快速發展的金融科技和共享經濟一直是產業革命的主力軍，給社會帶來了許多創新與變革。點對點的網路借貸（P2P lending）是一種新興的金融商業模式，它是金融科技和共享經濟的結合。與其他新興金融科技服務一樣，P2P網路貸款帶來了不少好處，也同樣產生了問題。例如，它能夠把更多以前不具備使用金融借貸服務條件的人包含在服務中，同時我們也觀察到P2P網路借貸公司和借款人之間產生的利益衝突。通過研究Lending Club數據及評估FICO信用評分，本研究分析了P2P網路借貸平台中傳統的信用評分（如FICO）在大數據中的價值，以及傳統信用評分對機器學習模型產生的影響。我們的方法是根據使用現有數據，建構機器學習預測模型，預測和計算的單個貸款個案的利潤值。然後我們將預測模型的預計利潤值與FICO評分在模型建構中的存在與否進行比較。我們的研究是基於Lending Club2012年至2015年的數據集，顯示FICO評分的價值在129美元至685美元範圍之間。

The rapid development of financial technology and the sharing economy has been the main force of industry revolutionary and making many changes to our society. Peer-to-Peer (P2P) lending is a new business model that is a combination of financial technology and the sharing economy. Like other new fintech services, P2P lending brings both benefits and issues. For example, P2P lending enhances financial inclusion by allowing more people access to financial services and resources, while we see conflicts of interest occurring between P2P lending companies and lenders. By studying Lending Club dataset and evaluating the FICO scores, this research analyzed the value of conventional credit scores (e.g. FICO) in the world of big data in the P2P lending platforms and how conventional credit scores effect on machine learning models. Our approach is to predict the projected individual loan value based on machine learning predictive models built with existing data; then we compared the projected profit values of the models predicted with the presence and absence of FICO scores. Our research, based on the Lending Club datasets (2012 – 2015), shows that the range of the value of FICO scores is between $129 to $685 USD.

Table of Contents
1 INTRODUCTION 3
1.1 RESEARCH BACKGROUND 3
1.2 RESEARCH MOTIVATION 4
1.3 RESEARCH OBJECTIVE 5
2 LITERATURE REVIEW 6
2.1 FINTECH 6
2.2 PEER-TO-PEER LENDING 6
2.3 CREDIT RISK 8
2.4 CREDIT SCORING 8
2.5 FICO SCORE 9
2.6 LAWS AND REGULATIONS OF FINANCIAL CREDIT SCORE SYSTEM 11
2.6.1 UNITED STATES - THE FAIR CREDIT REPORTING ACT 11
2.6.2 EUROPEAN UNION - THE GENERAL DATA PROTECTION REGULATION 12
2.6.3 CHINA 13
2.6.4 TAIWAN 13
3 DATA AND METHODOLOGY OVERVIEW 15
3.1 BASIC INFORMATION OF THE LENDING CLUB DATA SETS 15
3.2 BASIC METHODOLOGY REVIEW 18
3.2.1 FEATURE SELECTION IN MACHINE LEARNING 18
3.2.2 CHOICE OF ALGORITHMS 19
3.2.3 RANDOM FOREST 19
3.2.4 LOGISTIC REGRESSION WITH L1 PENALTY (LASSO LOGISTIC) 20
4 DATA DESCRIPTION 22
4.1 VARIABLES OVERVIEW 22
4.2 DESCRIPTIVE STATISTICS OF THE DATA SETS 23
5 DATA ANALYSIS 31
5.1 DATA PREPROCESSING PROCEDURE 31
5.2 MODEL PERFORMANCE EVALUATION CRITERIA 35
5.3 MODELING DESCRIPTION 37
6 RESULT 40
7 CONCLUSION AND LIMITATION 49
REFERENCE 51

Reference

Baesens, B. (2014). Analytics in a big data world: The essential guide to data science and its applications. John Wiley & Sons.
Chandrashekar, G., & Sahin, F. (2014). A survey on feature selection methods. Computers & Electrical Engineering, 40(1), 16-28.
Churpek, M. M., Yuen, T. C., Winslow, C., Meltzer, D. O., Kattan, M. W., & Edelson, D. P. (2016). Multicenter comparison of machine learning methods and conventional regression for predicting clinical deterioration on the wards. Critical care medicine, 44(2), 368.
Gomber, P., Kauffman, R. J., Parker, C., & Weber, B. W. (2018). On the fintech revolution: interpreting the forces of innovation, disruption, and transformation in financial services. Journal of Management Information Systems, 35(1), 220-265.
Han, J., Kamber, M., & Pei, J. (2012). Data mining: Concepts and techniques. Amsterdam: Elsevier/Morgan Kaufmann.
J. Li, S. Hsu, Z. Chen, and Y. Chen, “Risks of p2p lending platforms in china: Modeling failure using a cox hazard model,” The Chinese Economy, vol. 49, no. 3, pp. 161–172, 2016.
Lee, E., & Lee, B. (2012). Herding behavior in online P2P lending: An empirical investigation. Electronic Commerce Research and Applications, 11(5), 495-503.
Malekipirbazari, M., & Aksakalli, V. (2015). Risk assessment in social lending via random forests. Expert Systems with Applications, 42(10), 4621-4631.
Mezei, J., Byanjankar, A., & Heikkilä, M. (2018). Credit risk evaluation in peer-to-peer lending with linguistic data transformation and supervised learning.
Riza Emekter, Yanbin Tu, Benjamas Jirasakuldech & Min Lu (2015) Evaluating credit risk and loan performance in online Peer-to-Peer (P2P) lending, Applied Economics, 47:1, 54-70, DOI: 10.1080/00036846.2014.962222
Rodriguez-Galiano, V., Luque-Espinar, J., Chica-Olmo, M., & Mendes, M. (2018). Feature selection approaches for predictive modelling of groundwater nitrate pollution: An evaluation of filters, embedded and wrapper methods. Science of The Total Environment, 624, 661-672. doi:10.1016/j.scitotenv.2017.12.152
Rubini, Agustin. Fintech in a Flash: Financial Technology Made Easy (2018 edition) (p. 23). Banking Innovations.
Shalev-Shwartz, S., & Ben-David, S. (2014). Understanding machine learning: From theory to algorithms. Cambridge: Cambridge University Press.
Shmueli, G., Bruce, P. C., Yahav, I., Patel, N. R., & Lichtendahl, K. C. (2018). Data mining for business analytics: Concepts, techniques, and applications in R. Hoboken, NJ: John Wiley & Sons.
Siddiqi, N. (2012). Credit risk scorecards: developing and implementing intelligent credit scoring (Vol. 3). John Wiley & Sons.
Song, P., Chen, Y., Zhou, Z., & Wu, H. (2018). Performance Analysis of Peer-to-Peer Online Lending Platforms in China. Sustainability, 10(9), 2987.
Wang H. and Greiner M. E. (2011), “Prosper: the ebay for money in lending 2.0,” Communications of the Asso- ciation for Information Systems, vol. 29, no. 1, p. 13,
Wei, S. (2015). Internet lending in China: Status quo, potential risks and regulatory options. Computer Law & Security Review, 31(6), 793-809.
Ye, X., Dong, L. A., & Ma, D. (2018). Loan evaluation in P2P lending based on Random Forest optimized by genetic algorithm with profit score. Electronic Commerce Research and Applications, 32, 23-36
Yüzbaşı, B., & Arashi, M. (2019). Double shrunken selection operator. Communications in Statistics-Simulation and Computation, 48(3), 666-674.