Credit Risk Evaluation Based on Text Analysis

Credit Risk Evaluation Based on Text Analysis

Shuxia Wang (Beijing Institute of Petrochemical Technology, Beijing, China), Yuwei Qi (Peking University, Beijing, China), Bin Fu (Peking University, Beijing, China) and Hongzhi Liu (Peking University, Beijing, China)
Copyright: © 2016 |Pages: 11
DOI: 10.4018/IJCINI.2016010101

Abstract

The main difficulty of credit risk evaluation is to evaluate borrowers' willingness of repayment, which is a subjective factor depending on the thoughts and ideas of borrowers. Text description is a kind of human behavior which reflects the mental process of writers. The authors identify the characteristics of borrowers from their text descriptions and further use them to evaluate the credit risk of loans. Experimental results show that: (1) textual information is a good choice when traditional financial information is missing. The authors can achieve similar accuracy using only textual information as traditional methods which use financial information and credit information from the third party. (2) Textual information is a good complementary information source to traditional financial information sources. Using textual information can improve the performance of credit risk evaluation system when combined with traditional financial information.
Article Preview

Traditional credit risk evaluation methods focus on the financial information of borrowers. Emekter et al. (2015) studied the relationship between various financial factors and the default rate. Their results showed that credit grade, debt-to-income ratio, FICO score and revolving line utilization play an important role in loan defaults. Loans with lower credit grade and longer duration are associated with high default rate. Puro et al. (2010) studied the relationship between loan amount, interest rate and the funding success. Their experimental results showed that lower interest rates decrease the chance of getting the loan funded, while smaller loan amounts increase the chance of funded.

Potzsch and Bohme (2010) found that soft information can effectively alleviate the problem of missing hard information like financial information and improve the accuracy of credit risk evaluation. Social information is one kind of soft information. Recently, researchers begin to study the relationship between the online social information and credit risk. Brandes et al. (2011) studied the impact of social information on the interest rate of funded loans and found that lenders use social information to trust some borrowers more than it is suggested by their credit grades. Greiner and Wang (2009) studied the role of social capital in online P2P lending market and found that borrowers with more social capital tend to get loans funded and obtain lower interest rates. Herrero-Lopez (2009) studied the influence of social interaction in online P2P lending. Their results showed that fostering social features, such as joining in a trusted group, could increase the chance of getting a loan funded. Lin et al. (2013) studied the relation between the online friendships and transactional outcomes in the online P2P market and found that online friendships of borrowers act as signals of credit quality. Borrowers with friends are more likely to get their loans funded with lower interest rates.

Purchase this article to continue reading all 11 pages >

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 13: 4 Issues (2019): Forthcoming, Available for Pre-Order
Volume 12: 4 Issues (2018): 3 Released, 1 Forthcoming
Volume 11: 4 Issues (2017)
Volume 10: 4 Issues (2016)
Volume 9: 4 Issues (2015)
Volume 8: 4 Issues (2014)
Volume 7: 4 Issues (2013)
Volume 6: 4 Issues (2012)
Volume 5: 4 Issues (2011)
Volume 4: 4 Issues (2010)
Volume 3: 4 Issues (2009)
Volume 2: 4 Issues (2008)
Volume 1: 4 Issues (2007)
View Complete Journal Contents Listing