Kinasih, Shela Retno (2026) IMPLEMENTASI DECISION TREE FAKTOR YANG MEMPENGARUHI RISIKO GAGAL BAYAR PADA PLATFORM P2P LENDING DI JAWA TIMUR. S1 thesis, Universitas PGRI Madiun.
![[img]](https://eprint.unipma.ac.id/style/images/fileicons/text.png) |
Text
Halaman Depan.pdf Download (637kB) |
|
Text
ABSTRAK.pdf Download (205kB) |
|
Text
BAB I.pdf Download (478kB) |
|
Text
BAB II.pdf Download (566kB) |
|
Text
BAB III.pdf Restricted to Repository staff only Download (426kB) |
|
Text
BAB IV.pdf Restricted to Repository staff only Download (410kB) |
|
Text
BAB V.pdf Download (245kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (295kB) |
|
Text
Lampiran.pdf Restricted to Repository staff only Download (4MB) |
Abstract
This study aims to analyze the factors influencing default risk among Peer-to-Peer (P2P) lending platform users in East Java. Investigating this phenomenon is critical given that East Java Province ranks third in Indonesia for the highest accumulated online loan debt, which is predominantly driven by the productive age group. Using an explanatory quantitative approach, this research implements the C4.5 (J48) Decision Tree machine learning algorithm. Data were gathered through a questionnaire survey of 385 respondents, comprising Generation Z and Millennials across five major regions: Surabaya, Malang, Sidoarjo, Jember, and Banyuwangi. The independent variables analyzed include gender, interest rates, loan tenor, loan limit, and credit history. The attribute selection results indicate that credit history is the most dominant and significant predictor of default risk, yielding the highest Gain Ratio of 0.52372. Loan tenor also emerges as an important factor serving as an internal node with a Gain Ratio of 0.02037, where a medium-term tenor (6–12 months) is identified as the optimal threshold for risk mitigation. Conversely, gender, interest rates, and loan limits exhibit minimal influence and do not function as primary splitters in the decision tree structure. Model performance evaluation demonstrates high reliability, achieving an accuracy of 94.78% and an Area Under Curve (AUC) of 0.954, falling under the "Excellent Classification" category. This study concludes that historical financial behavior serves as the most accurate predictor of future repayment failures. Based on these findings, P2P lending platforms are recommended to exercise greater selectivity in tenor configurations and tighten the integration of credit history data to mitigate non-performing loans.
| Item Type: | Thesis/Skripsi/Tugas Akhir (S1) |
|---|---|
| Kata Kunci: | P2P Lending, Risiko Gagal Bayar, Decision Tree, Algoritma C4.5, Jawa Timur. P2P Lending, Default Risk, Decision Tree, C4.5 Algorithm, East Java. |
| Subjects: | A Social Sciences > HB Economic Theory A Social Sciences > HG Finance A Social Sciences > HM Sociology |
| Divisions: | Fakultas Ekonomi dan Bisnis > Manajemen |
| Depositing User: | KINASIH RETNO SHELA |
| Date Deposited: | 26 Oct 2026 03:44 |
| Last Modified: | 26 Oct 2026 03:44 |
| URI: | http://eprint.unipma.ac.id/id/eprint/10699 |
Actions (login required)
 |
View Item |
