Analisis Kredit Pembayaran Biaya Kuliah Dengan Pendekatan Pembelajaran Mesin

Penulis

  • Arliyanti Nurdin Institut Teknologi Telkom Surabaya, Surabaya
  • Rizqa Amelia Zunaidi Institut Teknologi Telkom Surabaya, Surabaya
  • Muhammad Arkan Fauzan Wicaksono Institut Teknologi Telkom Surabaya, Surabaya
  • Agi Lobita Japtara Martadinata Institut Teknologi Telkom Surabaya, Surabaya

DOI:

https://doi.org/10.25126/jtiik.20231026301

Abstrak

Salah satu tantangan dalam institusi keuangan adalah manajemen risiko kredit. Hal ini juga terjadi pada institusi pendidikan swasta dimana pengelolaan keuangan dilakukan secara mandiri serta sumber dana mayoritas berasal dari mahasiswa. Setiap institusi harus menjamin kesehatan finansial melalui monitoring cashflow. Adanya penundaan atau kredit pembayaran biaya kuliah mahasiswa akan mempengaruhi cashflow institusi. Oleh karena itu dibutuhkan analisis kredit sebagai tindakan preventif guna mencegah terjadinya kredit yang bermasalah dan meminimalkan risiko kredit lainnya yang timbul di kemudian hari. Pada penelitian ini, algoritma machine learning digunakan untuk analisis kredit pembayaran biaya kuliah pada perguruan tinggi. Dataset yang digunakan adalah data riwayat tagihan, transaksi pembayaran, dan data pengajuan kredit/ angsuran. Tahap perancangan sistem terdiri dari preprocessing, pemilihan fitur, pemodelan, pengujian dan evaluasi hasil. Berdasarkan hasil pengujian algoritma dengan kinerja terbaik adalah KNN dengan recall untuk prediksi “gagal bayar” sebesar 0,8 dan prediksi “berhasil” sebesar 0,76.  Model machine learning ini kemudian ditanamkan dalam sebuah sistem informasi analisis kredit biaya kuliah. Selain itu juga sistem akan memberikan skor setiap pengajuan berdasarkan metode scorecard. Semakin tinggi skor kredit semakin kecil risiko gagal bayarnya. Skor kredit ini berkisar antara 250 – 600. Jika kredit yang diajukan diprediksi “gagal bayar” dengan skor kredit rendah atau berpotensi menjadi piutang macet, sistem akan merekomendasikan untuk menilik ulang skema pengajuan kredit dari mahasiswa tersebut agar mahasiswa tetap dapat melanjutkan pendidikan dan cash collection ratio tetap baik.

 

Abstract

One of the challenges in financial institutions is credit risk management. This also occurs in private educational institutions where financial management is carried out independently and most of funding sources come from students. Each institution must ensure financial health through cashflow monitoring. Any delay or credit in paying student tuition fees will affect the institution's cashflow. Therefore, credit analysis is needed as a preventive measure to prevent non-performing loans and minimize other credit risks that arise in the future. In this study, machine learning algorithms are used for credit analysis for paying tuition fees activity at universities. The datasets used are billing history data, payment transactions, and credit/installment application data. The system design stage consists of preprocessing, feature selection, modeling, uji and evaluation of results. Based on the results of uji the algorithm with the best performance is KNN with a recall for the prediction of "failure to pay" of 0,8 and prediction of "success" of 0,76. This machine learning model is then embedded in a tuition credit analysis information system. In addition, the system will provide a score for each submission based on the scorecard method. The higher the credit score, the lower the risk of default. This credit score ranges from 250 – 600. If the proposed credit is predicted to be "in default" with a low credit score or has the potential to become bad debts, the system will recommend reviewing the student's credit application scheme so that students can continue their education and cash collection ratio remains good.


Downloads

Download data is not yet available.

Referensi

BARREAU, B. (n.d.). Machine Learning for Financial Products Recommendation. https://tel.archives-ouvertes.fr/tel-02974918

CHEN, Z., VAN KHOA, L. D., TEOH, E. N., NAZIR, A., KARUPPIAH, E. K., & LAM, K. S., 2018. Machine learning techniques for anti-money laundering (AML) solutions in suspicious transaction detection: a review. Knowledge and Information Systems, 57(2), 245–285. https://doi.org/10.1007/s10115-017-1144-z

DAMRONGSAKMETHEE, T., & NEAGOE, V.-E., 2019. Principal Component Analysis and ReliefF Cascaded with Decision Tree for Credit Scoring (pp. 85–95). https://doi.org/10.1007/978-3-030-19810-7_9

DASTILE, X., CELIK, T., & POTSANE, M., 2020. Statistical and machine learning models in credit scoring: A systematic literature survey. Applied Soft Computing, 91, 106263. https://doi.org/10.1016/j.asoc.2020.106263

DUMITRESCU, E., HUÉ, S., HURLIN, C., & TOKPAVI, S., 2022. Machine learning for credit scoring: Improving logistic regression with non-linear decision-tree effects. European Journal of Operational Research, 297(3), 1178–1192. https://doi.org/10.1016/j.ejor.2021.06.053

GOH, R. Y., & LEE, L. S., 2019. Credit Scoring: A Review on Support Vector Machines and Metaheuristic Approaches. Advances in Operations Research, 2019, 1–30. https://doi.org/10.1155/2019/1974794

HARRIS, T., 2015. Credit scoring using the clustered support vector machine. Expert Systems with Applications, 42(2), 741–750. https://doi.org/10.1016/j.eswa.2014.08.029

LEO, M., SHARMA, S., & MADDULETY, K., 2019. Machine learning in banking risk management: A literature review. Risks, 7(1). https://doi.org/10.3390/risks7010029

LIU, F., HUA, Z., & LIM, A., 2015. Identifying future defaulters: A hierarchical Bayesian method. European Journal of Operational Research, 241(1), 202–211. https://doi.org/10.1016/j.ejor.2014.08.008

LIU, W., FAN, H., & XIA, M., 2021. Step-wise multi-grained augmented gradient boosting decision trees for credit scoring. Engineering Applications of Artificial Intelligence, 97, 104036. https://doi.org/10.1016/j.engappai.2020.104036

MARINAKIS, Y., MARINAKI, M., DOUMPOS, M., MATSATSINIS, N., & ZOPOUNIDIS, C., 2008. Optimization of nearest neighbor classifiers via metaheuristic algorithms for credit risk assessment. Journal of Global Optimization, 42(2), 279–293. https://doi.org/10.1007/s10898-007-9242-1

MUNKHDALAI, L., WANG, L., PARK, H. W., & RYU, K. H., 2019. Advanced Neural Network Approach, Its Explanation with LIME for Credit Scoring Application (pp. 407–419). https://doi.org/10.1007/978-3-030-14802-7_35

NAEEM SIDDIQI., 2012. Credit risk scorecards: developing and implementing intelligent credit scoring (Vol. 3). John Willey & Sons.

ORESKI, S., & ORESKI, G., 2014. Genetic algorithm-based heuristic for feature selection in credit risk assessment. Expert Systems with Applications, 41(4), 2052–2064. https://doi.org/10.1016/j.eswa.2013.09.004

RITTER, G. (N.D.). MACHINE LEARNING FOR TRADING. https://ssrn.com/abstract=3015609

SHEN, F., WANG, R., & SHEN, Y., 2019. A COST-SENSITIVE LOGISTIC REGRESSION CREDIT SCORING MODEL BASED ON MULTI-OBJECTIVE OPTIMIZATION APPROACH. Technological and Economic Development of Economy, 26(2), 405–429. https://doi.org/10.3846/tede.2019.11337

VARMEDJA, D., KARANOVIC, M., SLADOJEVIC, S., ARSENOVIC, M., & ANDERLA, A., 2019. Credit Card Fraud Detection - Machine Learning methods. 2019 18th International Symposium INFOTEH-JAHORINA (INFOTEH), 1–5. https://doi.org/10.1109/INFOTEH.2019.8717766

Diterbitkan

14-04-2023

Terbitan

Bagian

Ilmu Komputer

Cara Mengutip

Analisis Kredit Pembayaran Biaya Kuliah Dengan Pendekatan Pembelajaran Mesin. (2023). Jurnal Teknologi Informasi Dan Ilmu Komputer, 10(2), 271-280. https://doi.org/10.25126/jtiik.20231026301