A novel framework of credit risk feature selection for SMEs during industry 4.0

With the development of industry 4.0, the credit data of SMEs are characterized by a large volume, high speed, diversity and low-value density. How to select the key features that affect the credit risk from the high-dimensional data has become the critical point to accurately measure the credit risk of SMEs and alleviate their financing constraints. In doing so, this paper proposes a credit risk feature selection approach that integrates the binary opposite whale optimization algorithm (BOWOA) and the Kolmogorov-Smirnov (KS) statistic. Furthermore, we use seven machine learning classifiers and three discriminant methods to verify the robustness of the proposed model by using three actual bank data from SMEs. The empirical results show that although no one artificial intelligence credit evaluation method is universal for different SMEs' credit data, the performance of the BOWOA-KS model proposed in this paper is better than other methods if the number of indicators in the optimal subset of indicators and the prediction performance of the classifier are considered simultaneously. By providing a high-dimensional data feature selection method and improving the predictive performance of credit risk, it could help SMEs focus on the factors that will allow them to improve their creditworthiness and more easily access loans from financial institutions. Moreover, it will also help government agencies and policymakers develop policies to help SMEs reduce their credit risks.

Optimal Adenosine Stress for Maximum Stress Perfusion, Coronary Flow Reserve, and Pixel Distribution of Coronary Flow Capacity by Kolmogorov-Smirnov Analysis

BACKGROUND

Different adenosine stress imaging protocols have not been systemically validated for absolute myocardial perfusion and coronary flow reserve (CFR) by positron emission tomography, where submaximal stress precludes assessing physiological severity of coronary artery disease.

METHODS AND RESULTS

In 127 volunteers, serial rest-stress positron emission tomography scans using rubidium-82 with various adenosine infusion protocols identified (1) the protocol with maximum stress perfusion and CFR, (2) test-retest precision in same subject, (3) stress perfusion and CFR after adenosine compared with dipyridamole, (4) heterogeneity of coronary flow capacity combining stress perfusion and CFR, and (5) potential relevance for patients with risk factors or coronary artery disease. The adenosine 6-minute infusion with rubidium-82 injection at 3 minutes caused CFR that was significantly 15.7% higher than the 4-minute adenosine infusion with rubidium-82 injection at 2 minutes and significantly more homogeneous by Kolmogorov-Smirnov analysis for histograms of 1344 pixel range of perfusion in paired positron emission tomographies. In a coronary artery disease cohort separate from volunteers of this study, compared with the 3/6-minute protocol, the 2/4-minute adenosine protocol would potentially have changed 332 of 1732 (19%) positron emission tomographies at low-risk physiological severity CFR ≥2.3 to CFR <2.0, thereby implying high-risk quantitative severity potentially appropriate for interventions but because of suboptimal stress of the 2/4 protocol in some patients.

CONCLUSIONS

The 6-minute adenosine infusion with rubidium-82 activation at 3 minutes produced CFR that averaged 15.7% higher than that in the 2/4-minute protocol, thereby potentially providing essential information for personalized management in some patients.