| Title | Application of Deep Learning Neural Network in Online Supply Chain Financial Credit Risk Assessment |
| Publication Type | Conference Paper |
| Year of Publication | 2020 |
| Authors | Xu, Rong-Zhen, He, Meng-Ke |
| Conference Name | 2020 International Conference on Computer Information and Big Data Applications (CIBDA) |
| Keywords | Companies, credit risk, data mining, DBN model, deep learning neural network, factor analysis, finance, Indexes, Metrics, online supply chain finance, pubcrawl, risk management, supply chain risk assessment, Supply chains, Support vector machines |
| Abstract | Under the background of "Internet +", in order to solve the problem of deeply mining credit risk behind online supply chain financial big data, this paper proposes an online supply chain financial credit risk assessment method based on deep belief network (DBN). First, a deep belief network evaluation model composed of Restricted Boltzmann Machine (RBM) and classifier SOFTMAX is established, and the performance evaluation test of three kinds of data sets is carried out by using this model. Using factor analysis to select 8 indicators from 21 indicators, and then input them into RBM for conversion to form a more scientific evaluation index, and finally input them into SOFTMAX for evaluation. This method of online supply chain financial credit risk assessment based on DBN is applied to an example for verification. The results show that the evaluation accuracy of this method is 96.04%, which has higher evaluation accuracy and better rationality compared with SVM method and Logistic method. |
| DOI | 10.1109/CIBDA50819.2020.00058 |
| Citation Key | xu_application_2020 |
Application of Deep Learning Neural Network in Online Supply Chain Financial Credit Risk Assessment