Sutrisno, Muhammad Rafli Dwi (2024) Analisis dan Penerapan Decision Tree Dengan Algoritma Cart dan Optimasi Hiperparameter Menggunakan Grid Search Nested Crossvalidation untuk Memprediksi Kecenderungan Penyebab Klaim pada Produk Asuransi Siswakoe – PT Asuransi Umum Bumiputera Muda 1967. Project Report. [s.n.], [s.l.]. (Unpublished)
![[thumbnail of 5006211080-Project_Report.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5006211080-Project_Report.pdf - Accepted Version Download (2MB) |
Abstract
Laporan ini membahas penerapan algoritma Classification and Regression Tree (CART) dan optimasi hyperparameter menggunakan Grid Search Nested Cross-Validation untuk memprediksi kecenderungan penyebab klaim pada produk asuransi SiswaKoe di PT Asuransi Umum Bumiputera Muda 1967. Penelitian ini dilakukan dengan tujuan memberikan solusi prediktif atas penyebab klaim asuransi berbasis data historis tahun 2023. Data yang digunakan meliputi jenjang pendidikan, jenis asuransi, kota asal, dan penyebab klaim. Proses analisis mencakup data preprocessing, exploratory data analysis, klasifikasi dengan CART, dan evaluasi model menggunakan metrik seperti accuracy, precision, recall, dan F1-score. Hasil menunjukkan bahwa model memiliki akurasi sebesar 83% dan memberikan interpretasi keputusan yang dapat digunakan untuk mendukung strategi bisnis dan manajemen risiko perusahaan. Meskipun hasil optimasi hyperparameter menunjukkan parameter yang serupa dengan nilai default, proses validasi silang tetap penting untuk memastikan performa model yang optimal.
====================================================================================================================================
This report discusses the application of the Classification and Regression Tree (CART) algorithm and hyperparameter optimization using Grid Search Nested Cross-Validation to predict the tendency of claim causes in the SiswaKoe insurance product at PT Asuransi Umum Bumiputera Muda 1967. The study aims to provide predictive solutions based on historical claim data from 2023. The dataset includes education level, insurance type, city of origin, and claim cause. The analysis process involves data preprocessing, exploratory data analysis, classification using CART, and model evaluation using metrics such as accuracy, precision, recall, and F1-score. The model achieved an accuracy of 83% and generated interpretable decision rules that can support the company’s business strategy and risk management. Although the hyperparameter optimization results were similar to the default values, the cross-validation process remains essential to ensure optimal model performance.
| Item Type: | Monograph (Project Report) |
|---|---|
| Uncontrolled Keywords: | Decision Tree, CART, Asuransi, Grid Search, Nested Cross-Validation, Machine Learning, Klasifikasi, Penyebab Klaim Decision Tree, CART, Insurance, Grid Search, Nested Cross-Validation, Machine Learning, Classification, Claim Causes |
| Subjects: | Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > Q Science (General) > Q370 Entropy (Information theory) Q Science > QA Mathematics > QA9.58 Algorithms Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics) |
| Divisions: | Faculty of Science and Data Analytics (SCIENTICS) > Actuaria > 94203-(S1) Undergraduate Thesis |
| Depositing User: | Muhammad Rafli Dwi Sutrisno |
| Date Deposited: | 18 Jul 2025 02:44 |
| Last Modified: | 18 Jul 2025 02:44 |
| URI: | http://repository.its.ac.id/id/eprint/120030 |
Actions (login required)
 |
View Item |