
In this study, we constructed the credit-scoring model of P2P loans by using several machine learning and artificial neural network (ANN) methods, including logistic regression (LR), a support vector machine, a decision tree, random forest, XGBoost, LightGBM and 2-layer neural networks. This study explores several hyperparameter settings for each method by performing a grid search and cross-validation to get the most suitable credit-scoring model in terms of training time and test performance. In this study, we get and clean the open P2P loan data from Lending Club with feature engineering concepts. In order to find significant default factors, we used an XGBoost method to pre-train all data and get the feature importance. The 16 selected features can provide economic implications for research about default prediction in P2P loans. Besides, the empirical result shows that gradient-boosting decision tree methods, including XGBoost and LightGBM, outperform ANN and LR methods, which are commonly used for traditional credit scoring. Among all of the methods, XGBoost performed the best.
Citation: An-Hsing Chang, Li-Kai Yang, Rua-Huan Tsaih, Shih-Kuei Lin. Machine learning and artificial neural networks to construct P2P lending credit-scoring model: A case using Lending Club data[J]. Quantitative Finance and Economics, 2022, 6(2): 303-325. doi: 10.3934/QFE.2022013
[1] | Honglei Wang, Wenliang Zeng, Xiaoling Huang, Zhaoyang Liu, Yanjing Sun, Lin Zhang . MTTLm6A: A multi-task transfer learning approach for base-resolution mRNA m6A site prediction based on an improved transformer. Mathematical Biosciences and Engineering, 2024, 21(1): 272-299. doi: 10.3934/mbe.2024013 |
[2] | Shanzheng Wang, Xinhui Xie, Chao Li, Jun Jia, Changhong Chen . Integrative network analysis of N6 methylation-related genes reveal potential therapeutic targets for spinal cord injury. Mathematical Biosciences and Engineering, 2021, 18(6): 8174-8187. doi: 10.3934/mbe.2021405 |
[3] | Pingping Sun, Yongbing Chen, Bo Liu, Yanxin Gao, Ye Han, Fei He, Jinchao Ji . DeepMRMP: A new predictor for multiple types of RNA modification sites using deep learning. Mathematical Biosciences and Engineering, 2019, 16(6): 6231-6241. doi: 10.3934/mbe.2019310 |
[4] | Yong Zhu, Zhipeng Jiang, Xiaohui Mo, Bo Zhang, Abdullah Al-Dhelaan, Fahad Al-Dhelaan . A study on the design methodology of TAC3 for edge computing. Mathematical Biosciences and Engineering, 2020, 17(5): 4406-4421. doi: 10.3934/mbe.2020243 |
[5] | Atefeh Afsar, Filipe Martins, Bruno M. P. M. Oliveira, Alberto A. Pinto . A fit of CD4+ T cell immune response to an infection by lymphocytic choriomeningitis virus. Mathematical Biosciences and Engineering, 2019, 16(6): 7009-7021. doi: 10.3934/mbe.2019352 |
[6] | Tamás Tekeli, Attila Dénes, Gergely Röst . Adaptive group testing in a compartmental model of COVID-19*. Mathematical Biosciences and Engineering, 2022, 19(11): 11018-11033. doi: 10.3934/mbe.2022513 |
[7] | Wenli Cheng, Jiajia Jiao . An adversarially consensus model of augmented unlabeled data for cardiac image segmentation (CAU+). Mathematical Biosciences and Engineering, 2023, 20(8): 13521-13541. doi: 10.3934/mbe.2023603 |
[8] | Tongmeng Jiang, Pan Jin, Guoxiu Huang, Shi-Cheng Li . The function of guanylate binding protein 3 (GBP3) in human cancers by pan-cancer bioinformatics. Mathematical Biosciences and Engineering, 2023, 20(5): 9511-9529. doi: 10.3934/mbe.2023418 |
[9] | Xin Yu, Jun Liu, Ruiwen Xie, Mengling Chang, Bichun Xu, Yangqing Zhu, Yuancai Xie, Shengli Yang . Construction of a prognostic model for lung squamous cell carcinoma based on seven N6-methylandenosine-related autophagy genes. Mathematical Biosciences and Engineering, 2021, 18(5): 6709-6723. doi: 10.3934/mbe.2021333 |
[10] | Tahir Rasheed, Faran Nabeel, Muhammad Bilal, Yuping Zhao, Muhammad Adeel, Hafiz. M. N. Iqbal . Aqueous monitoring of toxic mercury through a rhodamine-based fluorescent sensor. Mathematical Biosciences and Engineering, 2019, 16(4): 1861-1873. doi: 10.3934/mbe.2019090 |
In this study, we constructed the credit-scoring model of P2P loans by using several machine learning and artificial neural network (ANN) methods, including logistic regression (LR), a support vector machine, a decision tree, random forest, XGBoost, LightGBM and 2-layer neural networks. This study explores several hyperparameter settings for each method by performing a grid search and cross-validation to get the most suitable credit-scoring model in terms of training time and test performance. In this study, we get and clean the open P2P loan data from Lending Club with feature engineering concepts. In order to find significant default factors, we used an XGBoost method to pre-train all data and get the feature importance. The 16 selected features can provide economic implications for research about default prediction in P2P loans. Besides, the empirical result shows that gradient-boosting decision tree methods, including XGBoost and LightGBM, outperform ANN and LR methods, which are commonly used for traditional credit scoring. Among all of the methods, XGBoost performed the best.
The constituent members in a system mainly found in nature can be interacting with each other through cooperation and competition. Demonstrations for such systems involve biological species, countries, businesses, and many more. It's very much intriguing to investigate in a comprehensive manner numerous social as well as biological interactions existent in dissimilar species/entities utilizing mathematical modeling. The predation and the competition species are the most famous interactions among all such types of interactions. Importantly, Lotka [1] and Volterra [2] in the 1920s have announced individually the classic equations portraying population dynamics. Such illustrious equations are notably described as predator-prey (PP) equations or Lotka-Volterra (LV) equations. In this structure, PP/LV model represents the most influential model for interacting populations. The interplay between prey and predator together with additional factors has been a prominent topic in mathematical ecology for a long period. Arneodo et al. [3] have established in 1980 that a generalized Lotka-Volterra biological system (GLVBS) would depict chaos phenomena in an ecosystem for some explicitly selected system parameters and initial conditions. Additionally, Samardzija and Greller [4] demonstrated in 1988 that GLVBS would procure chaotic reign from the stabled state via rising fractal torus. LV model was initially developed as a biological concept, yet it is utilized in enormous diversified branches for research [5,6,7,8]. Synchronization essentially is a methodology of having different chaotic systems (non-identical or identical) following exactly a similar trajectory, i.e., the dynamical attributes of the slave system are locked finally into the master system. Specifically, synchronization and control have a wide spectrum for applications in engineering and science, namely, secure communication [9], encryption [10,11], ecological model [12], robotics [13], neural network [14], etc. Recently, numerous types of secure communication approaches have been explored [15,16,17,18] such as chaos modulation [18,19,20,21], chaos shift keying [22,23] and chaos masking [9,17,20,24]. In chaos communication schemes, the typical key idea for transmitting a message through chaotic/hyperchaotic models is that a message signal is nested in the transmitter system/model which originates a chaotic/ disturbed signal. Afterwards, this disturbed signal has been emitted to the receiver through a universal channel. The message signal would finally be recovered by the receiver. A chaotic model has been intrinsically employed both as receiver and transmitter. Consequently, this area of chaotic synchronization & control has sought remarkable considerations among differential research fields.
Most prominently, synchronization theory has been in existence for over 30 years due to the phenomenal research of Pecora and Carroll [25] established in 1990 using drive-response/master-slave/leader-follower configuration. Consequently, many authors and researchers have started introducing and studying numerous control and synchronization methods [9,26,27,28,29,30,31,32,33,34,35,36] etc. to achieve stabilized chaotic systems for possessing stability. In [37], researchers discussed optimal synchronization issues in similar GLVBSs via optimal control methodology. In [38,39], the researchers studied the adaptive control method (ACM) to synchronize chaotic GLVBSs. Also, researchers [40] introduced a combination difference anti-synchronization scheme in similar chaotic GLVBSs via ACM. In addition, authors [41] investigated a combination synchronization scheme to control chaos existing in GLVBSs using active control strategy (ACS). Bai and Lonngren [42] first proposed ACS in 1997 for synchronizing and controlling chaos found in nonlinear dynamical systems. Furthermore, compound synchronization using ACS was first advocated by Sun et al. [43] in 2013. In [44], authors discussed compound difference anti-synchronization scheme in four chaotic systems out of which two chaotic systems are considered as GLVBSs using ACS and ACM along with applications in secure communications of chaos masking type in 2019. Some further research works [45,46] based on ACS have been reported in this direction. The considered chaotic GLVBS offers a generalization that allows higher-order biological terms. As a result, it may be of interest in cases where biological systems experience cataclysmic changes. Unfortunately, some species will be under competitive pressure in the coming years and decades. This work may be comprised as a step toward preserving as many currently living species as possible by using the proposed synchronization approach which is based on master-slave configuration and Lyapunov stability analysis.
In consideration of the aforementioned discussions and observations, our primary focus here is to develop a systematic approach for investigating compound difference anti-synchronization (CDAS) approach in 4 similar chaotic GLVBSs via ACS. The considered ACS is a very efficient yet theoretically rigorous approach for controlling chaos found in GLVBSs. Additionally, in view of widely known Lyapunov stability analysis (LSA) [47], we discuss actively designed biological control law & convergence for synchronization errors to attain CDAS synchronized states.
The major attributes for our proposed research in the present manuscript are:
● The proposed CDAS methodology considers four chaotic GLVBSs.
● It outlines a robust CDAS approach based active controller to achieve compound difference anti-synchronization in discussed GLVBSs & conducts oscillation in synchronization errors along with extremely fast convergence.
● The construction of the active control inputs has been executed in a much simplified fashion utilizing LSA & master-salve/ drive-response configuration.
● The proposed CDAS approach in four identical chaotic GLVBSs of integer order utilizing ACS has not yet been analyzed up to now. This depicts the novelty of our proposed research work.
This manuscript is outlined as follows: Section 2 presents the problem formulation of the CDAS scheme. Section 3 designs comprehensively the CDAS scheme using ACS. Section 4 consists of a few structural characteristics of considered GLVBS on which CDAS is investigated. Furthermore, the proper active controllers having nonlinear terms are designed to achieve the proposed CDAS strategy. Moreover, in view of Lyapunov's stability analysis (LSA), we have examined comprehensively the biological controlling laws for achieving global asymptotical stability of the error dynamics for the discussed model. In Section 5, numerical simulations through MATLAB are performed for the illustration of the efficacy and superiority of the given scheme. Lastly, we also have presented some conclusions and the future prospects of the discussed research work in Section 6.
We here formulate a methodology to examine compound difference anti-synchronization (CDAS) scheme viewing master-slave framework in four chaotic systems which would be utilized in the coming up sections.
Let the scaling master system be
˙wm1= f1(wm1), | (2.1) |
and the base second master systems be
˙wm2= f2(wm2), | (2.2) |
˙wm3= f3(wm3). | (2.3) |
Corresponding to the aforementioned master systems, let the slave system be
˙ws4= f4(ws4)+U(wm1,wm2,wm3,ws4), | (2.4) |
where wm1=(wm11,wm12,...,wm1n)T∈Rn, wm2=(wm21,wm22,...,wm2n)T∈Rn, wm3=(wm31,wm32,...,wm3n)T∈Rn, ws4=(ws41,ws42,...,ws4n)T∈Rn are the state variables of the respective chaotic systems (2.1)–(2.4), f1,f2,f3,f4:Rn→Rn are four continuous vector functions, U=(U1,U2,...,Un)T:Rn×Rn×Rn×Rn→Rn are appropriately constructed active controllers.
Compound difference anti-synchronization error (CDAS) is defined as
E=Sws4+Pwm1(Rwm3−Qwm2), |
where P=diag(p1,p2,.....,pn),Q=diag(q1,q2,.....,qn),R=diag(r1,r2,.....,rn),S=diag(s1,s2,.....,sn) and S≠0.
Definition: The master chaotic systems (2.1)–(2.3) are said to achieve CDAS with slave chaotic system (2.4) if
limt→∞‖E(t)‖=limt→∞‖Sws4(t)+Pwm1(t)(Rwm3(t)−Qwm2(t))‖=0. |
We now present our proposed CDAS approach in three master systems (2.1)–(2.3) and one slave system (2.4). We next construct the controllers based on CDAS approach by
Ui= ηisi−(f4)i−KiEisi, | (3.1) |
where ηi=pi(f1)i(riwm3i−qiwm2i)+piwm1i(ri(f3)i−qi(f2)i), for i=1,2,...,n.
Theorem: The systems (2.1)–(2.4) will attain the investigated CDAS approach globally and asymptotically if the active control functions are constructed in accordance with (3.1).
Proof. Considering the error as
Ei= siws4i+piwm1i(riwm3i−qiwm2i),fori=1,2,3,.....,n. |
Error dynamical system takes the form
˙Ei= si˙ws4i+pi˙wm1i(riwm3i−qiwm2i)+piwm1i(ri˙wm3i−qi˙wm2i)= si((f4)i+Ui)+pi(f1)i(riwm3i−qiwm2i)+piwm1i(ri(f3)i−qi(f2)i)= si((f4)i+Ui)+ηi, |
where ηi=pi(f1)i(riwm3i−qiwm2i)+piwm1i(ri(f3)i−qi(f2)i), i=1,2,3,....,n. This implies that
˙Ei= si((f4)i−ηisi−(f4)i−KiEisi)+ηi= −KiEi | (3.2) |
The classic Lyapunov function V(E(t)) is described by
V(E(t))= 12ETE= 12ΣE2i |
Differentiation of V(E(t)) gives
˙V(E(t))=ΣEi˙Ei |
Using Eq (3.2), one finds that
˙V(E(t))=ΣEi(−KiEi)= −ΣKiE2i). | (3.3) |
An appropriate selection of (K1,K1,.......,Kn) makes ˙V(E(t)) of eq (3.3), a negative definite. Consequently, by LSA [47], we obtain
limt→∞Ei(t)=0,(i=1,2,3). |
Hence, the master systems (2.1)–(2.3) and slave system (2.4) have attained desired CDAS strategy.
We now describe GLVBS as the scaling master system:
{˙wm11=wm11−wm11wm12+b3w2m11−b1w2m11wm13,˙wm12=−wm12+wm11wm12,˙wm13=b2wm13+b1w2m11wm13, | (4.1) |
where (wm11,wm12,wm13)T∈R3 is state vector of (4.1). Also, wm11 represents the prey population and wm12, wm13 denote the predator populations. For parameters b1=2.9851, b2=3, b3=2 and initial conditions (27.5,23.1,11.4), scaling master GLVBS displays chaotic/disturbed behaviour as depicted in Figure 1(a).
The base master systems are the identical chaotic GLVBSs prescribed respectively as:
{˙wm21=wm21−wm21wm22+b3w2m21−b1w2m21wm23,˙wm22=−wm22+wm21wm22,˙wm23=b2wm23+b1w2m21wm23, | (4.2) |
where (wm21,wm22,wm23)T∈R3 is state vector of (4.2). For parameter values b1=2.9851, b2=3, b3=2, this base master GLVBS shows chaotic/disturbed behaviour for initial conditions (1.2,1.2,1.2) as displayed in Figure 1(b).
{˙wm31=wm31−wm31wm32+b3w2m31−b1w2m31wm33,˙wm32=−wm32+wm31wm32,˙wm33=b2wm33+b1w2m31wm33, | (4.3) |
where (wm31,wm32,wm33)T∈R3 is state vector of (4.3). For parameters b1=2.9851, b2=3, b3=2, this second base master GLVBS displays chaotic/disturbed behaviour for initial conditions (2.9,12.8,20.3) as shown in Figure 1(c).
The slave system, represented by similar GLVBS, is presented by
{˙ws41=ws41−ws41ws42+b3w2s41−b1w2s41ws43+U1,˙ws42=−ws42+ws41ws42+U2,˙ws43=b2ws43+b1w2s41ws43+U3, | (4.4) |
where (ws41,ws42,ws43)T∈R3 is state vector of (4.4). For parameter values, b1=2.9851, b2=3, b3=2 and initial conditions (5.1,7.4,20.8), the slave GLVBS exhibits chaotic/disturbed behaviour as mentioned in Figure 1(d).
Moreover, the detailed theoretical study for (4.1)–(4.4) can be found in [4]. Further, U1, U2 and U3 are controllers to be determined.
Next, the CDAS technique has been discussed for synchronizing the states of chaotic GLVBS. Also, LSA-based ACS is explored & the necessary stability criterion is established.
Here, we assume P=diag(p1,p2,p3), Q=diag(q1,q2,q3), R=diag(r1,r2,r3), S=diag(s1,s2,s3). The scaling factors pi,qi,ri,si for i=1,2,3 are selected as required and can assume the same or different values.
The error functions (E1,E2,E3) are defined as:
{E1=s1ws41+p1wm11(r1wm31−q1wm21),E2=s2ws42+p2wm12(r2wm32−q2wm22),E3=s3ws43+p3wm13(r3wm33−q3wm23). | (4.5) |
The major objective of the given work is the designing of active control functions Ui,(i=1,2,3) ensuring that the error functions represented in (4.5) must satisfy
limt→∞Ei(t)=0for(i=1,2,3). |
Therefore, subsequent error dynamics become
{˙E1=s1˙ws41+p1˙wm11(r1wm31−q1wm21)+p1wm11(r1˙wm31−q1˙wm21),˙E2=s2˙ws42+p2˙wm12(r2wm32−q2wm22)+p2wm12(r2˙wm32−q2˙wm22),˙E3=s3˙ws43+p3˙wm13(r3wm33−q3wm23)+p3wm13(r3˙wm33−q3˙wm23). | (4.6) |
Using (4.1), (4.2), (4.3), and (4.5) in (4.6), the error dynamics simplifies to
{˙E1=s1(ws41−ws41ws42+b3w2s41−b1w2s41ws43+U1)+p1(wm11−wm11wm12+b3w2m11−b1w2m11wm13)(r1wm31−q1wm21)+p1wm11(r1(wm31−wm31wm32+b3w2m31−b1w2m31wm33)−q1(wm21−wm21wm22+b3w2m21−b1w2m21wm23),˙E2=s2(−ws42+ws41ws42+U2)+p2(−wm12+wm11wm12)(r2wm32−q2wm22)+p2wm12(r2(−wm32+wm31wm32)−q2(−wm22+wm21wm22)),˙E3=s3(b2ws43+b1w2s41ws43+U3)+p3(b2wm13+b1w2m11wm13)(r3wm33−q3wm23)+p3wm13(r3(b2wm33+b1w2m31wm33)−q3(b2wm23+b1w2m21wm23)). | (4.7) |
Let us now choose the active controllers:
U1= η1s1−(f4)1−K1E1s1, | (4.8) |
where η1=p1(f1)1(r1wm31−q1wm21)+p1wm11(r1(f3)1−q1(f2)1), as described in (3.1).
U2= η2s2−(f4)2−K2E2s2, | (4.9) |
where η2=p2(f1)2(r2wm32−q2wm22)+p2wm12(r2(f3)2−q2(f2)2).
U3= η3s3−(f4)3−K3E3s3, | (4.10) |
where η3=p3(f1)3(r3wm33−q3wm23)+p3wm13(r3(f3)3−q3(f2)3) and K1>0,K2>0,K3>0 are gaining constants.
By substituting the controllers (4.8), (4.9) and (4.10) in (4.7), we obtain
{˙E1=−K1E1,˙E2=−K2E2,˙E3=−K3E3. | (4.11) |
Lyapunov function V(E(t)) is now described by
V(E(t))= 12[E21+E22+E23]. | (4.12) |
Obviously, the Lyapunov function V(E(t)) is +ve definite in R3. Therefore, the derivative of V(E(t)) as given in (4.12) can be formulated as:
˙V(E(t))= E1˙E1+E2˙E2+E3˙E3. | (4.13) |
Using (4.11) in (4.13), one finds that
˙V(E(t))= −K1E21−K2E22−K3E23<0, |
which displays that ˙V(E(t)) is -ve definite.
In view of LSA [47], we, therefore, understand that CDAS error dynamics is globally as well as asymptotically stable, i.e., CDAS error E(t)→0 asymptotically for t→∞ to each initial value E(0)∈R3.
This section conducts a few simulation results for illustrating the efficacy of the investigated CDAS scheme in identical chaotic GLVBSs using ACS. We use 4th order Runge-Kutta algorithm for solving the considered ordinary differential equations. Initial conditions for three master systems (4.1)–(4.3) and slave system (4.4) are (27.5,23.1,11.4), (1.2,1.2,1.2), (2.9,12.8,20.3) and (14.5,3.4,10.1) respectively. We attain the CDAS technique among three masters (4.1)–(4.3) and corresponding one slave system (4.4) by taking pi=qi=ri=si=1, which implies that the slave system would be entirely anti-synchronized with the compound of three master models for i=1,2,3. In addition, the control gains (K1,K2,K3) are taken as 2. Also, Figure 2(a)–(c) indicates the CDAS synchronized trajectories of three master (4.1)–(4.3) & one slave system (4.4) respectively. Moreover, synchronization error functions (E1,E2,E3)=(51.85,275.36,238.54) approach 0 as t tends to infinity which is exhibited via Figure 2(d). Hence, the proposed CDAS strategy in three masters and one slave models/systems has been demonstrated computationally.
In this work, the investigated CDAS approach in similar four chaotic GLVBSs using ACS has been analyzed. Lyapunov's stability analysis has been used to construct proper active nonlinear controllers. The considered error system, on the evolution of time, converges to zero globally & asymptotically via our appropriately designed simple active controllers. Additionally, numerical simulations via MATLAB suggest that the newly described nonlinear control functions are immensely efficient in synchronizing the chaotic regime found in GLVBSs to fitting set points which exhibit the efficacy and supremacy of our proposed CDAS strategy. Exceptionally, both analytic theory and computational results are in complete agreement. Our proposed approach is simple yet analytically precise. The control and synchronization among the complex GLVBSs with the complex dynamical network would be an open research problem. Also, in this direction, we may extend the considered CDAS technique on chaotic systems that interfered with model uncertainties as well as external disturbances.
The authors gratefully acknowledge Qassim University, represented by the Deanship of Scientific Research, on the financial support for this research under the number 10163-qec-2020-1-3-I during the academic year 1441 AH/2020 AD.
The authors declare there is no conflict of interest.
[1] | Aldrich JH, Nelson FD (1984) Quantitative Applications in the Social Sciences: Linear Probability, Logit, and Probit Models, Thousand Oaks, CA: SAGE Publications. https://doi.org/10.4135/9781412984744 |
[2] | Alexander VE, Clifford CC (1996) Categorical Variables in Developmental Research: Methods of Analysis, Elsevier. https://doi.org/10.1016/B978-012724965-0/50003-1 |
[3] |
Arya S, Eckel C, Wichman C (2013) Anatomy of the Credit Score. J Econ Behav Organ 95: 175–185. https://doi.org/10.1016/j.jebo.2011.05.005 doi: 10.1016/j.jebo.2011.05.005
![]() |
[4] |
Baesens B, Van Gestel T, Viaene S, et al. (2003) Benchmarking state-of-the-art classification algorithms for credit scoring. J Oper Res Soc 54: 627–635. https://doi.org/10.1057/palgrave.jors.2601545 doi: 10.1057/palgrave.jors.2601545
![]() |
[5] |
Baesens B, Gestel TV, Stepanova M, et al. (2004) Neural Network Survival Analysis for Personal Loan Data. J Oper Res Soc 56: 1089–1098. https://doi.org/10.1057/palgrave.jors.2601990 doi: 10.1057/palgrave.jors.2601990
![]() |
[6] | Bishop CM (2006) Pattern Recognition and Machine Learning, Springer. https://doi.org/10.1007/978-0-387-45528-0_5 |
[7] | Bolton C (2010) Logistic Regression and its Application in Credit Scoring, University of Pretoria. Available from: http://hdl.handle.net/2263/27333. |
[8] |
Breiman L (1996) Bagging Predictors. Mach Learn 24: 123–140. https://doi.org/10.1007/BF00058655 doi: 10.1007/BF00058655
![]() |
[9] | Breiman L, Friedman J, Stone CJ, et al. (1984) Classification and Regression Trees, Taylor & Francis. https://doi.org/10.1201/9781315139470 |
[10] | Brown M, Grundy M, Lin D, et al. (1999) Knowledge-Base Analysis of Microarray Gene Expression Data Using Support Vector Machines, University of California in Santa Cruz. https://doi.org/10.1073/pnas.97.1.262 |
[11] | Byanjankar A, Heikkilä M, Mezei J (2015) Predicting credit risk in peer-to-peer lending: A neural network approach. In 2015 IEEE symposium series on computational intelligence, IEEE, 719–725. https://doi.org/10.1109/SSCI.2015.109 |
[12] |
Cao A, He H, Chen Z, et al. (2018) Performance evaluation of machine learning approaches for credit scoring. Int J Econ Financ Manage Sci 6: 255–260. https://doi.org/10.11648/j.ijefm.20180606.12 doi: 10.11648/j.ijefm.20180606.12
![]() |
[13] | Chen S, Wang Q, Liu S (2019) Credit risk prediction in peer-to-peer lending with ensemble learning framework. In 2019 Chinese Control And Decision Conference (CCDC), IEEE, 4373–4377. https://doi.org/10.1109/CCDC.2019.8832412 |
[14] | Chen TQ, Guestrin C (2016) XGBoost: A Scalable Tree Boosting System. KDD'16 Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794. https://doi.org/10.1145/2939672.2939785 |
[15] |
Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20: 273–297. https://doi.org/10.1007/BF00994018 doi: 10.1007/BF00994018
![]() |
[16] | Crouhy M, Galai D, Mark R (2014) The Essentials of Risk Management, 2nd Edition. McGraw-Hill. Available from: https://www.mhprofessional.com/9780071818513-usa-the-essentials-of-risk-management-second-edition-group. |
[17] |
Cybenko G (1989) Approximation by Superpositions of a Sigmoidal Function Mathematics of Control. Signals Syst 2: 303–314. https://doi.org/10.1007/BF02551274 doi: 10.1007/BF02551274
![]() |
[18] |
Duan J (2019) Financial system modeling using deep neural networks (DNNs) for effective risk assessment and prediction. J Franklin Inst 356: 4716–4731. https://doi.org/10.1016/j.jfranklin.2019.01.046 doi: 10.1016/j.jfranklin.2019.01.046
![]() |
[19] |
Duchi J, Hazan E, Singer Y (2011) Adaptive Subgradient Methods for Online Learning and Stochastic Optimization. J Mach Learn Rese 12: 2121–2159. https://doi.org/10.5555/1953048.2021068 doi: 10.5555/1953048.2021068
![]() |
[20] | Elrahman SMA, Abraham A (2013) A Review of Class Imbalance Problem. J Network Innov Comput 1: 332–340. http://ias04.softcomputing.net/jnic2.pdf |
[21] |
Everett CR (2015) Group Membership, Relationship Banking and Loan Default Risk: the Case of Online Social Lending. Bank Financ Rev 7: 15–54. https://doi.org/10.2139/ssrn.1114428 doi: 10.2139/ssrn.1114428
![]() |
[22] |
Friedman JH (2001) Greedy Function Approximation: A Gradient Boosting Machine. Ann Stat 29: 1189–1232. https://doi.org/10.1214/aos/1013203451 doi: 10.1214/aos/1013203451
![]() |
[23] |
Genuer R, Poggi JM, Tuleau-Malot C (2010) Variable selection Using Random Forests. Pattern Recogn Lett 31: 2225–2236. https://doi.org/10.1016/j.patrec.2010.03.014 doi: 10.1016/j.patrec.2010.03.014
![]() |
[24] | Glorot X, Bengio Y (2010) Understanding the Difficulty of Training Deep Feedforward Neural Networks. J Mach Learn Res 9: 249–256. http://proceedings.mlr.press/v9/glorot10a.html |
[25] | Guyon I, ElNoeeff A (2003) An Introduction to Variable and Feature Selection. J Mach Learn Res 3: 1157–1182. https://www.jmlr.org/papers/v3/guyon03a.html |
[26] | Hastie T, Tibshirani R, Friedman JH (2009) The elements of statistical learning: data mining, inference, and prediction, Springer. https://doi.org/10.1007/978-0-387-84858-7 |
[27] | He KM, Zhang XY, Ren SQ, et al. (2015) Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification. IEEE international conference on computer vision. https://doi.org/10.1109/ICCV.2015.123 |
[28] | Ho TK (1995) Random Decision Forest, Proceeding of the 3rd International Conference on Document Analysis and Recognition, 278–282. https://doi.org/10.1109/ICDAR.1995.598994 |
[29] |
Ho TK (1998) The Random Subspace Method for Constructing Decision Forests. IEEE T Pattern Anal 20: 832–844. https://doi.org/10.1109/34.709601 doi: 10.1109/34.709601
![]() |
[30] | Hochreiter S, Bengio Y, Frasconi P, et al. (2001) Gradient Flow in Recurrent Nets: the Difficulty of Learning Long-Term Dependencies. In A Field Guide to Dynamical Recurrent Networks, IEEE, 237–243. https://doi.org/10.1109/9780470544037.ch14. |
[31] | Hsu CW, Chang CC, Lin CJ (2003) A Practical Guide to Support Vector Classification. National Taiwan University, 1–12. Available from: https://www.csie.ntu.edu.tw/~cjlin/papers/guide/guide.pdf. |
[32] |
Ioffe S, Szegedy C (2015) Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. International conference on machine learning, 448–456. https://doi.org/10.48550/arXiv.1502.03167 doi: 10.48550/arXiv.1502.03167
![]() |
[33] | Iyer R, Khwaja AI, Luttmer EF, et al. (2009) Screening in New Credit Markets: Can Individual Lenders Infer Borrower Creditworthiness in Peer-to-Peer Lending? AFA 2011 Denver Meetings Paper. https://doi.org/10.2139/ssrn.1570115 |
[34] |
Kang H (2013) The Prevention and Handling of the Missing Data. Korean J Anesthesiol 64: 402–406. https://doi.org/10.4097/kjae.2013.64.5.402 doi: 10.4097/kjae.2013.64.5.402
![]() |
[35] | Ke GL, Meng Q, Finley T, et al. (2017) LightGBM: A highly Efficient Gradient Boosting Decision Tree, Neural Information Processing Systems, 3149–3157. Available from: https://proceedings.neurips.cc/paper/2017/file/6449f44a102fde848669bdd9eb6b76fa-Paper.pdf. |
[36] | Keogh E, Mueen A (2017) Curse of Dimensionality. Encyclopedia of Machine Learning and Data Mining, Boston: Springer. https://doi.org/10.1007/978-1-4899-7687-1_192 |
[37] | Kingma DP, Ba JL (2015) Adam: a Method for Stochastic Optimization. International Conference on Learning Representations, 1–13. https://doi.org/10.48550/arXiv.1412.6980 |
[38] | Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet Classification with Deep Convolutional Neural Networks, Advances in Neural Information Processing Systems, 1097–1105. https://doi.org/10.1145/3065386 |
[39] | Lantz B (2013) Machine Learning with R. Packt Publishing Limited. Available from: https://edu.kpfu.ru/pluginfile.php/278552/mod_resource/content/1/MachineLearningR__Brett_Lantz.pdf. |
[40] | Li LH, Sharma AK, Ahmad R, Chen RC (2021) Predicting the Default Borrowers in P2P Platform Using Machine Learning Models. In International Conference on Artificial Intelligence and Sustainable Computing. https://doi.org/10.1007/978-3-030-82322-1_20 |
[41] | Lin HT, Lin CJ (2003) A Study on Sigmoid Kernels for SVM and the Training of Non-PSD Kernels by SMO-type Methods. National Taiwan University. Available from: https://www.csie.ntu.edu.tw/~cjlin/papers/tanh.pdf |
[42] | Lu L, Shin YJ, Su YH, et al. (2019) Dying ReLU and Initialization: Theory and Numerical Examples. arXiv preprint arXiv: 1903.06733. https://doi.org/10.4208/cicp.OA-2020-0165 |
[43] | Maas AL, Hannun AY, Ng AY (2013) Rectifier Nonlinearities Improve Neural Network Acoustic Models. ICML Workshop on Deep Learning for Audio, Speech, and Language Processing. Available from: https://ai.stanford.edu/~amaas/papers/relu_hybrid_icml2013_final.pdf. |
[44] | Madasamy K, Ramaswami M (2017) Data Imbalance and Classifiers: Impact and Solutions from a Big Data Perspective. Int J Comput Intell Res 13: 2267–2281. Available from: https://www.ripublication.com/ijcir17/ijcirv13n9_09.pdf. |
[45] |
McCulloch WS, Pitts W (1943) A Logical Calculus of the Ideas Immanent in Nervous Activity. Bull Math Biophys 5: 115–133. https://doi.org/10.2307/2268029 doi: 10.2307/2268029
![]() |
[46] | Mester LJ (1997) What's the Point of Credit Scoring? Bus Rev 3: 3–16. Available from: https://www.philadelphiafed.org/-/media/frbp/assets/economy/articles/business-review/1997/september-october/brso97lm.pdf. |
[47] | Mijwel MM (2018) Artificial Neural Networks Advantages and Disadvantages. Available from: https://www.linkedin.com/pulse/artificial-neural-networks-advantages-disadvantages-maad-m-mijwel/. |
[48] | Mills KG, McCarthy B (2016) The State of Small Business Lending: Innovation and Technology and the Implications for Regulation. HBS Working Paper No. 17-042. https://doi.org/10.2139/ssrn.2877201 |
[49] |
Mountcastle VB (1957) Modality and Topographic Properties of Single Neurons of Cat's Somatic Sensory Cortex. J Neurophysiol 20: 408–434. https://doi.org/10.1152/jn.1957.20.4.408 doi: 10.1152/jn.1957.20.4.408
![]() |
[50] |
Ohlson JA (1980) Financial Ratios and the Probabilistic Prediction of Bankruptcy. J Account Res 18: 109–131. https://doi.org/10.2307/2490395 doi: 10.2307/2490395
![]() |
[51] | Patro SGK, Sahu KK (2015) Normalization: A Preprocessing Stage. https://doi.org/10.17148/IARJSET.2015.2305 |
[52] |
Pontil M, Verri A (1998) Support Vector Machines for 3D Object Recognition. IEEE Trans PAMI 20: 637–646. https://doi.org/10.1109/34.683777 doi: 10.1109/34.683777
![]() |
[53] |
Qian N (1999) On the Momentum Term in Gradient Descent Learning Algorithms. Neural Networks 12: 145–151. https://doi.org/10.1016/S0893-6080(98)00116-6 doi: 10.1016/S0893-6080(98)00116-6
![]() |
[54] |
Quinlan JR (1987) Simplifying Decision Trees. Int J Man-Mach Stud 27: 221–234. https://doi.org/10.1016/S0020-7373(87)80053-6 doi: 10.1016/S0020-7373(87)80053-6
![]() |
[55] | Quinlan JR (1992) C4.5: Programs for Machine Learning. San Mateo: Morgan Kaufmann Publishers Inc. Available from: https://www.elsevier.com/books/c45/quinlan/978-0-08-050058-4. |
[56] |
Rajan U, Seru A, Vig V (2015) The Failure of Models that Predict Failure: Distance, Incentives, and Defaults. J Financ Econ 115: 237–260. https://doi.org/10.1016/j.jfineco.2014.09.012 doi: 10.1016/j.jfineco.2014.09.012
![]() |
[57] |
Rosenblatt F (1958) The Perceptron: A Probabilistic Model for Information Storage and Organization in the Brain. Psychol Rev 65: 386–408. https://doi.org/10.1037/h0042519 doi: 10.1037/h0042519
![]() |
[58] | Ruder S (2017) An Overview of Gradient Descent Optimization Algorithms. arXiv preprint arXiv: 1609.04747. https://doi.org/10.6919/ICJE.202102_7(2).0058 |
[59] |
Rumelhart DE, Hinton GE, Williams RJ (1986) Learning Representations by Back-Propagating Errors. Nature 323: 533–536. https://doi.org/10.1038/323533a0 doi: 10.1038/323533a0
![]() |
[60] | Samitsu A (2017) The Structure of P2P Lending and Legal Arrangements: Focusing on P2P Lending Regulation in the UK. IMES Discussion Paper Series, No. 17-J-3. Available from: https://www.boj.or.jp/en/research/wps_rev/lab/lab17e06.htm/ |
[61] | Serrano-Cinca C, Gutierrez-Nieto B, López-Palacios L (2015) Determinants of Default in P2P Lending. PloS One 10: e0139427. https://doi.org/10.1371/journal.pone.0139427 |
[62] |
Serrano-Cinca C, Gutiérrez-Nieto B (2016) The use of profit scoring as an alternative to credit scoring systems in peer-to-peer (P2P) lending. Deci Support Syst 89: 113–122. https://doi.org/10.1016/j.dss.2016.06.014 doi: 10.1016/j.dss.2016.06.014
![]() |
[63] |
Shannon C (1948) A Mathematical Theory of Communication. Bell Syst Tech J 27: 379–423. https://doi.org/10.1002/j.1538-7305.1948.tb01338.x doi: 10.1002/j.1538-7305.1948.tb01338.x
![]() |
[64] | Shelke MS, Deshmukh PR, Shandilya VK (2017) A Review on Imbalanced Data Handling using Undersampling and Oversampling Technique. Int J Recent Trends Eng Res. https://doi.org/10.23883/IJRTER.2017.3168.0UWXM |
[65] | Singh S, Gupta P (2014) Comparative Study Id3, Cart and C4.5 Decision Tree Algorithm: A Survey. Int J Adv Inf Sci Technol (IJAIST) 27: 97–103. https://doi.org/10.15693/ijaist/2014.v3i7.47-52 |
[66] | Srivastava N, Hinton G, Krizhevsky A, et al. (2014) Dropout: A Simple Way to Prevent Neural Networks from Overfitting. J Mach Learn Res 15: 1929–1958. https://doi.org/https://jmlr.org/papers/v15/srivastava14a.html |
[67] |
Thomas LC (2000) A Survey of Credit and Behavioural Scoring: Forecasting Financial Risk of Lending to Consumers. Int J Forecast 16: 149–172. https://doi.org/10.1016/S0169-2070(00)00034-0 doi: 10.1016/S0169-2070(00)00034-0
![]() |
[68] |
Tibshirani R (1996) Regression shrinkage and selection via the lasso. J Royal Stat Soc (Methodological) 58: 267–288. https://doi.org/10.1111/j.2517-6161.1996.tb02080.x doi: 10.1111/j.2517-6161.1996.tb02080.x
![]() |
[69] | Tieleman T, Hinton G (2012) Lecture 6.5—RMSProp, COURSERA: Neural Networks for Machine Learning. |
[70] | Verified Market Research (2021) Global Peer to Peer (P2P) Lending Market Size by Type, by End User, by Geographic Scope and Forecast. Available from: https://www.verifiedmarketresearch.com/product/peer-to-peer-p2p-lending-market/. |
[71] |
Wang Z, Cui P, Li FT, et al. (2014) A Data-Driven Study of Image Feature Extraction and Fusion. Inf Sci 281: 536–558. https://doi.org/10.1016/j.ins.2014.02.030 doi: 10.1016/j.ins.2014.02.030
![]() |
1. | Muhammad Zubair Mehboob, Arslan Hamid, Jeevotham Senthil Kumar, Xia Lei, Comprehensive characterization of pathogenic missense CTRP6 variants and their association with cancer, 2025, 25, 1471-2407, 10.1186/s12885-025-13685-0 |