RESUMO
Background: the credit scoring model is an effective tool for banks and other financial institutions to distinguish potential default borrowers. The credit scoring model represented by machine learning methods such as deep learning performs well in terms of the accuracy of default discrimination, but the model itself also has many shortcomings such as many hyperparameters and large dependence on big data. There is still a lot of room to improve its interpretability and robustness. Methods: the deep forest or multi-Grained Cascade Forest (gcForest) is a decision tree depth model based on the random forest algorithm. Using multidimensional scanning and cascading processing, gcForest can effectively identify and process high-dimensional feature information. At the same time, gcForest has fewer hyperparameters and has strong robustness. So, this paper constructs a two-stage hybrid default discrimination model based on multiple feature selection methods and gcForest algorithm, and at the same time, it optimizes the parameters for the lowest type II error as the first principle, and the highest AUC and accuracy as the second and third principles. GcForest can not only reflect the advantages of traditional statistical models in terms of interpretability and robustness but also take into account the advantages of deep learning models in terms of accuracy. Results: the validity of the hybrid default discrimination model is verified by three real open credit data sets of Australian, Japanese, and German in the UCI database. Conclusions: the performance of the gcForest is better than the current popular single classifiers such as ANN, and the common ensemble classifiers such as LightGBM, and CNNs in type II error, AUC, and accuracy. Besides, in comparison with other similar research results, the robustness and effectiveness of this model are further verified.
RESUMO
BACKGROUND: Recently, type 2 diabetic osteoporosis (T2DOP) has become a research hotspot for the complications of diabetes, but the specific mechanism of its occurrence and development remains unknown. Ferroptosis caused by iron overload is con-sidered an important cause of T2DOP. Polycytosine RNA-binding protein 1 (PCBP1), an iron ion chaperone, is considered a protector of ferroptosis. AIM: To investigate the existence of ferroptosis and specific role of PCBP1 in the development of type 2 diabetes. METHODS: A cell counting kit-8 assay was used to detect changes in osteoblast viability under high glucose (HG) and/or ferroptosis inhibitors at different concentrations and times. Transmission electron microscopy was used to examine the morphological changes in the mitochondria of osteoblasts under HG, and western blotting was used to detect the expression levels of PCBP1, ferritin, and the ferroptosis-related protein glutathione peroxidase 4 (GPX4). A lentivirus silenced and overexpressed PCBP1. Western blotting was used to detect the expression levels of the osteoblast functional proteins osteoprotegerin (OPG) and osteocalcin (OCN), whereas flow cytometry was used to detect changes in reactive oxygen species (ROS) levels in each group. RESULTS: Under HG, the viability of osteoblasts was considerably decreased, the number of mitochondria undergoing atrophy was considerably increased, PCBP1 and ferritin expression levels were increased, and GPX4 expression was decreased. Western blotting results demonstrated that infection with lentivirus overexpressing PCBP1, increased the expression levels of ferritin, GPX4, OPG, and OCN, compared with the HG group. Flow cytometry results showed a reduction in ROS, and an opposite result was obtained after silencing PCBP1. CONCLUSION: PCBP1 may protect osteoblasts and reduce the harm caused by ferroptosis by promoting ferritin expression under a HG environment. Moreover, PCBP1 may be a potential therapeutic target for T2DOP.