah Keluarga Indonesia berupa “Data Kepesertaan Asuransi” pada tahun 2016-2021 dan klasifikasi akan dikategorikan ke dalam polis aktif dan polis tidak aktif (lapse). Hasil analisis klasifikasi pada metode Regresi Logistik Biner memperoleh nilai akurasi sebesar 66,79%, pada metode Naïve Bayes nilai akurasinya sebesar 92,66% dan pada metode Support Vector Machine (SVM) nilai akurasinya sebesar 99%. Berdasarkan nilai akurasi terbesar didapatkan bahwa pada penelitian klasifikasi status keterangan polis asuransi jiwa syariah di PT Asuransi Syariah Keluarga Indonesia hasil analisis klasifikasi terbaik adalah menggunakan metode Support Vector Machine (SVM).
========================================================================================================================================
All provisions on insurance are written in the insurance policy, one of which is the due date for payment of the contribution fee. If the contribution fee is not paid by the specified maturity, the status of the insurance policy statement will change from an active policy to an inactive policy (lapse). This is one of the causes of the change in the status of the policy statement, which has an impact that can be detrimental to both parties. Therefore, this study will be conducted on the classification status of the notification sharia life insurance policy statement at PT Asuransi Syariah Keluarga Indonesia in order to maintain clarity of the status problems of the customer's insurance policy statement. The analysis uses Binary Logistics Regression, Nave Bayes and Support Vector Machine (SVM) using data obtained from PT Asuransi Syariah Keluarga Indonesia. This is of "Insurance Participation Data" in 2016–2021, and the classification will be categorized into active policies and non-active policies (lapse). The results of the classification analysis on the Binary Logistics Regression method obtained an accuracy value of 66.79%, the Naïve Bayes method has an accuracy value of 92.66%, and the Support Vector Machine (SVM) method has an accuracy value of 99%. Based on the greatest accuracy value, it was found that in the study of the classification status of notification sharia life insurance policy at PT Asuransi Syariah Keluarga Indonesia, the best classification analysis was performed using the Support Vector Machine (SVM) method.
Item Type: | Thesis (Other) |
---|---|
Additional Information: | RSAk 519.536 Fah k-1 2022 |
Uncontrolled Keywords: | Naïve Bayes, Regresi Logistik Biner, Status keterangan polis, SVM |
Subjects: | Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression |
Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis |
Depositing User: | - Davi Wah |
Date Deposited: | 30 Oct 2023 09:01 |
Last Modified: | 05 Nov 2024 01:51 |
URI: | http://repository.its.ac.id/id/eprint/105014 |
Actions (login required)
View Item |