Fintech Score: 핀테크 서비스 데이터가 개인신용평가에 미치는 영향 연구
Fintech Score: The Effect of Fintech Service Data on Personal Credit Assessment
1
조회
0
다운로드
4
인용
0
참고문헌
키워드
초록
핀테크 비즈니스 환경에서 포용적 금융 서비스를 위해서는 금융이력 부족자(Thin-filer)에 대한 고려가 필수적이다. 현재 개인신용평가모형은 전통적인 금융이력 데이터를 기반으로 신용등급이 산출되어, 금융이력 부족자의 경우 정상적인 개인신용평가를 진행하는데 한계가 존재한다. 또한, 원활한 금융 서비스를 위해서는 고객 세분화를 통한 자본적정성 및 수익성 개선이 필요하다. 하지만, 전통적인 금융이력 데이터만을 가지고 산출된 등급은 고객을 세분화하기 어렵고, 특정 등급에 분포해 있어 다양한 비즈니스 전략에 적용하기에는 한계가 존재한다. 본 논문에서는 금융소외계층의 금융접근성을 개선하고 고객 세분화를 통해서 다양한 금융 서비스를 제공하고자 핀테크 스코어(Fintech Score)를 제안한다. 핀테크 스코어는 핀테크 서비스 사용 로그 데이터와 온라인 및 오프라인 간편 결제 데이터 및 고객들 간의 간편 송금 데이터를 중심으로 모형을 개발하였다. 핀테크 스코어 모형은 핀테크 서비스 데이터만 가지고 변별력 있는 변수를 생성하였으며, 고객 세분화를 통해서 기존 금융이력 부족자들에게 다양한 금융서비스를 제공할 수 있다. 또한, 개인신용평가모형 개발방법론에 준용하여 모형을 설계 및 개발하였으며 개발된 모형의 안정성, 변별력을 계량적으로 검증하였다.
Abstract
Consideration of thin-filer is essential for inclusive financial services in the fintech business environment. The current personal credit rating model calculates credit ratings based on traditional financial history data, so there is a limit to conducting a normal personal credit evaluation for those who lack financial history. In addition, for smooth financial services, it is necessary to improve capital adequacy and profitability through customer segmentation. However, it is difficult to segment customers based only on traditional financial history data, and because they are distributed in specific grades, it is difficult to apply them to various business strategies. In this paper, we propose a Fintech Score to improve financial accessibility of the financially underprivileged and provide various financial services through customer segmentation. The fintech score was developed based on fintech service usage log data, online and offline simple payment data, and simple remittance data between customers. The fintech score model creates a variable with only fintech service data, and through customer segmentation, various financial services can be provided to those with insufficient financial history. In addition, the model was designed and developed using development methodology of the personal credit evaluation model, and the stability and discrimination power of the developed model were quantitatively verified.
참고문헌 (18)
- The Global Findex Database 2017: Measuring Financial Inclusion and the Fintech Revolution, "D-K. Asli", World Bank Group, ~ (2017)
- A. J. Ohlson, "Financial Ratios and the Probabilistic Prediction of Bankruptcy", Journal of Accounting Research, 18(1), pp.109~131 (1980)
- J-Y. Kim, "Development of a Personal Credit Scoring Model(TELCO Score) Using Big Data in Telecommunications", PhD, pp.2019
- 김종윤, "온라인 상거래 데이터를 반영한 개인신용평가모형(커머스스코어) 개발", 정보화연구, 16(1), pp.45~55 (2019), 한국엔터프라이즈아키텍처학회[DOI]
- D. Durand, "Risk Elements in Consumer Instalment Financing" (1941), National Bureau of Economic Research, Inc
- V. B. Djeundje, "Enhancing credit scoring with alternative dat", Expert Systems with Applications, 163 (2021)
- T. Groenfeldt, "Lenddo Creates Credit Scores Using Social Media", ~ (2015)
- J. S. Pedro, "MobiScore:Towards Universal Credit Scoring from Mobile Phone Data, Conference of User Modeling", Adaptation and Personalization (UMAP), ~ (2015)
- C. Sarraute, "A study of age and gender seen through mobile phone usage patterns in mexico", Advances in Social Networks Analysis and Mining (ASONAM), 2014 IEEE, 2014, ~
- Y. Leo, "Alvarez-Hamelin JI, Sarraute C, Karsai M. Socioeconomic correlations and stratification in social- communication networks", Journal of The Royal Society Interface (2016)
- M. Haenlein, "A social network analysis of customerlevel revenue distribution", Marketing letters (2011)
- M. Óskarsdóttir, "Social network analytics for churn prediction in telco: Model building, evaluation and network architecture", Expert Systems with Applications (2017)
- A. Backiel, "Combining local and social network classifiers to improve churn prediction", Proceedings of the 2015 IEEE, 2015, ~[DOI]
- G. Guo, "From Footprint to Evidence: An Exploratory Study of Mining Social Data for Credit Scoring", ACM Trans. Web 10, 4, Article 22, ~ (2016)