Penentuan Rekomendasi Investasi Berdasarkan Model Evaluasi Kredit Pada Peer-to-Peer Lending

Apriwibowo, Tubagus Irkham Izzata (2021) Penentuan Rekomendasi Investasi Berdasarkan Model Evaluasi Kredit Pada Peer-to-Peer Lending. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 05111740000012-Undergraduate_Thesis.pdf] Text
05111740000012-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (2MB) | Request a copy

Abstract

Pesatnya perkembangan zaman dan teknologi memudahkan berbagai macam sektor dalam melakukan kegiatan transaksional, salah satunya dalam hal penanaman modal atau biasa disebut investasi. Salah satu media untuk melakukan proses transaksi keuangan tersebut adalah Peer-to-Peer (P2P) Lending. P2P lending adalah salah satu bentuk pinjaman daring yang dapat memberikan modal kepada individu atau bisnis serta untuk investasi. Akan tetapi, P2P lending dapat memberikan risiko kerugian yang tinggi terhadap pemberi pinjaman (lender) karena tiap peminjam (borrower) memiliki kemungkinan untuk gagal bayar. Untuk dapat menurunkan risiko kerugian, dibutuhkan suatu analisis dan rekomendasi investasi untuk lender agar dapat mengetahui mitra mana yang dapat memberikan keuntungan. Analisis pada Tugas Akhir ini adalah membangun sebuah model risiko kredit menggunakan algoritma Logistic Regression, Gaussian Naïve Bayes, dan XGBoost. Evaluasi yang dilakukan adalah membandingkan hasil akurasi dari ketiga model. Setelah mendapatkan model dengan hasil akurasi terbaik, dilakukan perekomendasian berdasarkan prediksi kelas peminjam serta perhitungan expected loss, expected gain, dan expected return untuk lender. Berdasarkan hasil eksperimen algoritma, XGBoost memberikan akurasi 97% yang mana memiliki selisih 2-3% lebih tinggi dibanding kedua algoritma lainnya.
=====================================================================================================
The rapid development of times and technology makes it easier for various sectors to carry out transactional activities, one of which is in terms of investment or commonly called investment. One of the media to process these financial transactions is Peer-to-Peer (P2P) Lending. P2P lending is a form of online loan that can provide capital to individuals or businesses as well as for investment. However, P2P lending can pose a high risk of loss to lenders because each borrower has the possibility of default. To be able to reduce the risk of loss, an analysis and investment recommendations are needed for lenders in order to find out which partners can provide benefits. The analysis in this final project is to build a credit risk model using Logistic Regression, Gaussian Naïve Bayes, and XGBoost algorithms. The evaluation carried out is to compare the results of the accuracy of the three models. After getting the model with the best accuracy results, recommendations are made based on the predictions of the borrower class and the calculation of expected loss, expected gain, and expected return for lenders. Based on the experimental results of the algorithm, XGBoost provides 97% accuracy which has a difference of 2-3% higher than the other two algorithms.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: P2P lending, Risiko kredit, Rekomendasi, Investasi P2P lending, Credit risk, Recommendation, Investment
Subjects: H Social Sciences > HG Finance > HG4910 Investments
Q Science > Q Science (General) > Q325.5 Machine learning.
Q Science > QA Mathematics > QA76.9.D343 Data mining. Querying (Computer science)
T Technology > T Technology (General) > T57.5 Data Processing
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information Technology > 59201-(S1) Undergraduate Thesis
Depositing User: Tubagus Irkham Izzata Apriwibowo
Date Deposited: 11 Aug 2021 03:28
Last Modified: 11 Aug 2021 03:28
URI: http://repository.its.ac.id/id/eprint/85384

Actions (login required)

View Item View Item