As the development of the modern economy is increasingly insep-arable from credit support,the traditional credit investigation mode has yet to meet this demand.Because of the difficulties in conventional credit data s...As the development of the modern economy is increasingly insep-arable from credit support,the traditional credit investigation mode has yet to meet this demand.Because of the difficulties in conventional credit data sharing among credit investigation agencies,poor data portability,and centralized supervision,this paper proposes a data-sharing scheme for credit investigation agencies based on a double blockchain.Given the problems such as difficult data sharing,difficult recovery of damaged data,and accessible data leakage between institutions and users with non-traditional credit inves-tigation data other than credit,this paper proposes a data-sharing scheme for credit investigation subjects based on the digital envelope.Based on the above two solutions,this paper designs a double blockchain credit data-sharing plat-form based on the“public chain+alliance chain”from credit investigation agencies’and visiting subjects’perspectives.The sharing platform uses the alliance chain as the management chain to solve the problem of complex data sharing between credit bureaus and centralized supervision,uses the public chain as the use chain to solve the problem of complex data sharing between the access subject and the credit bureaus,uses the interplanetary file system and digital envelope and other technologies to solve the problem of difficult recovery of damaged data,data leakage,and other issues.After the upload test,the average upload speed reaches 80.6 M/s.The average download speed of the system is 88.7 M/s after the download test.The multi-thread stress test tests the linkage port on the system package,and the average response time for the hypertext transfer protocol(HTTP)is 0.6 ms.The system performance and security analysis show that the sharing platform can provide safe and reliable credit-sharing services for organizations and users and high working efficiency.展开更多
Credit card fraudulent data is highly imbalanced, and it has presented an overwhelmingly large portion of nonfraudulent transactions and a small portion of fraudulent transactions. The measures used to judge the verac...Credit card fraudulent data is highly imbalanced, and it has presented an overwhelmingly large portion of nonfraudulent transactions and a small portion of fraudulent transactions. The measures used to judge the veracity of the detection algorithms become critical to the deployment of a model that accurately scores fraudulent transactions taking into account case imbalance, and the cost of identifying a case as genuine when, in fact, the case is a fraudulent transaction. In this paper, a new criterion to judge classification algorithms, which considers the cost of misclassification, is proposed, and several undersampling techniques are compared by this new criterion. At the same time, a weighted support vector machine (SVM) algorithm considering the financial cost of misclassification is introduced, proving to be more practical for credit card fraud detection than traditional methodologies. This weighted SVM uses transaction balances as weights for fraudulent transactions, and a uniformed weight for nonfraudulent transactions. The results show this strategy greatly improve performance of credit card fraud detection.展开更多
Personal credit risk assessment is an important part of the development of financial enterprises. Big data credit investigation is an inevitable trend of personal credit risk assessment, but some data are missing and ...Personal credit risk assessment is an important part of the development of financial enterprises. Big data credit investigation is an inevitable trend of personal credit risk assessment, but some data are missing and the amount of data is small, so it is difficult to train. At the same time, for different financial platforms, we need to use different models to train according to the characteristics of the current samples, which is time-consuming. <span style="font-family:Verdana;">In view of</span><span style="font-family:Verdana;"> these two problems, this paper uses the idea of transfer learning to build a transferable personal credit risk model based on Instance-based Transfer Learning (Instance-based TL). The model balances the weight of the samples in the source domain, and migrates the existing large dataset samples to the target domain of small samples, and finds out the commonness between them. At the same time, we have done a lot of experiments on the selection of base learners, including traditional machine learning algorithms and ensemble learning algorithms, such as decision tree, logistic regression, </span><span style="font-family:Verdana;">xgboost</span> <span style="font-family:Verdana;">and</span><span style="font-family:Verdana;"> so on. The datasets are from P2P platform and bank, the results show that the AUC value of Instance-based TL is 24% higher than that of the traditional machine learning model, which fully proves that the model in this paper has good application value. The model’s evaluation uses AUC, prediction, recall, F1. These criteria prove that this model has good application value from many aspects. At present, we are trying to apply this model to more fields to improve the robustness and applicability of the model;on the other hand, we are trying to do more in-depth research on domain adaptation to enrich the model.</span>展开更多
基金supported in part by the Advanced and High-level Discipline Construction Fund of Universities in Beijing(No.3201023)in part by the Beijing Electronic Science and Technology Institute of Basic Research Funds Outstanding Master Training Project(No.328202233)in part by the National First-class Undergraduate Discipline Construction of”Communication Engineering”and“Electronic Information Engineering,”and in part by the National Cryptography Science Foundation of China.
文摘As the development of the modern economy is increasingly insep-arable from credit support,the traditional credit investigation mode has yet to meet this demand.Because of the difficulties in conventional credit data sharing among credit investigation agencies,poor data portability,and centralized supervision,this paper proposes a data-sharing scheme for credit investigation agencies based on a double blockchain.Given the problems such as difficult data sharing,difficult recovery of damaged data,and accessible data leakage between institutions and users with non-traditional credit inves-tigation data other than credit,this paper proposes a data-sharing scheme for credit investigation subjects based on the digital envelope.Based on the above two solutions,this paper designs a double blockchain credit data-sharing plat-form based on the“public chain+alliance chain”from credit investigation agencies’and visiting subjects’perspectives.The sharing platform uses the alliance chain as the management chain to solve the problem of complex data sharing between credit bureaus and centralized supervision,uses the public chain as the use chain to solve the problem of complex data sharing between the access subject and the credit bureaus,uses the interplanetary file system and digital envelope and other technologies to solve the problem of difficult recovery of damaged data,data leakage,and other issues.After the upload test,the average upload speed reaches 80.6 M/s.The average download speed of the system is 88.7 M/s after the download test.The multi-thread stress test tests the linkage port on the system package,and the average response time for the hypertext transfer protocol(HTTP)is 0.6 ms.The system performance and security analysis show that the sharing platform can provide safe and reliable credit-sharing services for organizations and users and high working efficiency.
文摘Credit card fraudulent data is highly imbalanced, and it has presented an overwhelmingly large portion of nonfraudulent transactions and a small portion of fraudulent transactions. The measures used to judge the veracity of the detection algorithms become critical to the deployment of a model that accurately scores fraudulent transactions taking into account case imbalance, and the cost of identifying a case as genuine when, in fact, the case is a fraudulent transaction. In this paper, a new criterion to judge classification algorithms, which considers the cost of misclassification, is proposed, and several undersampling techniques are compared by this new criterion. At the same time, a weighted support vector machine (SVM) algorithm considering the financial cost of misclassification is introduced, proving to be more practical for credit card fraud detection than traditional methodologies. This weighted SVM uses transaction balances as weights for fraudulent transactions, and a uniformed weight for nonfraudulent transactions. The results show this strategy greatly improve performance of credit card fraud detection.
文摘Personal credit risk assessment is an important part of the development of financial enterprises. Big data credit investigation is an inevitable trend of personal credit risk assessment, but some data are missing and the amount of data is small, so it is difficult to train. At the same time, for different financial platforms, we need to use different models to train according to the characteristics of the current samples, which is time-consuming. <span style="font-family:Verdana;">In view of</span><span style="font-family:Verdana;"> these two problems, this paper uses the idea of transfer learning to build a transferable personal credit risk model based on Instance-based Transfer Learning (Instance-based TL). The model balances the weight of the samples in the source domain, and migrates the existing large dataset samples to the target domain of small samples, and finds out the commonness between them. At the same time, we have done a lot of experiments on the selection of base learners, including traditional machine learning algorithms and ensemble learning algorithms, such as decision tree, logistic regression, </span><span style="font-family:Verdana;">xgboost</span> <span style="font-family:Verdana;">and</span><span style="font-family:Verdana;"> so on. The datasets are from P2P platform and bank, the results show that the AUC value of Instance-based TL is 24% higher than that of the traditional machine learning model, which fully proves that the model in this paper has good application value. The model’s evaluation uses AUC, prediction, recall, F1. These criteria prove that this model has good application value from many aspects. At present, we are trying to apply this model to more fields to improve the robustness and applicability of the model;on the other hand, we are trying to do more in-depth research on domain adaptation to enrich the model.</span>
