CN113011632B - Enterprise risk assessment method, device, equipment and computer readable storage medium - Google Patents
Enterprise risk assessment method, device, equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN113011632B CN113011632B CN202110128783.1A CN202110128783A CN113011632B CN 113011632 B CN113011632 B CN 113011632B CN 202110128783 A CN202110128783 A CN 202110128783A CN 113011632 B CN113011632 B CN 113011632B
- Authority
- CN
- China
- Prior art keywords
- enterprise
- service platform
- government
- data
- federal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/20—Ensemble learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Abstract
The invention discloses a method, a device and equipment for enterprise risk assessment and a computer readable storage medium, wherein the method comprises the following steps: acquiring first enterprise data of a target enterprise which proposes a business application in the bank service platform; and performing federal calculation on the first enterprise data and second enterprise data by adopting a federal model obtained by pre-training through a combined government service platform to obtain a risk assessment result of the target enterprise, wherein the second enterprise data is the enterprise data of the target enterprise in the government service platform. According to the method and the system, the bank service platform can combine the first enterprise data and the second enterprise data to carry out risk assessment under the condition that the government side data are not leaked, so that the data basis for carrying out risk assessment on the enterprises is expanded, the accuracy of the enterprise risk assessment result is improved, and the accuracy of bank customer screening is further improved.
Description
Technical Field
The invention relates to the technical field of wind control, in particular to an enterprise risk assessment method, device and equipment and a computer readable storage medium.
Background
When a bank applies for actual business to an enterprise, risk assessment needs to be performed on the enterprise by combining data of all aspects of the enterprise, and then the enterprise is screened to determine whether business application of the enterprise is received. In order to improve the accuracy of a risk assessment result and the accuracy of screening customers, a bank needs to participate in the risk assessment by means of external data, and at present, the bank usually adopts a mode of purchasing an external data source, takes an original value of the external data or a mapped interval value as a characteristic, and brings credit risk data of the bank into a wind control model together. However, important index information of some small and medium-sized enterprises is difficult to acquire in government institutions and government non-public information, and data transmission has risks in data security, privacy protection and the like, so that important external data is difficult to acquire, and therefore the bank customer screening accuracy is low and the wind control management is limited.
Disclosure of Invention
The invention mainly aims to provide an enterprise risk assessment method, an enterprise risk assessment device, enterprise risk assessment equipment and a computer readable storage medium, and aims to solve the problem that the bank customer screening accuracy is low due to the fact that the data acquisition difficulty of government institutions is high at present.
In order to achieve the above object, the present invention provides an enterprise risk assessment method, which is applied to a banking service platform, and comprises the following steps:
acquiring first enterprise data of a target enterprise which proposes a business application in the bank service platform;
and performing federal calculation on the first enterprise data and second enterprise data by adopting a federal model obtained by pre-training through a combined government service platform to obtain a risk assessment result of the target enterprise, wherein the second enterprise data is the enterprise data of the target enterprise in the government service platform.
Optionally, before the step of obtaining the first enterprise data of the target enterprise that proposed the business application in the banking service platform, the method further includes:
and receiving the business application sent by the branch business system, wherein the branch business system sends the business application to the bank service platform after receiving the business application triggered by the target enterprise based on the enterprise client.
Optionally, after the step of performing federal calculation on the first enterprise data and the second enterprise data by using a federal model obtained by pre-training by the united government service platform to obtain a risk assessment result of the target enterprise, the method further includes:
and sending the risk evaluation result to the branch business system so that the branch business system screens the target enterprise based on the risk evaluation result to obtain an application result of the business application.
Optionally, before the step of obtaining the first enterprise data of the target enterprise that proposes the business application in the banking service platform, the method further includes:
receiving government agency data fields sent by the branch service system;
sending a modeling application and the government agency data fields to the government service platform for the government service platform to review the modeling application and the government agency data fields;
and after the fact that the audit of the government service platform is passed is detected, the federal model is obtained by combining the government service platform and performing federal modeling.
Optionally, the step of obtaining the federal model by federating the government service platform to perform federal modeling includes:
calling a federal learning service node of the bank service platform to obtain first enterprise sample data in a local database;
performing encryption sample alignment on the first enterprise sample data and the government service platform to obtain a first training sample, wherein the government service platform adopts second enterprise sample data corresponding to the government agency data field in the other-end database to participate in encryption sample alignment to obtain a second training sample aligned with the first training sample in a sample dimension;
and carrying out federal modeling by adopting the first training samples and the government service platform to obtain the federal model, wherein the government service platform adopts the second training samples to participate in the federal modeling.
Optionally, the federal model is obtained by using the first training sample to jointly perform federal modeling on the government service platform, wherein the step of using the second training sample by the government service platform to participate in federal modeling includes:
receiving a first encrypted intermediate result sent by the government service platform, wherein the government service platform calculates to obtain a first intermediate result based on the second training sample, and encrypts the first intermediate result by using a public key received from a third party platform to obtain the first encrypted intermediate result;
calling the federal learning service node to calculate to obtain a first encryption gradient value and a second encryption intermediate result based on the first encryption intermediate result and the first training sample, sending the first encryption gradient value to the third-party platform to be decrypted to obtain a first gradient value, and updating the local-end federal model to be trained by adopting the first gradient value;
calling the federal learning service node to send the second encrypted intermediate result to the government service platform so that the government service platform can calculate a second encrypted gradient value according to the second encrypted intermediate result, sending the second encrypted gradient value to the third party platform to decrypt to obtain a second gradient value, and updating the federal model to be trained at the other end by adopting the second gradient value;
and calling the federal learning service node to obtain the federal model based on the updated local-end to-be-trained federal model.
Optionally, after the step of performing federal calculation on the first enterprise data and the second enterprise data by using a federal model obtained by pre-training by the combined government service platform to obtain the risk assessment result of the target enterprise, the method further includes:
when detecting the modification or deletion operation of the federal model, acquiring operation data;
and calling a block chain link point of the bank service platform to upload the operation data to a block chain, wherein the block chain is used for recording the operation data of the bank service platform and the government service platform on the federal model.
In order to achieve the above object, the present invention further provides an enterprise risk assessment apparatus, where the enterprise risk assessment apparatus is deployed on a banking service platform, and the enterprise risk assessment apparatus includes:
the acquisition module is used for acquiring first enterprise data of a target enterprise which proposes a business application in the bank service platform;
and the federal calculation module is used for performing federal calculation on the first enterprise data and the second enterprise data by adopting a federal model obtained by pre-training through a united government service platform to obtain a risk assessment result of the target enterprise, wherein the second enterprise data is the enterprise data of the target enterprise in the government service platform.
In order to achieve the above object, the present invention further provides an enterprise risk assessment apparatus, including: a memory, a processor, and an enterprise risk assessment program stored on the memory and executable on the processor, the enterprise risk assessment program when executed by the processor implementing the steps of the enterprise risk assessment method as described above.
In addition, to achieve the above object, the present invention further provides a computer readable storage medium, which stores an enterprise risk assessment program, and when the enterprise risk assessment program is executed by a processor, the computer implements the steps of the enterprise risk assessment method as described above.
In the invention, first enterprise data of a target enterprise which provides a business application in a bank service platform is obtained through the bank service platform, and the first enterprise data and second enterprise data of the target enterprise in the government service platform are subjected to federal calculation by adopting a federal model obtained by pre-training in combination with the government service platform, so that a risk evaluation result of the target enterprise is obtained; because the bank service platform and the government service platform adopt the federal model to carry out federal calculation to obtain enterprise risk assessment results, federal learning is a parameter exchange mode under an encryption mechanism to protect the privacy of user data, the data and the model cannot be transmitted, and the original data of the bottom layer of the government cannot be reversely pushed, so that the possibility of leakage does not exist in the data layer, and the requirements of laws and regulations for related data protection are met, so that the bank service platform can carry out risk assessment by combining the first enterprise data and the second enterprise data under the condition of ensuring that the data of the government is not leaked, thereby expanding the data basis for carrying out risk assessment on enterprises, improving the accuracy of enterprise risk assessment results, and further improving the accuracy of bank screening.
Drawings
FIG. 1 is a schematic diagram of a hardware operating environment according to an embodiment of the present invention;
FIG. 2 is a schematic flowchart of a first embodiment of a method for enterprise risk assessment according to the present invention;
FIG. 3 is a schematic diagram of an enterprise risk assessment architecture according to an embodiment of the present invention;
FIG. 4 is a functional diagram of an enterprise risk assessment device according to a preferred embodiment of the present invention.
The implementation, functional features and advantages of the present invention will be further described with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not limit the invention.
As shown in fig. 1, fig. 1 is a schematic device structure diagram of a hardware operating environment according to an embodiment of the present invention.
It should be noted that, the enterprise risk assessment device in the embodiment of the present invention may be a smart phone, a personal computer, a server, and the like, and is not limited herein.
As shown in fig. 1, the enterprise risk assessment device may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a communication bus 1002. Wherein a communication bus 1002 is used to enable connective communication between these components. The user interface 1003 may include a Display screen (Display), an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., a WI-FI interface). The memory 1005 may be a high-speed RAM memory or a non-volatile memory such as a disk memory. The memory 1005 may alternatively be a storage device separate from the processor 1001 described previously.
Those skilled in the art will appreciate that the configuration of the device shown in FIG. 1 does not constitute a limitation of an enterprise risk assessment device, and may include more or fewer components than shown, or some components in combination, or a different arrangement of components.
As shown in fig. 1, a memory 1005, which is a type of computer storage medium, may include an operating system, a network communication module, a user interface module, and an enterprise risk assessment program therein.
In the device shown in fig. 1, the user interface 1003 is mainly used for data communication with a client; the network interface 1004 is mainly used for establishing communication connection with other terminals participating in federal learning; and processor 1001 may be configured to invoke an enterprise risk assessment program stored in memory 1005 and perform the following operations:
the enterprise risk assessment method is applied to a bank service platform, and comprises the following steps:
acquiring first enterprise data of a target enterprise which proposes a business application in the bank service platform;
and performing federal calculation on the first enterprise data and second enterprise data by adopting a federal model obtained by pre-training through a combined government service platform to obtain a risk assessment result of the target enterprise, wherein the second enterprise data is the enterprise data of the target enterprise in the government service platform.
Further, before the step of obtaining the first enterprise data of the target enterprise which proposes the business application in the banking service platform, the method further includes:
and receiving the business application sent by the branch business system, wherein the branch business system sends the business application to the bank service platform after receiving the business application triggered by the target enterprise based on the enterprise client.
Further, after the step of performing federal calculation on the first enterprise data and the second enterprise data by using a federal model obtained by pre-training by the united government service platform to obtain a risk assessment result of the target enterprise, the method further includes:
and sending the risk evaluation result to the branch business system so that the branch business system screens the target enterprise based on the risk evaluation result to obtain an application result of the business application.
Further, before the step of obtaining the first enterprise data of the target enterprise which proposes the business application in the banking service platform, the method further includes:
receiving government agency data fields sent by the branch service system;
sending a modeling application and the government agency data fields to the government service platform for the government service platform to review the modeling application and the government agency data fields;
and after the fact that the audit of the government service platform is passed is detected, the federal model is obtained by combining the government service platform and performing federal modeling.
Further, the step of obtaining the federal model by federating the government service platform to perform federal modeling comprises the following steps:
calling a federal learning service node of the bank service platform to obtain first enterprise sample data in a local database;
performing encryption sample alignment on the first enterprise sample data and the government service platform to obtain a first training sample, wherein the government service platform adopts second enterprise sample data corresponding to the government agency data field in the other-end database to participate in encryption sample alignment to obtain a second training sample aligned with the first training sample in a sample dimension;
and performing federal modeling by combining the first training sample and the government service platform to obtain the federal model, wherein the government service platform adopts the second training sample to participate in the federal modeling.
Further, the federal model is obtained by using the first training sample in combination with the government service platform for federal modeling, wherein the step of using the second training sample by the government service platform to participate in federal modeling includes:
receiving a first encrypted intermediate result sent by the government service platform, wherein the government service platform calculates to obtain the first intermediate result based on the second training sample, and encrypts the first intermediate result by adopting a public key received from a third party platform to obtain the first encrypted intermediate result;
calling the federal learning service node to calculate to obtain a first encryption gradient value and a second encryption intermediate result based on the first encryption intermediate result and the first training sample, sending the first encryption gradient value to the third-party platform to be decrypted to obtain a first gradient value, and updating the local-end federal model to be trained by adopting the first gradient value;
calling the federal learning service node to send the second encrypted intermediate result to the government service platform so that the government service platform can calculate according to the second encrypted intermediate result to obtain a second encrypted gradient value, sending the second encrypted gradient value to the third party platform to decrypt to obtain a second gradient value, and updating the federal model to be trained at the other end by adopting the second gradient value;
and calling the federal learning service node to obtain the federal model based on the updated local-end to-be-trained federal model.
Further, after the step of performing federal calculation on the first enterprise data and the second enterprise data by using a federal model obtained by pre-training by the joint government service platform to obtain the risk assessment result of the target enterprise, the method further includes:
when detecting the modification or deletion operation of the federal model, acquiring operation data;
and calling a block chain link point of the bank service platform to upload the operation data to a block chain, wherein the block chain is used for recording the operation data of the bank service platform and the government service platform on the federal model.
Based on the structure, various embodiments of the enterprise risk assessment method are provided.
While a logical order is shown in the flowcharts, in some cases, the steps shown or described may be performed in an order different than here. The federal learning privacy data processing method in the first embodiment of the present invention is applied to a bank service platform, and the bank service platform and the government service platform in the embodiments of the present invention may be a smart phone, a personal computer, a server, and other devices, and are not particularly limited herein. Referring to fig. 2, fig. 2 is a flowchart illustrating a first embodiment of the enterprise risk assessment method according to the present invention. In this embodiment, the method for enterprise risk assessment includes:
step S10, acquiring first enterprise data of a target enterprise which proposes a business application in the bank service platform;
the bank provides various business application items, such as various loan application items, to the enterprise, the enterprise can submit business applications to the bank in various ways, and the bank needs to perform risk assessment on the enterprise submitting the business applications so as to filter the enterprise based on assessment results and determine whether to receive the business applications of the enterprise. The bank service platform can be deployed in a bank, and after receiving the business application or after determining a target enterprise which provides the business application, the bank service platform can acquire first enterprise data corresponding to the target enterprise in the bank service platform. The business application submitted by the enterprise can carry related enterprise data, such as data of enterprise legal persons, enterprise assets and the like, and the bank service platform can take the data in the business application as first enterprise data; or, the bank or the bank branch stores enterprise data of the target enterprise, such as the number of enterprise loans, the corporate legal property, the number of historical defaults, and the like, in the course of historical business transactions with the target enterprise, the enterprise data is stored in a database of a bank service platform or a branch service platform, and the bank service platform can call the enterprise data as the first enterprise data.
And S20, performing federal calculation on the first enterprise data and second enterprise data by using a federal model obtained by pre-training through a united government service platform to obtain a risk assessment result of the target enterprise, wherein the second enterprise data is the enterprise data of the target enterprise in the government service platform.
The government organization can deploy a government service platform, wherein the database of the government service platform stores enterprise data of each enterprise, such as industrial and commercial information, social security payment information, tax information, various red and black list information and the like of the enterprise, and the data fields of the enterprise data are different or partially different from those of the enterprise data of the bank side. The bank service platform and the government service platform can adopt enterprise data of different data fields of the database to construct training samples, adopt the training samples to participate in federal modeling, and train to obtain a federal model, wherein after the federal model is trained, part of the trained federal model is deployed on the bank service platform, and the other part of the trained federal model is deployed on the government service platform. The method for federal modeling by both parties can refer to the existing longitudinal federal modeling method, and details are not described in this embodiment. The model structure used for modeling may be a machine learning model structure such as a decision tree model, a logistic regression model, or a gradient lifting tree, which is not limited in this embodiment. The output of the federal model may be set as a result related to the risk of the enterprise, and may be different according to the specific requirements of the business application project, for example, it may be set as outputting the risk value of the enterprise, or outputting the estimated loan amount for the enterprise, etc.
After the bank service platform obtains the first enterprise data of the target enterprise, the combined government service platform performs federal calculation on the first enterprise data and the second enterprise data by adopting the pre-trained federal model to obtain a risk assessment result of the target enterprise. And the second enterprise data is the enterprise data of the target enterprise on the administrative service platform. Specifically, the banking service platform may send a federal calculation request to the government service platform, where the request carries an identifier of the target enterprise, such as a name or a unique number of the target enterprise. And after receiving the federal calculation request, the government service platform searches the enterprise data of the target enterprise from local or other equipment to serve as second enterprise data. It is to be understood that the second enterprise data is data under data fields employed by the target enterprise in federal modeling conducted by the government service platform, and the first enterprise data is data under data fields employed by the target enterprise in federal modeling conducted by the bank service platform. And after the government service platform finds the second enterprise data, carrying out federal calculation with the bank service platform. In an embodiment, if the government service platform does not find the second enterprise data of the target enterprise, feedback information indicating that federal calculation cannot be performed may be fed back to the bank service platform, and after receiving the feedback information, the bank service platform performs risk assessment of the target enterprise according to the first enterprise data.
After the risk assessment result is obtained, the bank service platform can output the risk assessment result, so that business personnel of a bank can screen the target enterprise according to the risk assessment result to determine whether to accept business application of the target enterprise. Or, the banking service platform may also directly screen the target enterprise according to the risk assessment result to obtain an application result indicating whether to accept the business application, for example, when the risk assessment result is a risk value of the enterprise, the banking service platform may compare the obtained risk value with a preset threshold, and when the risk value is greater than the preset threshold, determine to reject the business application, otherwise determine to receive the business application.
Further, in an embodiment, data fields adopted when enterprise risk assessment is performed on different business application projects may be different, so that the bank service platform can perform multi-time federal modeling by using data of different data fields in combination with the government service platform to obtain federal models for different business application projects. When a business application is received, a bank service platform firstly determines a data field and a federal model corresponding to the business application, and then a united government service platform adopts the federal model to carry out federal calculation on enterprise data under the data field to obtain a risk assessment result of a target enterprise aiming at the business application, so that the relevance between the risk assessment result and an actual business is improved, namely the accuracy of the risk assessment result of the bank in a specific subdivision business is improved, and the passenger screening accuracy in the specific subdivision business can be improved.
In the embodiment, first enterprise data of a target enterprise which provides a business application in a bank service platform is obtained through the bank service platform, and the first enterprise data and second enterprise data of the target enterprise in the government service platform are subjected to federal calculation by adopting a federal model obtained by pre-training through a combined government service platform, so that a risk assessment result of the target enterprise is obtained; because the bank service platform and the government service platform adopt the federation model to carry out federation calculation to obtain an enterprise risk assessment result, federation learning is a parameter exchange mode under an encryption mechanism to protect the privacy of user data, the data and the model cannot be transmitted, and the original data at the bottom of the government cannot be reversely deduced, the possibility of leakage does not exist in a data layer, and the requirements of laws and regulations for related data protection are met, so that the bank service platform can carry out risk assessment by combining the first enterprise data and the second enterprise data under the condition of ensuring that the data at the government side is not leaked, the data basis for carrying out risk assessment on enterprises is expanded, the accuracy of enterprise risk assessment results is improved, and the accuracy of bank screening is further improved.
Further, based on the first embodiment, a second embodiment of the enterprise risk assessment method according to the present invention is provided, in this embodiment, before the step S10, the method further includes:
and step S30, receiving the service application sent by the branch service system, wherein the branch service system sends the service application to the bank service platform after receiving the service application triggered by the target enterprise based on the enterprise client.
In this embodiment, the banking service platform may be a head office deployed in a bank, and a branch business system is deployed in each branch of the bank, and is used for receiving a business application sent by an enterprise client. The enterprise can trigger business application in the enterprise client, the enterprise client sends the business application to a corresponding branch business system, and the branch business system sends the business application of the enterprise to the bank service platform when detecting that the risk assessment is needed to be carried out on the enterprise. Specifically, the branch service system may determine whether risk assessment needs to be performed on the enterprise by detecting whether the type of the service application belongs to a preset service application type that needs risk assessment.
After calculating the risk assessment result of the target enterprise, the bank service platform may send the risk assessment result to the branch business system, or may screen the target enterprise based on the risk assessment result, and send the application result to the branch business system after obtaining the application result corresponding to the business application. After receiving the risk assessment result, the branch business system can output the risk assessment result, so that business personnel can screen the target enterprise according to the risk assessment result to determine whether to accept the business application of the target enterprise. Or after receiving the application result of the service application, the branch service system can send the application result to the enterprise client end submitting the service application so that the target enterprise can obtain the application result.
Further, in an embodiment, after the step S20, the method further includes:
and S40, sending the risk evaluation result to the branch business system so that the branch business system can screen the target enterprise based on the risk evaluation result to obtain an application result of the business application.
After the bank service platform calculates the risk assessment result, the bank service platform can send the risk assessment result to a branch business system submitting business application. After receiving the risk assessment result, the branch service system screens the target enterprise based on the risk assessment result to obtain an application result of the service application, for example, when the risk assessment result is a risk value of the enterprise, the branch service system may compare the obtained risk value with a preset threshold, and when the risk value is greater than the preset threshold, determine to reject the service application, otherwise determine to receive the service application.
Further, based on the second embodiment, a third embodiment of the enterprise risk assessment method according to the present invention is provided, and in this embodiment, the method further includes:
step S50, receiving government agency data fields sent by the branch business system;
in this embodiment, before the bank service platform and the government service platform perform federal modeling, the branch business system may send, to the bank service platform, the government agency data fields required for federal modeling based on business requirements. The data field refers to a certain data item describing enterprise information, such as enterprise assets and enterprise credit values, belonging to the data field, and the government agency data field refers to a data field owned by a government agency. The branch service personnel may select the government agency data fields required for modeling, i.e., what types of data are required by the government agency to participate in modeling, in the administrative menu provided by the branch service system based on the requirements of the service application project. And after receiving the government agency data fields selected by the business personnel, the branch business system sends the selected government agency data fields to the bank service platform.
Or in other embodiments, the head office business system may provide a management menu, and business personnel of each branch can log in the management menu according to the authority, and select the government agency data fields required by modeling in the management menu; and the head office business system receives the government agency data fields selected by the branch business personnel in the management menu, and then carries out the next operation.
Step S60, sending a modeling application and the government agency data fields to the government service platform so that the government service platform can check the modeling application and the government agency data fields;
after receiving the government agency data fields, the banking service platform may send the modeling application and the government agency data fields to the government service platform. Wherein the modeling application is used to indicate a need for joint modeling with a government service platform. And after receiving the modeling application and the government agency data fields, the government service platform checks the modeling application and the government agency data fields. Specifically, the government service platform may check whether the data field of the government agency belongs to a preset data field allowed to be used, if so, determine that the check is passed, and if not, determine that the check fails. The government service platform can send the auditing result to the bank service platform; if the audit result is that the audit is passed, the government service platform can also directly send an instruction of agreeing to carry out federal modeling to the bank service platform.
And S70, after the government service platform is checked to be passed, performing federal modeling by combining the government service platform to obtain the federal model.
After the bank service platform receives the auditing result of the approval or receives the instruction of agreeing to carry out the federal modeling, namely, after the bank service platform detects that the government service platform approves the approval, the bank service platform can carry out the federal modeling in combination with the government service platform to obtain the federal model. In federal modeling, a government services platform uses enterprise data received under the data fields of the government agencies to participate in federal modeling.
Further, the step S70 of performing federal modeling in combination with the government service platform to obtain the federal model includes:
step S701, calling a federal learning service node of the bank service platform to obtain first enterprise sample data in a local database;
in one embodiment, the banking service platform comprises a federal learning service node and a database, wherein the federal learning service node is used for performing federal modeling with a government service platform, and the database is used for storing enterprise data. The banking service platform can call the federal learning service node to obtain the first enterprise sample data in a local database (namely, a database in the banking service platform). The first enterprise sample data comprises a plurality of pieces of sample data, and each piece of sample data comprises data of an enterprise under a data field of a bank side.
Step S702, performing encryption sample alignment on the first enterprise sample data and the government service platform to obtain a first training sample, wherein the government service platform adopts second enterprise sample data corresponding to the government agency data field in the other-end database to participate in the encryption sample alignment to obtain a second training sample aligned with the first training sample in a sample dimension;
the bank service platform adopts the first enterprise sample data and the government service platform to align the encrypted samples to obtain a first training sample. The government service platform also comprises a federal learning service node and a database, which are called as other-end federal learning service node and other-end database for showing differences. The database of the government service platform and the database of the bank service platform can adopt the same data structure, so the database of the government service platform and the database of the bank service platform can be called as a mirror database. And the government service platform calls the other-end federal learning service node to acquire second enterprise sample data from the other-end database, wherein the second enterprise sample data also comprises a plurality of pieces of sample data, and each piece of sample data comprises data of an enterprise under the received government structure data field. The government service platform adopts the second enterprise sample data to participate in the alignment with the encrypted sample of the bank service platform, so as to obtain a second training sample aligned with the first training sample in the sample dimension, namely, the same enterprise and different enterprises exist in the first enterprise sample data and the second enterprise sample data, after the encrypted samples are aligned, different enterprises are removed, and the obtained enterprises in the first training sample are the same as those in the second training sample. Specifically, the encrypted sample alignment method may refer to an encrypted sample alignment method in the existing longitudinal federal learning, and details are not described in detail in this embodiment.
Step S703, performing federal modeling by using the first training sample and the government service platform to obtain the federal model, wherein the government service platform uses the second training sample to participate in the federal modeling.
After the encrypted samples are aligned, the bank service platform adopts the first training sample to perform federal modeling in combination with a government service platform to obtain a federal model. Wherein the government service platform uses the second training sample to participate in federal modeling. In particular, the banking service platform may invoke its federal learning service node in conjunction with the federal learning service node of a government service platform to federate modeling for the first training sample and the second training sample.
Further, in an embodiment, the step S703 includes:
step S7031, receiving a first encrypted intermediate result sent by the government service platform, wherein the government service platform calculates to obtain the first intermediate result based on the second training sample, and encrypts the first intermediate result by using a public key received from a third party platform to obtain the first encrypted intermediate result;
step S7032, the federal learning service node is called to calculate a first encryption gradient value and a second encryption intermediate result based on the first encryption intermediate result and the first training sample, the first encryption gradient value is sent to the third-party platform to be decrypted to obtain a first gradient value, and the first gradient value is adopted to update the federal model to be trained at the local terminal;
step S7033, the federal learning service node is called to send the second encrypted intermediate result to the government service platform, so that the government service platform can calculate a second encrypted gradient value according to the second encrypted intermediate result, send the second encrypted gradient value to the third party platform to be decrypted to obtain a second gradient value, and update the federal model to be trained of the other end by adopting the second gradient value;
and S7034, calling the Federal learning service node to obtain the Federal model based on the updated Federal model to be trained at the home terminal.
The bank service platform and the government service platform respectively comprise a federal model to be trained, and the model structures of the two federal models to be trained are complementary and combined to form a complete model. The bank service platform calls a federal learning service node in the bank service platform to participate in modeling. Specifically, a private key and a public key for encryption and decryption are set in the third-party platform, and the third-party platform sends the public key to the bank service platform and the government service platform. The government service platform inputs the second training sample into a federal model to be trained of the other end (the other end refers to the government service platform) to be processed to obtain model output, and the model output is used as a first intermediate result; the government service platform encrypts the first intermediate result by adopting the public key to obtain a first encrypted intermediate result, and sends the first encrypted intermediate result to the bank service platform. After receiving the first encrypted intermediate result, the bank service platform calls a federal learning service node to input the first encrypted intermediate result and a first training sample into a to-be-trained federal model of a home terminal (the home terminal refers to the bank service platform) for processing, and a risk assessment result corresponding to the training sample is obtained; calling a federal learning service node to calculate a loss value according to a risk evaluation result and a real risk label corresponding to a training sample, and calculating to obtain a gradient value corresponding to a first encryption intermediate result and a gradient value corresponding to each parameter in a federal model to be trained at the home terminal according to the loss value, wherein the first encryption intermediate result is a ciphertext state, so the calculated gradient value is also the ciphertext state, the gradient value corresponding to the first encryption intermediate result of the ciphertext state is used as a second encryption intermediate result, and the gradient value corresponding to each parameter of the ciphertext state is used as a first encryption gradient value; calling the federal learning service node to send the first encryption gradient value to a third-party platform, decrypting the first encryption gradient value by a private key by the third-party platform and then returning the first gradient value, and updating each parameter in the federal model to be trained at the home terminal by the federal learning service node according to the first gradient value so as to complete one round of updating of the federal model to be trained at the home terminal; and the federal learning service node sends the second encrypted intermediate result to the government service platform. After receiving the second encrypted intermediate result, the government service platform calculates a second encrypted gradient value corresponding to the parameter in the federal model to be trained at the other end by adopting the second encrypted intermediate result according to a gradient back propagation algorithm; the second encryption gradient value is sent to a third party platform, and the third party platform decrypts the second encryption gradient value by using a private key and returns the second encryption gradient value; and the government service platform updates the parameters in the federal model to be trained of the other end by adopting the second gradient value so as to complete one round of updating of the federal model to be trained of the other end.
And after updating respective federal models to be trained by the two parties, if the convergence condition is detected to be met, taking the updated federal model to be trained as the federal model of the local terminal, and if the convergence condition is detected to be not met, performing a round of model updating on the basis of the updated federal model to be trained until the convergence condition is detected to be met. Wherein, the convergence condition may be that the loss value set to the federal model is less than a certain threshold. Specifically, the bank service platform may send the loss value to the third party platform after calculating the loss value (in a ciphertext state), and the third party platform may determine whether the loss value is smaller than a certain threshold value after decrypting the loss value by using a private key, and send the determination result to the bank service platform and the government service platform. After training is finished, the bank service platform and the government service platform can respectively have respective models, and the two models are combined to be used as an enterprise risk assessment model; or the bank service platform and the government service platform send each part of model to the other party, so that both parties have a complete enterprise risk assessment model.
In the embodiment, the bank service platform and the government service platform adopt respective training samples to carry out federal modeling, and the bank service platform and the government service platform do not directly interact original sample data in the modeling process but send encrypted intermediate results for calculating the gradient value, so that the original data of the two parties cannot be leaked in the modeling process, the bank can construct a federal model by using enterprise data shared by government institutions, the prediction accuracy of the federal model is improved, and the risk assessment result obtained according to the federal model is more accurate.
Further, based on the first, second and/or third embodiments, a third embodiment of the enterprise risk assessment method according to the present invention is provided, in this embodiment, the method further includes:
s80, acquiring operation data when detecting the modification or deletion operation of the federal model;
in this embodiment, the banking service platform may further include a blockchain node, and the government service platform may also include a blockchain node. When the bank service platform detects the modification or deletion operation of the federal model, the operation data can be acquired. The operation data may include time of modification or deletion operation, data object, operator, and other data.
And step S90, calling a block chain node of the bank service platform to upload the operation data to a block chain, wherein the block chain is used for recording the operation data of the bank service platform and the government service platform on the federal model.
And the bank service platform calls the block chain link to upload the operation data to the block chain. And recording the operation data of the bank service platform and the government service platform on the federal model by the blockchain. And the government service platform also acquires operation data when detecting the modification or deletion operation of the federal model of the government service platform, and calls the block chain link point of the government service platform to upload the operation data to the block chain. The method for operating the data upload block chain node may refer to the existing block chain data upload method, which is not described in detail in this embodiment.
In this embodiment, block chain nodes are arranged on a bank service platform and a government service platform, and when detecting that the federal model of the self party is modified or deleted, operation data is uploaded to the block chain, so that the bank service platform and the government service platform can have complete behavior data, the traceability after verification supervision and problems are conveniently traced, the bank and the government are connected through the block chain nodes to form an independent digital account book, the situation that the partner tampers with data and improperly uses the data can be effectively avoided, and the authenticity and traceability of the data are guaranteed.
In one embodiment, as shown in fig. 3, a schematic diagram of an enterprise risk assessment architecture is shown. The bank service platform comprises three parts: a federated learning service node a, a block chain node a and a mirror database a. The government service platform also includes three parts: federated learning service node b, block chain node b and mirror database b. The federated learning service node is an architecture core component and comprises functions of data joint modeling, data authority auditing, federated learning calculation and the like; the block chain node is used for chaining the user behavior data to facilitate supervision; the mirror database is used for storing local data. The "comet" service platform of the head office is the bank service platform in each embodiment, and the "comet" service platform of the government is the government service platform in each embodiment.
In addition, an embodiment of the present invention further provides an enterprise risk assessment apparatus, where the enterprise risk assessment apparatus is deployed on a banking service platform, and referring to fig. 4, the enterprise risk assessment apparatus includes:
the acquiring module 10 is configured to acquire first enterprise data of a target enterprise that proposes a business application in the banking platform;
and the federal calculation module 20 is configured to perform federal calculation on the first enterprise data and second enterprise data by using a federal model obtained through pre-training in a combined government service platform to obtain a risk assessment result of the target enterprise, where the second enterprise data is enterprise data of the target enterprise in the government service platform.
Further, the apparatus further comprises:
the first receiving module is used for receiving the business application sent by the branch business system, wherein the branch business system sends the business application to the bank service platform after receiving the business application triggered by a target enterprise based on an enterprise client.
Further, the apparatus further comprises:
and the first sending module is used for sending the risk evaluation result to the branch business system so that the branch business system can screen the target enterprise based on the risk evaluation result to obtain the application result of the business application.
Further, the apparatus further comprises:
the second receiving module is used for receiving government agency data fields sent by the branch service system;
the second sending module is used for sending a modeling application and the government agency data fields to the government service platform so that the government service platform can check the modeling application and the government agency data fields;
and the federal modeling module is used for combining the government service platform to carry out federal modeling to obtain the federal model after the audit of the government service platform is passed.
Further, the federated modeling module includes:
the acquisition unit is used for calling a federal learning service node of the bank service platform to acquire first enterprise sample data in a local database;
the alignment unit is used for aligning the first enterprise sample data with the government service platform to obtain a first training sample, wherein the government service platform adopts second enterprise sample data corresponding to the government agency data field in the other-end database to participate in the alignment of the encrypted sample, so as to obtain a second training sample aligned with the first training sample in a sample dimension;
and the modeling unit is used for carrying out federal modeling by adopting the first training sample and the government service platform to obtain the federal model, wherein the government service platform adopts the second training sample to participate in the federal modeling.
Further, the modeling unit includes:
the receiving subunit is configured to receive a first encrypted intermediate result sent by the government service platform, where the government service platform obtains the first intermediate result by computing based on the second training sample, and encrypts the first intermediate result by using a public key received from a third party platform to obtain the first encrypted intermediate result;
the calculation subunit is configured to call the federal learning service node to calculate a first encrypted gradient value and a second encrypted intermediate result based on the first encrypted intermediate result and the first training sample, send the first encrypted gradient value to the third-party platform to be decrypted to obtain a first gradient value, and update the federal model to be trained at the local end with the first gradient value;
the sending subunit is configured to call the federal learning service node to send the second encrypted intermediate result to the government service platform, so that the government service platform calculates a second encrypted gradient value according to the second encrypted intermediate result, sends the second encrypted gradient value to the third-party platform to decrypt the second encrypted gradient value to obtain a second gradient value, and updates the federal model to be trained at the other end by using the second gradient value;
and the determining subunit is used for calling the federal learning service node to obtain the federal model based on the updated state of the federal model to be trained at the home terminal.
Further, the obtaining module 10 is further configured to obtain operation data when a modification or deletion operation on the federated model is detected;
the device further comprises:
and the uploading module is used for calling a block chain link point of the bank service platform to upload the operation data to a block chain, wherein the block chain is used for recording the operation data of the bank service platform and the government service platform on the federal model.
The expansion content of the specific implementation of the enterprise risk assessment device of the present invention is basically the same as that of each embodiment of the enterprise risk assessment method, and is not described herein again.
In addition, an embodiment of the present invention further provides a computer-readable storage medium, where an enterprise risk assessment program is stored on the storage medium, and when being executed by a processor, the enterprise risk assessment program implements the steps of the enterprise risk assessment method as described below.
The embodiments of the enterprise risk assessment apparatus and the computer-readable storage medium of the present invention may refer to the embodiments of the enterprise risk assessment method of the present invention, and are not described herein again.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one of 8230, and" comprising 8230does not exclude the presence of additional like elements in a process, method, article, or apparatus comprising the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the description of the foregoing embodiments, it is clear to those skilled in the art that the method of the foregoing embodiments may be implemented by software plus a necessary general hardware platform, and certainly may also be implemented by hardware, but in many cases, the former is a better implementation. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention, and all equivalent structures or equivalent processes performed by the present invention or directly or indirectly applied to other related technical fields are also included in the scope of the present invention.
Claims (9)
1. An enterprise risk assessment method is applied to a bank service platform, and comprises the following steps:
when a business application is received, determining a data field and a federal model corresponding to the business application, wherein the federal model is a federal model obtained by pre-training and aiming at the business application, and the bank service platform and a government service platform jointly use data of different data fields to carry out multi-time federal modeling to obtain a federal model aiming at different business application projects;
acquiring first enterprise data of a target enterprise which proposes the business application in the bank service platform;
performing federal calculation on the first enterprise data and second enterprise data by adopting the federal model through a united government service platform to obtain a risk assessment result of the target enterprise, wherein the second enterprise data is enterprise data of the target enterprise in the government service platform;
before the step of obtaining the first enterprise data of the target enterprise which proposes the business application in the banking service platform, the method further includes:
receiving government agency data fields sent by a branch business system, wherein branch business personnel select the government agency data fields required by modeling in a management menu provided by the branch business system based on the requirements of business application projects, and after receiving the government agency data fields selected by the business personnel, the branch business system sends the selected government agency data fields to the bank service platform; or business personnel of each branch log in a management menu provided by a head office according to the authority, government agency data fields required by modeling are selected in the management menu, and the bank service platform receives the government agency data fields selected by the branch business personnel in the management menu;
sending a modeling application and the government agency data field to the government service platform so that the government service platform can check the modeling application and the government agency data field, wherein the government service platform checks whether the government agency data field belongs to a preset data field allowed to be used, if so, the checking is passed, and if not, the checking is failed;
and after the fact that the audit of the government service platform is passed is detected, the federal model is obtained by combining the government service platform and performing federal modeling.
2. The enterprise risk assessment method of claim 1, wherein the step of obtaining the first enterprise data of the target enterprise that proposed the business application in the banking platform further comprises:
and receiving the business application sent by the branch business system, wherein the branch business system sends the business application to the bank service platform after receiving the business application triggered by the target enterprise based on the enterprise client.
3. The enterprise risk assessment method of claim 2, wherein the federated government service platform federately calculates the first enterprise data and the second enterprise data using the federated model, and after the step of obtaining the target enterprise risk assessment result, further comprises:
and sending the risk evaluation result to the branch business system so that the branch business system screens the target enterprise based on the risk evaluation result to obtain an application result of the business application.
4. The enterprise risk assessment method of claim 1, wherein the step of federately modeling the government service platform to obtain the federal model comprises:
calling a federal learning service node of the bank service platform to obtain first enterprise sample data in a local database;
performing encryption sample alignment on the first enterprise sample data and the government service platform to obtain a first training sample, wherein the government service platform adopts second enterprise sample data corresponding to the government agency data field in the other-end database to participate in the encryption sample alignment to obtain a second training sample aligned with the first training sample in a sample dimension;
and performing federal modeling by combining the first training sample and the government service platform to obtain the federal model, wherein the government service platform adopts the second training sample to participate in the federal modeling.
5. The enterprise risk assessment method of claim 4, wherein the federal model is derived from the federal modeling using the first training sample in conjunction with the government service platform, wherein the step of the government service platform using the second training sample to participate in federal modeling comprises:
receiving a first encrypted intermediate result sent by the government service platform, wherein the government service platform calculates to obtain a first intermediate result based on the second training sample, and encrypts the first intermediate result by using a public key received from a third party platform to obtain the first encrypted intermediate result;
calling the federated learning service node to calculate a first encrypted gradient value and a second encrypted intermediate result based on the first encrypted intermediate result and the first training sample, sending the first encrypted gradient value to the third-party platform for decryption to obtain a first gradient value, and updating the local to-be-trained federated model by adopting the first gradient value;
calling the federal learning service node to send the second encrypted intermediate result to the government service platform so that the government service platform can calculate according to the second encrypted intermediate result to obtain a second encrypted gradient value, sending the second encrypted gradient value to the third party platform to decrypt to obtain a second gradient value, and updating the federal model to be trained at the other end by adopting the second gradient value;
and calling the federal learning service node to obtain the federal model based on the updated local-end model to be trained.
6. The enterprise risk assessment method of any one of claims 1-5, wherein the federated government service platform federately calculates the first enterprise data and the second enterprise data using the federated model to obtain the target enterprise risk assessment result, after the step of obtaining the target enterprise risk assessment result, further comprises:
when detecting the modification or deletion operation of the federal model, acquiring operation data;
and calling a block chain link point of the bank service platform to upload the operation data to a block chain, wherein the block chain is used for recording the operation data of the bank service platform and the government service platform on the federal model.
7. An enterprise risk assessment device deployed on a banking platform, the enterprise risk assessment device comprising:
the bank service platform and government united service platform adopts data of different data fields to carry out multi-time federal modeling to obtain federal models for different service application projects;
the acquisition module is used for acquiring first enterprise data of a target enterprise which provides the business application in the bank service platform;
the federal calculation module is used for performing federal calculation on the first enterprise data and second enterprise data by adopting the federal model through a united government service platform to obtain a risk assessment result of the target enterprise, wherein the second enterprise data is enterprise data of the target enterprise in the government service platform;
the enterprise risk assessment device further comprises:
the second receiving module is used for receiving government agency data fields sent by a branch business system, wherein branch business personnel select the government agency data fields required by modeling in a management menu provided by the branch business system based on the requirements of business application projects, and the branch business system sends the selected government agency data fields to the bank service platform after receiving the government agency data fields selected by the business personnel; or business personnel of each branch log in a management menu provided by a head office according to the authority, government agency data fields required by modeling are selected in the management menu, and the bank service platform receives the government agency data fields selected by the branch business personnel in the management menu;
the second sending module is used for sending a modeling application and the government agency data field to the government service platform so that the government service platform can check the modeling application and the government agency data field, wherein the government service platform checks whether the government agency data field belongs to a preset data field allowed to be used, if so, the checking is passed, and if not, the checking is failed;
and the federal modeling module is used for combining the government service platform to carry out federal modeling to obtain the federal model after the audit of the government service platform is passed.
8. An enterprise risk assessment device, comprising: a memory, a processor, and an enterprise risk assessment program stored on the memory and executable on the processor, the enterprise risk assessment program when executed by the processor implementing the steps of the enterprise risk assessment method of any one of claims 1-6.
9. A computer readable storage medium, having stored thereon an enterprise risk assessment program which, when executed by a processor, implements the steps of the enterprise risk assessment method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110128783.1A CN113011632B (en) | 2021-01-29 | 2021-01-29 | Enterprise risk assessment method, device, equipment and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110128783.1A CN113011632B (en) | 2021-01-29 | 2021-01-29 | Enterprise risk assessment method, device, equipment and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113011632A CN113011632A (en) | 2021-06-22 |
| CN113011632B true CN113011632B (en) | 2023-04-07 |
Family
ID=76385424
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110128783.1A Active CN113011632B (en) | 2021-01-29 | 2021-01-29 | Enterprise risk assessment method, device, equipment and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113011632B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113887741B (en) * | 2021-11-05 | 2022-09-30 | 深圳市电子商务安全证书管理有限公司 | Data generation method, device, equipment and storage medium based on federal learning |
| CN114205155B (en) * | 2021-12-07 | 2023-09-15 | 四川启睿克科技有限公司 | Provider risk assessment system and method based on secure multiparty calculation |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111079182A (en) * | 2019-12-18 | 2020-04-28 | 北京百度网讯科技有限公司 | Data processing method, device, equipment and storage medium |
| CN112200713A (en) * | 2020-10-28 | 2021-01-08 | 支付宝(杭州)信息技术有限公司 | Business data processing method, device and equipment in federated learning |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10140277B2 (en) * | 2016-07-15 | 2018-11-27 | Intuit Inc. | System and method for selecting data sample groups for machine learning of context of data fields for various document types and/or for test data generation for quality assurance systems |
| CN110956275B (en) * | 2019-11-27 | 2021-04-02 | 支付宝(杭州)信息技术有限公司 | Risk prediction and risk prediction model training method and device and electronic equipment |
| CN111461874A (en) * | 2020-04-13 | 2020-07-28 | 浙江大学 | Credit risk control system and method based on federal mode |
-
2021
- 2021-01-29 CN CN202110128783.1A patent/CN113011632B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111079182A (en) * | 2019-12-18 | 2020-04-28 | 北京百度网讯科技有限公司 | Data processing method, device, equipment and storage medium |
| CN112200713A (en) * | 2020-10-28 | 2021-01-08 | 支付宝(杭州)信息技术有限公司 | Business data processing method, device and equipment in federated learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113011632A (en) | 2021-06-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11934540B2 (en) | System and method for multiparty secure computing platform | |
| CN110377239B (en) | Data signature method, device, server, system and storage medium | |
| US10142113B2 (en) | Identifying and maintaining secure communications | |
| JP6514218B2 (en) | Client authentication using social data | |
| US10491593B2 (en) | Multi-dimensional framework for defining criteria that indicate when authentication should be revoked | |
| US20210406386A1 (en) | System and method for multiparty secure computing platform | |
| CN107332844B (en) | Private information using method and personal credit assessment scoring method | |
| CN110462658A (en) | For providing system and method for the digital identity record to verify the identity of user | |
| KR102065993B1 (en) | Systems and methods to verify ownership of a telephone number and to track ownership reassignments | |
| CN113011632B (en) | Enterprise risk assessment method, device, equipment and computer readable storage medium | |
| CN110912712A (en) | Service operation risk authentication method and system based on block chain | |
| CN110633963A (en) | Electronic bill processing method, electronic bill processing device, computer-readable storage medium and equipment | |
| CN113240524A (en) | Method and device for detecting abnormality of account in federal learning system and electronic equipment | |
| CN105763547A (en) | Third-party authorization method and third-party authorization system | |
| CN108256814A (en) | Item information processing method, device, server and storage medium | |
| CN110837653A (en) | Label prediction method, device and computer readable storage medium | |
| WO2022216317A1 (en) | Apparatus, systems and methods for storing user profile data on a distributed database for anonymous verification | |
| CN109754322A (en) | A kind of data service system | |
| CN112766896A (en) | Electronic contract signing system based on Internet | |
| CN111383113A (en) | Suspicious client prediction method, device, equipment and readable storage medium | |
| CN112600830B (en) | Service data processing method and device, electronic equipment and storage medium | |
| CN104618306A (en) | Novel integrated mobile phone data transmission key signal encryption method | |
| CN114358767A (en) | Data transaction flow compliance notarization method and device, electronic equipment and storage medium | |
| CN110971511B (en) | Instant messaging processing method, device, equipment and computer readable storage medium | |
| CN114154965A (en) | Collaborative law enforcement and case handling management method and system based on block chain |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |