• The Journal of Society for e-Business Studies
• /
• v.23 no.3
• /
• pp.207-224
• /
• 2018

In this article, an empirical study was conducted by using public dataset from Lending Club Corporation, the largest online peer-to-peer (P2P) lending in the world. We explore significant predictor variables related to P2P lending default that housing situation, length of employment, average current balance, debt-to-income ratio, loan amount, loan purpose, interest rate, public records, number of finance trades, total credit/credit limit, number of delinquent accounts, number of mortgage accounts, and number of bank card accounts are significant factors to loan funded successful on Lending Club platform. We developed online P2P lending default prediction models using discriminant analysis, logistic regression, neural networks, and decision trees (i.e., CART and C5.0) in order to predict P2P loan default. To verify the feasibility and effectiveness of P2P lending default prediction models, borrower loan data and credit data used in this study. Empirical results indicated that neural networks outperforms other classifiers such as discriminant analysis, logistic regression, CART, and C5.0. Neural networks always outperforms other classifiers in P2P loan default prediction.

• The Journal of Information Systems
• /
• v.25 no.2
• /
• pp.79-110
• /
• 2016

Purpose Online Peer-to-peer lending (hereafter P2P lending), is a new method of lending money to unrelated individuals through an online financial intermediary. Usually in the online P2P transaction, individuals who would like to borrow money (hereafter borrowers) and those who would like to lend money (hereafter lenders) have no previous relationship. Based on enormous previous studies, this study develops an integrated model, particularly for the online P2P lending environment in China, to better understand the critical factors that influence lenders' intention to lend money through the online P2P lending platform. Design/methodology/approach In order to verify the hypotheses, we develop a questionnaire with 42 survey items. We measured all the items on a five-point Likert-type scale. We use Sojump.com to collect questionnaire and gather 246 valid responses from registered members of Renrendai.com. We analyzed the main survey data by using SPSS 18.0 and AMOS 20.0. We first estimated the reliability, validity, composite reliability and AVE and then conduct common method bias test. The mediating role of trust in platform and in borrower has been tested. Last we tested the hypotheses through the structural model. Findings The results reveal that service quality, information quality, structural assurance, awareness and reputation significantly impact lenders' trust in the online P2P lending platform. Second, awareness, reputation and perceived risk significantly impact lenders' trust in borrower and lending intention. Third, trust propensity has a positive effect on lenders' trust on borrower. Last, awareness, reputation, perceived risk, platform trust and borrower trust can directly impact lenders' lending intention.

• Knowledge Management Research
• /
• v.21 no.3
• /
• pp.229-247
• /
• 2020

In the remarkable growth of P2P financial platform in the field of knowledge management, only companies with big data and machine learning technologies are surviving in fierce competition. The ability to analyze borrowers' credit is most important, and platform companies are also recognizing this capability as the most important business asset, so they are building a credit evaluation system based on artificial intelligence. Nonetheless, online P2P platform providers that offer related services only act as intermediaries to apply for investors and borrowers, and all the risks associated with the investments are attributable to investors. For investors, the only way to verify the safety of investment products depends on the reputation of P2P companies from newspaper and online website. Time series information such as delinquency rate is not enough to evaluate the early stage of Korean P2P makers' credit analysis capability. This study examines the credit analysis procedure of P2P loan platform using artificial intelligence through the case analysis method for well known the top three companies that are focusing on the credit lending market and the kinds of information data to use. Through this, we will improve the understanding of credit analysis techniques through artificial intelligence, and try to examine limitations of credit analysis methods through artificial intelligence.

• Journal of Digital Convergence
• /
• v.20 no.4
• /
• pp.185-192
• /
• 2022

This study investigates the effect of the semi-supervised learning(SSL) method on predicting default risk of peer-to-peer(P2P) loans. Despite its proven performance, the supervised learning(SL) method requires labeled data, which may require a lot of effort and resources to collect. With the rapid growth of P2P platforms, the number of loans issued annually that have no clear final resolution is continuously increasing leading to abundance in unlabeled data. The research data of P2P loans used in this study were collected on the LendingClub platform. This is why an SSL model is needed to predict the default risk by using not only information from labeled loans(fully paid or defaulted) but also information from unlabeled loans. The results showed that in terms of default risk prediction and despite the use of a small number of labeled data, the SSL method achieved a much better default risk prediction performance than the SL method trained using a much larger set of labeled data.

• The Journal of Society for e-Business Studies
• /
• v.15 no.4
• /
• pp.285-301
• /
• 2010

Peer-to-peer (P2P) lending is an open marketplace for loans not from bank but from individuals online. Financial transactions are facilitated directly between individuals ("peers") without any intermediation of a traditional financial institution. A market study by renowned research company forecasts that P2P lending will grow very fast and a couple of P2P lending sites in Korea also are getting attentions by providing the alternative financial services. In P2P lending market, Lender will enjoy higher income generated from the loans in the form of interest than interest that can be earned by financial products provided by official financial institutions. Furthermore, lenders are able to decide who they would lend the money for themselves. Meanwhile, borrowers with low credit scores are able to finance their liquidity requirement with low cost and convenient access to the Internet. The objective of this paper is to introduce P2P lending and its issues of information asymmetry. We provide the insights from the case study of one of P2P lending sites in Korea and review the issues in P2P lending market as research topics. Specifically, information asymmetry issues in both traditional financial institutions and P2P lending are discussed.

• Journal of Digital Convergence
• /
• v.17 no.9
• /
• pp.71-78
• /
• 2019

This study aims to identify good borrowers within the context of P2P lending. P2P lending is a growing platform that allows individuals to lend and borrow money from each other. Inherent in any loans is credit risk of borrowers and needs to be considered before any lending. Specifically in the context of P2P lending, traditional models fall short and thus this study aimed to rectify this as well as explore the problem of class imbalances seen within credit risk data sets. This study implemented an over-sampling technique known as Synthetic Minority Over-sampling Technique (SMOTE). To test our approach, we implemented five benchmarking classifiers such as support vector machines, logistic regression, k-nearest neighbor, random forest, and deep neural network. The data sample used was retrieved from the publicly available LendingClub dataset. The proposed SMOTE revealed significantly improved results in comparison with the benchmarking classifiers. These results should help actors engaged within P2P lending to make better informed decisions when selecting potential borrowers eliminating the higher risks present in P2P lending.

• Journal of the Korean Society for information Management
• /
• v.9 no.2
• /
• pp.118-153
• /
• 1992

This study att.empts to conduct a system an;~lysis and dcsign for circt~lation control in a univcr~ity library. In this proccss, structured sysltcm analys15 tcrhn~clucs Iikr data flolv diagrnms. data dictiomiry, and entity - relationship analysis, arc. vmploycd to construct both concqitual and p h p c a l data models of current and new circulation systems. Thc whole design aims at an al~tomntctf c~r.culation system on thc [nlcro-compurcr --basis in a mid dle - sized library.

• Economic Analysis
• /
• v.25 no.4
• /
• pp.1-24
• /
• 2019

This study examines the labor market performance of graduates who had student loans. Compared to earlier studies, we extended analyses to all jobs that were experienced for more than 18 months after graduation. First, we found that students who had student loans earned 2.81% less at their first job compared to their counterparts without student loans. Second, the wage gap decreased over time, a reduction of 0.66%p due to labor market turnovers. Third, when we compared cumulated labor income, however, the amount for borrowers were continuously higher. This is because the job searching period of a borrower was shorter, despite relatively lower wages at the first job, and borrowers also made more frequent job turnovers, accompanying relatively more wage increases. These results suggest that the negative effects of college loans on earnings, reported in previous studies, may have exaggerated the negative impact to some extent of having loans. However, when we look at the quality of jobs beyond simply wages, the proportion of borrowers working at large companies as regular workers was consistently low. Given that job conditions at the earlier stages of one's career may lead to gaps over time, our findings call for more systematic investigations into the effects that student loans have on long-term labor performance.

• 한국벤처창업학회:학술대회논문집
• /
• 2017.04a
• /
• pp.52-52
• /
• 2017

글로벌 금융위기 이후 다양한 형태로 등장한 금융상품과 ICT의 결합은 그 동안 생각하지 못한 방식으로 전 세계에 다양한 수요를 충족시키면서 폭발적으로 성장했다. 하지만 IT강국이라고 자부하는 대한민국은 다양한 규제와 시스템의 복잡성 때문에 은행상품이 온라인에서 거래되는 것은 아직까지 익숙하지 않다. 다행히 이러한 규제가 조금씩 완화되어 가면서 2016년은 모바일 송금, 금융상품 추천 플랫폼 등 비 금융업체 주도의 금융시장 온라인화가 소극적으로 이루어지는 과도기로 볼 수 있다. 이러한 시점에서 기존 오프라인 채널이 아닌 온라인 채널을 통해 금융상품을 구매하거나 가입하는 고객의 만족요인에 대해 연구하는 것은 향후 폭발적으로 증가할 수요에 앞서 연구하고, 현상을 주도할 기업에서도 소비자의 만족요인을 미리 파악한다는 점에서 시기적으로 적절하다. 해당 연구는 신용대출, 정기예금, 전세대출, 주택담보대출, 정기적금, 그리고 P2P투자 상품 별 만족도에 영향을 미치는 요인과 영향력을 SERVPERF 모델을 이용하여 분석한 뒤, 회귀분석과 텍스트간의 공동 출현단어에 대해 파이선을 통해 메트릭스를 형성하고, 사회연결망 분석으로 네트워크 중심성을 분석하여 단어간의 관계를 살펴보았다. 해당 연구는 국내 최초 온라인 금융상품 비교 추천 플랫폼인 "Finda"의 리뷰/평점데이터를 이용하였다.

• Industry Promotion Research
• /
• v.7 no.4
• /
• pp.109-119
• /
• 2022

In recent years, China's Internet finance industry is hot. There is no doubt that Internet finance has been fully integrated into China, forming a new form of financing, and rapidly becoming a new channel for investment and financing in China, shouldering the responsibility of inclusive financing and building China's real economy. However, with investment, there are risks. Based on the panel data of China's Internet financial platform, this paper uses the random effect model to study the influencing factors of Internet financial risks, and draws three conclusions: (1) The user funds and platform funds of the financial platform will be managed separately by the bank, which can effectively reduce the risk of financial transactions on the Internet; (2) The risk of Internet financial transactions can be effectively reduced by avoiding the concentration of platform funds in the hands of a few borrowers through regulatory policies; (3) The liquidity control of funds effectively reduces the risk of Internet financial transactions. Based on the conclusions, we propose optimization strategies for regulatory policies to achieve the healthy and sustainable development of Internet finance.