CN114969073A

CN114969073A - Wind control method and device based on block chain

Info

Publication number: CN114969073A
Application number: CN202210635263.4A
Authority: CN
Inventors: 孙宝林
Original assignee: Ant Blockchain Technology Shanghai Co Ltd
Current assignee: Ant Blockchain Technology Shanghai Co Ltd
Priority date: 2022-06-07
Filing date: 2022-06-07
Publication date: 2022-08-30

Abstract

One or more embodiments of the specification disclose a block chain-based wind control method and apparatus. The method comprises the following steps: whether missing data exists in a first data source which is provided by a first data provider and aims at a plurality of wind control objects is detected, and the first data provider is one or more of a plurality of data providers which are accessed to a block chain service platform. If so, acquiring associated data of data sources respectively provided by each data provider from the block chain service platform, and determining a predicted value corresponding to the missing data according to the associated data, wherein the associated data comprises correlation information among the data sources. Filling the missing data by using the predicted value to obtain a target data source corresponding to the first data source; and performing risk assessment on each wind control object according to the target data source, other data sources and a preset wind control model.

Description

Wind control method and device based on block chain

Technical Field

The present disclosure relates to the field of block chain technologies, and in particular, to a method and an apparatus for wind control based on a block chain.

Background

The Blockchain (Blockchain) is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism, an encryption algorithm and the like. Blockchains are an important concept of bitcoin, which is essentially a decentralized database. In the block chain system, data blocks are combined into a chain data structure in a sequential connection mode according to a time sequence, and a distributed account book which is not falsifiable and counterfeitable is ensured in a cryptographic mode. Because the block chain has the characteristics of decentralization, information non-tamper property, autonomy, independence and the like, the block chain is more and more valued and applied by people.

In the field of uplink wind control, since a blockchain can safely, privately and reliably access different data providers, the range of data is extremely wide, which is both an advantage and a burden. The reason is that the covered data sources are wide, although the breadth of the feature pool can be effectively provided, and the features which are really meaningful are mined, the stability of the wind control model is a great challenge, and the instability, fluctuation or absence of any data source can cause great fluctuation of the final model score, so that downstream business is influenced. Therefore, how to improve performance indexes and stability of on-chain wind control under the condition of accessing a plurality of data providers is a very key technology.

Disclosure of Invention

In one aspect, one or more embodiments of the present specification provide a method for block chain-based wind control, including: whether missing data exists in a first data source which is provided by a first data provider and aims at a plurality of wind control objects is detected, and the first data provider is one or more of a plurality of data providers which are accessed to a block chain service platform. If so, acquiring associated data of data sources respectively provided by each data provider from the block chain service platform, and determining a predicted value corresponding to the missing data according to the associated data, wherein the associated data comprises correlation information among the data sources. Filling the missing data by using the predicted value to obtain a target data source corresponding to the first data source; and performing risk assessment on each wind control object according to the target data source, other data sources and a preset wind control model.

In another aspect, one or more embodiments of the present specification provide a method for block chain-based wind control, including: whether new data exists in a third data source which is provided by a third data provider and aims at a plurality of wind control objects is detected, and the third data provider is one or more of a plurality of data providers which are accessed to a block chain service platform. If yes, updating a sample data set corresponding to a first wind control model stored in the block chain service platform according to the newly added data, wherein the sample data set comprises the newly added data, and the first wind control model is used for risk assessment of the wind control object. And updating the first wind control model according to the sample data set to obtain a second wind control model. And performing the risk assessment on each target wind control object by using the second wind control model.

In another aspect, one or more embodiments of the present specification provide a wind control device based on a block chain, including: the system comprises a first detection module and a second detection module, wherein the first detection module is used for detecting whether missing data exists in a first data source which is provided by a first data provider and aims at a plurality of wind control objects, and the first data provider is one or more of a plurality of data providers accessing to a block chain service platform. And if so, acquiring associated data of data sources respectively provided by the data providers from the blockchain service platform, and determining a predicted value corresponding to the missing data according to the associated data, wherein the associated data comprises correlation information among the data sources. And the filling module is used for filling the missing data by using the predicted value to obtain a target data source corresponding to the first data source. And the first wind control module is used for carrying out risk assessment on each wind control object according to the target data source, other data sources and a preset wind control model.

In another aspect, one or more embodiments of the present specification provide a wind control device based on a block chain, including: the second detection module is used for detecting whether newly added data exist in a third data source which is provided by a third data provider and aims at the plurality of wind control objects, and the third data provider is one or more of the plurality of data providers accessing the block chain service platform. And if so, updating a sample data set corresponding to a first wind control model stored in the block chain service platform according to the newly added data, wherein the sample data set comprises the newly added data, and the first wind control model is used for risk assessment of the wind control object. And the second updating module is used for updating the first wind control model according to the sample data set to obtain a second wind control model. And the second wind control module is used for carrying out the risk assessment on each target wind control object by utilizing the second wind control model.

In yet another aspect, one or more embodiments of the present specification provide a blockchain-based wind control device, including a processor and a memory electrically connected to the processor, the memory storing a computer program, the processor being configured to call and execute the computer program from the memory to implement: whether missing data exists in a first data source, provided by a first data provider, aiming at a plurality of wind control objects is detected, wherein the first data provider is one or more of a plurality of data providers accessed by a blockchain service platform. If so, acquiring associated data of data sources respectively provided by each data provider from the block chain service platform, and determining a predicted value corresponding to the missing data according to the associated data, wherein the associated data comprises correlation information among the data sources. Filling the missing data by using the predicted value to obtain a target data source corresponding to the first data source; and performing risk assessment on each wind control object according to the target data source, other data sources and a preset wind control model.

In yet another aspect, one or more embodiments of the present specification provide a blockchain-based wind control device, including a processor and a memory electrically connected to the processor, the memory storing a computer program, the processor being configured to call and execute the computer program from the memory to implement: whether new data exists in a third data source which is provided by a third data provider and aims at a plurality of wind control objects is detected, and the third data provider is one or more of a plurality of data providers accessed by a blockchain service platform. If yes, updating a sample data set corresponding to a first wind control model stored in the block chain service platform according to the newly added data, wherein the sample data set comprises the newly added data, and the first wind control model is used for risk assessment of the wind control object. And updating the first wind control model according to the sample data set to obtain a second wind control model. And performing the risk assessment on each target wind control object by using the second wind control model.

In another aspect, the present specification provides a storage medium for storing a computer program, where the computer program is executable by a processor to implement the following processes: whether missing data exists in a first data source, provided by a first data provider, aiming at a plurality of wind control objects is detected, wherein the first data provider is one or more of a plurality of data providers accessed by a blockchain service platform. If so, acquiring associated data of data sources respectively provided by each data provider from the block chain service platform, and determining a predicted value corresponding to the missing data according to the associated data, wherein the associated data comprises correlation information among the data sources. Filling the missing data by using the predicted value to obtain a target data source corresponding to the first data source; and performing risk assessment on each wind control object according to the target data source, other data sources and a preset wind control model.

In another aspect, the present specification provides a storage medium for storing a computer program, where the computer program is executable by a processor to implement the following processes: whether new data exists in a third data source which is provided by a third data provider and aims at a plurality of wind control objects is detected, and the third data provider is one or more of a plurality of data providers accessed by a blockchain service platform. If yes, updating a sample data set corresponding to a first wind control model stored in the block chain service platform according to the newly added data, wherein the sample data set comprises the newly added data, and the first wind control model is used for risk assessment of the wind control object. And updating the first wind control model according to the sample data set to obtain a second wind control model. And performing the risk assessment on each target wind control object by using the second wind control model.

Drawings

In order to more clearly illustrate one or more embodiments or technical solutions in the prior art in the present specification, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in one or more embodiments of the present specification, and other drawings can be obtained by those skilled in the art without inventive efforts.

Fig. 1 is a schematic scene diagram of a block chain-based wind control system according to an embodiment of the present specification;

FIG. 2 is a schematic flow chart diagram of a method for blockchain-based wind control according to an embodiment of the present disclosure;

fig. 3 is a schematic diagram of correlation information between data sources in a block chain-based wind control method according to an embodiment of the present specification;

FIG. 4 is a schematic flow chart diagram of a method for blockchain-based wind control according to another embodiment of the present disclosure;

FIG. 5 is a schematic flow chart diagram of a block chain based wind control method according to a further embodiment of the present disclosure;

fig. 6 is a schematic block diagram of a block chain-based wind control device according to an embodiment of the present specification;

FIG. 7 is a schematic block diagram of a blockchain based wind control apparatus according to another embodiment of the present disclosure;

fig. 8 is a schematic block diagram of a blockchain-based wind control device according to an embodiment of the present disclosure.

Detailed Description

One or more embodiments of the present disclosure provide a block chain-based wind control method and apparatus, so as to solve the problems of poor wind control performance and poor stability of the existing chain.

In order to make those skilled in the art better understand the technical solutions in one or more embodiments of the present disclosure, the technical solutions in one or more embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in one or more embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all embodiments. All other embodiments that can be derived by a person skilled in the art from one or more of the embodiments of the present disclosure without making any creative effort shall fall within the protection scope of one or more of the embodiments of the present disclosure.

Fig. 1 is a schematic scene diagram of a block chain-based wind control system according to an embodiment of the present specification. As shown in fig. 1, a blockchain-based wind control system (hereinafter referred to as a wind control system) includes a blockchain service platform and a plurality of data providers accessing the blockchain service platform. The block chain service platform comprises a plurality of block chain nodes. Each data provider is used for providing a respective data source for the blockchain service platform, and the data source comprises wind control data corresponding to one or more wind control objects, such as credit investigation data, payment data, communication payment records and the like. And the block chain service platform carries out risk assessment on the wind control object based on the data sources provided by the data providers. Because there are many data providers accessed by the block chain service platform and the data range is wide, it is inevitable that the data source is unstable, fluctuated or missing, for example, missing data or newly added data occurs, and the wind control performance of the wind control system is reduced. In the embodiment of the application, the blockchain service platform analyzes the correlation information between the data sources provided by the data providers and fills missing data based on the correlation information, so that the condition that the wind control effect is influenced by the wind control system due to the missing data is avoided. In addition, the block chain service platform updates the wind control model based on the newly added data by updating the sample data set corresponding to the wind control model and updating the wind control model based on the updated sample data set, so that the performance of the wind control model based on the wind control process is better, and the risk assessment result is more accurate.

Fig. 2 is a schematic flowchart of a method for block chain based wind control according to an embodiment of the present disclosure, and as shown in fig. 2, the method is applicable to the block chain service platform shown in fig. 1, and includes the following steps S202 to S208:

s202, whether missing data exist in a first data source, provided by a first data provider, for a plurality of wind control objects is detected, and the first data provider is one or more of the plurality of data providers.

The first data source may have missing data, which may be missing part of data in the first data source, or missing the entire first data source. Alternatively, if a plurality of data sources provided by the data providers form a data set, and each data source is respectively used as a column of data in the data set, for a first data source provided by a first data provider, if missing data occurs, part of the data in the column where the first data source is located may be missing, or the entire column of data where the first data source is located may be missing.

And S204, if the first data source has missing data, acquiring associated data of the data sources respectively provided by the data providers from the block chain service platform, and determining a predicted value corresponding to the missing data according to the associated data, wherein the associated data comprises correlation information among the data sources.

Optionally, the associated data may include at least one of a mean value of data not missing in the first data source, data not missing adjacent to the missing data in the first data source, and the like, in addition to the correlation information between the data sources. If the associated data includes a plurality of types, the predicted value corresponding to the missing data can be comprehensively predicted according to the weight corresponding to each type of associated data.

In this embodiment, the associated data of the data source provided by each data provider is stored in the blockchain service platform. In another embodiment, the associated data of the data sources respectively provided by the data providers may be stored in other locations, such as in a third party platform, wherein the third party platform may or may not access the blockchain service platform.

And S206, filling the missing data by using the predicted value to obtain a target data source corresponding to the first data source.

In this step, the missing data is filled with the predicted value, i.e., the predicted value is filled to the position where the missing data is located. For example, if a plurality of data sources provided by data providers form a data set, each data source is a column of data in the data set, and if the position of missing data in the data set is the x-th row and the y-th column, the predicted value corresponding to the missing data can be directly filled in the x-th row and the y-th column, thereby completing the data filling operation.

And S208, performing risk assessment on each wind control object according to the target data source, other data sources and a preset wind control model.

The preset wind control model is used for risk assessment of each wind control object, and input data of the wind control model is a data source provided by at least one data provider accessed by the block chain service platform. The data source comprises wind control data corresponding to one or more wind control objects, such as credit investigation data, payment data, communication payment records and the like. The output data of the wind control model can have various forms, and the embodiment of the specification does not limit the output data. Optionally, the output data (i.e., the risk assessment result) of the wind control model is a risk assessment score of the wind control subject, and the higher the risk assessment score is, the higher the risk degree of the wind control subject is. Optionally, the output data of the wind control model is a risk level of the wind control object, such as high level, medium level, low level, and the like. Optionally, the output data of the wind control model is a probability value that the wind control object belongs to a risk user, and the higher the probability value is, the higher the risk degree of the wind control object is.

The other data sources include data sources provided by one or more other data providers of the plurality of data providers other than the first data provider. When risk assessment is performed on each wind control object according to the target data source, other data sources and the preset wind control model, the target data source and other data sources can be input into the preset wind control model for risk assessment, and therefore a risk assessment result of the wind control object is obtained.

By adopting the technical scheme of one or more embodiments of the specification, whether missing data exist in a first data source, provided by a first data provider, for a plurality of wind control objects is detected, and when the missing data exist, a predicted value corresponding to the missing data is determined according to associated data of the data sources provided by the data providers respectively, wherein the associated data comprises correlation information among the data sources, and then the missing data is filled by using the predicted value, so that a target data source corresponding to the first data source is obtained. Therefore, when the first data source has a data missing phenomenon, missing data can be predicted and filled through correlation information among the data sources, including correlation information between the first data source and other data sources, so that the first data source is processed into a complete target data source, and complete and accurate data support is provided for subsequent risk assessment. And the associated data of each data source is stored in the blockchain service platform, and based on the non-tampering property and the traceability of the data in the blockchain service platform, the accuracy of the associated data can be ensured, so that the accuracy of the target data source predicted according to the associated data is ensured, and the accuracy of risk assessment on the wind control object is ensured. Further, risk assessment is performed on the wind control object according to the target data source, other data sources and the preset wind control model obtained after filling, and because input data (including the target data source and the other data sources) of the wind control model are all complete data, the situation that the risk assessment result is inaccurate due to inaccurate model output caused by data loss can be avoided. Therefore, the accuracy of risk assessment is improved, a series of problems caused by data fluctuation to the wind control system can be solved, and the performance stability of the wind control system based on the block chain is ensured.

In one embodiment, the method may include detecting whether the first data source has missing data upon detecting a preset exception event occurring during the first data provider providing the first data source. The preset exception event may include at least one of: access interruption by the data provider, data source transmission failure, etc. Whatever the predetermined exception event, the resulting result is to cause the data source to fluctuate. For example, if the access of the first data provider is interrupted (i.e., the connection with the blockchain service platform is disconnected), the first data source may fluctuate, i.e., the wind-controlled data in the first data source suddenly changes, and it may be determined that a predetermined abnormal event occurs during the first data provider provides the first data source.

Optionally, the blockchain service platform may send connection detection information to the first data provider according to a preset detection period, where the connection detection information is used to detect whether an abnormality occurs in a process of providing data to the blockchain service platform by the first data provider. For example, the blockchain service platform sends connection detection information to the first data provider once every 24 hours, and if response information of the first data provider to the connection detection information is received within a preset time period, it can be determined that no abnormality occurs in the process of providing data to the blockchain service platform by the first data provider; if the response information of the first data provider to the connection detection information is not received within the preset time, it can be determined that the process of providing data to the block chain service platform by the first data provider is abnormal, and at this time, it is detected whether the first data source has missing data.

When a preset abnormal event occurs in the process that a first data provider provides a first data source, whether the first data source has missing data or not is detected, so that the data missing condition of the first data source caused by the preset abnormal event can be detected in time, and the missing data is filled in time when the missing data exists, so that a complete target data source is obtained. Therefore, the missing data is filled in time, the data missing problem is solved quickly and efficiently, and a series of problems caused by the missing data to the wind control system are avoided, for example, the system stops running due to data change, and more seriously, the connection between other data providers and the block chain service platform is disconnected, and the data source cannot be provided. Therefore, in the embodiment, even if a preset abnormal event occurs in the process of providing the first data source by the first data provider, the normal operation of the wind control system is not affected, and the stability of the wind control performance of the wind control system is ensured.

In one embodiment, the correlation information between the data sources includes: and correlation information between the first data source and a second data source for the wind-controlled object provided by a second data provider, wherein the second data provider is one or more other data providers in the plurality of data providers except the first data provider. Based on this, when the predicted value corresponding to the missing data is determined according to the associated data of the data source provided by each data provider, the predicted value corresponding to the missing data can be determined according to the correlation information between the first data source and the second data source.

Since the second data provider is one or more other data providers of the plurality of data providers except the first data provider, the first data source may have a correlation with some or all of the data sources provided by the other data providers. When missing data is predicted according to the relevance information between the first data source and other data sources, part or all of the relevance information can be selected from all the relevance information corresponding to the first data source for prediction.

For example, a plurality of data providers accessed by the blockchain service platform respectively provide a first data source, a second data source and a third data source for the blockchain service platform, wherein the first data source and the second data source have data correlation therebetween, and the third data source does not have data correlation therebetween. Missing data may then be predicted based on the correlation information between the first data source and the second data source.

For another example, a plurality of data providers accessed by the blockchain service platform respectively provide a first data source, a second data source and a third data source for the blockchain service platform, and the first data source respectively has data correlation with the second data source and the third data source, so that missing data can be predicted according to correlation information between the first data source and the second data source, missing data can also be predicted according to correlation information between the first data source and the third data source, and missing data can also be predicted according to correlation information between the first data source and the second data source and correlation information between the first data source and the third data source.

In one embodiment, before determining the predicted value corresponding to the missing data according to the correlation information between the first data source and the second data source, correlation analysis needs to be performed on the first data source and the second data source to obtain the correlation information between the first data source and the second data source. Optionally, the non-missing data in the first data source and the second data source are obtained first, and then the deep learning correlation model is used to perform correlation analysis on the non-missing data in the first data source and the second data source, so as to obtain correlation information between the first data source and the second data source. The data correlation between the data sources may be linear or non-linear. If the correlation is linear, the correlation can be positive correlation or negative correlation; if a non-linear relationship, the non-linear relationship may be characterized by the form of a correlation function.

Optionally, the deep learning correlation model may be stored in a blockchain service platform, and the blockchain service platform may implement correlation analysis on the first data source and the second data source in an intelligent contract manner. Specifically, a first intelligent contract deployed on a blockchain service platform is called to perform correlation analysis on the non-missing data in the first data source and the second data source by using a deep learning correlation model, so as to obtain correlation information, such as a correlation function, between the first data source and the second data source.

The deep learning correlation model can be any existing model capable of analyzing the correlation between data. The input data of the deep learning correlation model can be a plurality of groups of wind control data, and each group of wind control data comprises wind control data corresponding to the same wind control object in the first data source and the second data source. For example, the wind control object includes a and B, the wind control data in the first data source and the second data source respectively corresponding to the wind control object a is used as one group of wind control data, and the wind control data in the first data source and the second data source respectively corresponding to the wind control object B is used as the other group of wind control data. And respectively taking the multiple groups of wind control objects as model data to perform correlation analysis to obtain output data, namely correlation information between the first data source and the second data source.

Furthermore, in addition to analyzing the correlation information between the data sources by using the deep learning correlation model, the correlation information between the data sources may also be analyzed by using other existing algorithms, for example, a Pearson product difference correlation coefficient between the data sources is analyzed by using a product difference correlation analysis algorithm, and the Pearson product difference correlation coefficient can represent the correlation between the data sources. As another example, the correlation between the data sources is analyzed by calculating a consistency coefficient and a harmony coefficient (i.e., Kendall correlation coefficient) between the data sources. Since various correlation analysis algorithms are well known technologies, they are not described herein again.

In one embodiment, the correlation information between data sources may be analyzed in advance and stored in the blockchain service platform. Specifically, before determining a predicted value corresponding to missing data according to associated data of data sources provided by each data provider, correlation analysis is performed on a first data source and a second data source to obtain correlation information between the first data source and the second data source, and then provider identification information and the correlation information corresponding to the first data provider are stored in a block chain service platform in an associated manner. In this way, when the predicted value corresponding to the missing data is determined according to the correlation information between the first data source and the second data source, the correlation information stored in association with the provider identification information can be acquired from the blockchain service platform according to the provider identification information corresponding to the first data provider, and then the predicted value corresponding to the missing data is determined according to the acquired correlation information.

In this embodiment, the manner of analyzing the correlation information between the data sources in advance is similar to the analysis method listed in the above embodiment. Specifically, the non-missing data in the first data source and the second data source may be obtained first, and then the deep learning correlation model is used to perform correlation learning on the non-missing data in the first data source and the second data source, so as to obtain correlation information between the first data source and the second data source. The correlation between the data sources may be linear or non-linear. If the correlation is linear, the correlation can be positive correlation or negative correlation; if a non-linear relationship, the non-linear relationship may be characterized by the form of a correlation function.

Furthermore, in addition to analyzing the correlation information between the data sources by using the deep learning correlation model, the correlation information between the data sources may also be analyzed by using other existing algorithms, for example, a Pearson product difference correlation coefficient between the data sources is analyzed by using a product difference correlation analysis algorithm, and the Pearson product difference correlation coefficient can represent the correlation between the data sources. As another example, the correlation between the data sources is analyzed by calculating a consistency coefficient and a harmony coefficient (i.e., Kendall correlation coefficient) between the data sources. Since various correlation analysis algorithms are known technologies, they are not described herein again.

In this embodiment, the correlation information between the first data source and the second data source may be stored in association with provider identification information corresponding to the first data source and/or the second data source, where the provider identification information corresponding to the data source is provider identification information of a data provider for providing the data source, and may be any one or more of a registration name, a registration account, a unique identification code, and the like of the data provider, which can uniquely identify the identity of the data provider. The correlation information between the first data source and the second data source can also be stored in a block of the blockchain service platform, which is specially used for storing the correlation information, and the correlation information is stored in association with the provider identification information corresponding to the first data source and the provider identification information corresponding to the second data source. When the missing data in the first data source needs to be predicted by using the correlation information, the correlation information stored in association with the provider identification information can be searched and acquired from the block according to the provider identification information of the first data source.

In one embodiment, after analyzing the correlation information between the first data source and the second data source, a predicted value corresponding to the missing data in the first data source is determined based on the correlation information. Optionally, the correlation information is a correlation function, the correlation function is used for characterizing a correlation between the first data source and the second data, and the function parameter of the correlation function includes wind control data in the first data source and wind control data in the second data source. By using the wind control data in any one of the data sources as the known parameters of the correlation function, the wind control data in the other data, namely the unknown parameters of the correlation function, can be calculated by using the correlation function.

Assuming that the wind control data in the second data source is complete data, missing data exists in the first data source. When the correlation function is used for predicting the missing data, optionally, the complete wind control data in the second data source can be used as the known parameters of the correlation function, the missing data in the first data source can be used as the unknown parameters of the correlation function, and the predicted value of the missing data can be determined through function calculation.

Optionally, the target wind control data corresponding to the missing data in the second data source may also be used as a known parameter of the correlation function, and the missing data in the first data source may also be used as an unknown parameter of the correlation function, so that the predicted value of the missing data may be determined through function calculation. The target wind control data corresponding to the missing data refers to the same wind control object corresponding to both the missing data and the target wind control data. Specifically, when providing the data source, the data provider may associate the wind control data with the identification information of each wind control object and transmit the associated wind control data. Therefore, when missing data is predicted, a target wind control object corresponding to the missing data can be determined, and the target wind control object is one or more of the plurality of wind control objects. And secondly, determining target wind control data corresponding to the target wind control objects in a second data source, wherein the second data source comprises wind control data corresponding to the wind control objects. And finally, predicting the related data by taking the target wind control data as the known parameters of the related function to obtain a predicted value corresponding to the missing data.

FIG. 3 illustrates dependency information between data sources in one embodiment of the present description. As shown in fig. 3, assuming that the wind control service platform has access to 3 data providers 1, 2, and 3, if each rectangular box represents a data source provided by one data provider, the box on the left side in fig. 3 represents a data source a provided by the data provider 1, the box on the middle represents a data source B provided by the data provider 2, and the box on the right side represents a data source C provided by the data provider 3. Assume that there is missing data a1 in data source a provided by data provider 2, missing data B1 in data source B provided by data provider 2, and missing data C1 in data source C provided by data provider 3.

When the blockchain service platform detects that the missing data a1 exists in the data source a, the block chain service platform determines a predicted value of the missing data a1 according to the correlation information between the data source a and other data sources (including the data source B and/or the data source C), and then fills the missing data a1 with the predicted value of the missing data a1 to obtain a complete data source a. Illustratively, the existence of dependency information between data sources is indicated by dashed arrows in FIG. 3.

Optionally, the correlation information between the first data source and the second data source includes correlation information between the wind control data corresponding to the plurality of wind control objects, respectively. For example shown in fig. 3, assuming that the wind control data may include credit investigation data, payment data, communication payment data, and the like of the wind control object, the data provider 1 provides a credit investigation data source, which includes credit investigation data corresponding to a plurality of wind control objects; the data provider 2 provides a payment data source which comprises payment data corresponding to a plurality of wind control objects; and the communication payment data source provided by the data provider 3 comprises communication payment data corresponding to a plurality of wind control objects. Therefore, the correlation information between the data source a and the data source B includes correlation information between credit investigation data and payment data corresponding to each wind control object, and the correlation information between the data source a and the data source C includes correlation information between credit investigation data and communication payment data corresponding to each wind control object. For each wind control object, the credit investigation data, the payment data, the communication payment data and the like corresponding to the wind control object have a certain degree of correlation, and the missing data corresponding to the wind control object is predicted based on the correlation, so that the predicted value of the missing data is more accurate. For example, if the missing data in the data source a is credit investigation data of the wind control object x, the credit investigation data of the wind control object x can be predicted by using the correlation information between the credit investigation data, the payment data and the communication payment data of the wind control object x.

In this embodiment, when it is detected that the first data source has missing data, the missing data can be predicted according to correlation information between the first data source and other data sources, and because in the risk assessment field, the correlation between the wind control data provided by different data providers is in a certain degree, the missing data can be predicted by using the correlation, so that a series of problems caused by data missing to the wind control system, such as system shutdown caused by data change, are avoided, and stability of the wind control performance of the wind control system is ensured.

In an embodiment, when determining the predicted value corresponding to the missing data according to the associated data of the data sources respectively provided by each data provider, if the associated data includes a mean value of the data that is not missing in the first data source, the predicted value corresponding to the missing data may be determined according to the mean value by calculating the mean value of the data that is not missing in each first data source. Optionally, the calculated mean and other associated data are combined to synthetically predict missing data, or the calculated mean is directly determined as missing data.

If the associated data comprises non-missing data adjacent to the missing data in the first data source, determining the non-missing data adjacent to the missing data in the first data source, and further determining a predicted value corresponding to the missing data according to the adjacent non-missing data. Alternatively, the missing data is comprehensively predicted by combining the non-missing data adjacent to the missing data with other associated data, or the non-missing data adjacent to the missing data is directly determined as the missing data.

If the missing data in the first data source is predicted according to the multiple kinds of associated data, the correlation information between the first data source and the second data source, the mean value of the data that is not missing in the first data source, and the prediction weight values corresponding to the data that is not missing and adjacent to the missing data in the first data source may be determined. And determining a predicted value corresponding to the missing data according to the predicted weight value, the correlation information, the average value and the non-missing data adjacent to the missing data.

The present embodiment does not limit the magnitude of the predicted weight value, and optionally, the predicted weight value corresponding to the relevance information may be set as a higher weight value, and the mean value and the predicted weight value corresponding to the non-missing data adjacent to the missing data are set as a lower weight value, so as to highlight the important role of the relevance information on the stability of the missing data.

In the embodiment, the missing data is comprehensively predicted through various associated data, so that the prediction result of the missing data is more accurate, the accuracy of a data source for risk assessment is improved, and the wind control performance of the wind control system is further improved.

Fig. 4 is a schematic flowchart of a method for block chain based wind control according to another embodiment of the present disclosure, and as shown in fig. 4, the method is applied to the block chain service platform shown in fig. 1, and includes the following steps S401 to S407:

s401, when a preset abnormal event is detected to occur in the process that a first data provider provides a first data source, whether the first data source has missing data or not is detected.

The preset abnormal event may include at least one of the following: access interruption by the data provider, data source transmission failure, etc. Whatever the predetermined exception event, the resulting result is a fluctuation in the data source. For example, if the access of the first data provider is interrupted (i.e., the connection with the blockchain service platform is disconnected), the first data source may fluctuate, i.e., the wind-controlled data in the first data source suddenly changes, and it may be determined that a predetermined abnormal event occurs during the first data provider provides the first data source.

S402, if the missing data exists, acquiring the un-missing data in the first data source and a second data source provided by a second data provider.

The second data provider is one or more other data providers except the first data provider in the plurality of data providers accessed in the blockchain service platform.

And S403, analyzing a correlation function between the first data source and the second data source according to the non-missing data in the first data source and the second data source provided by the second data provider.

The correlation function is used for characterizing correlation information between the first data source and the second data source, and the correlation information can be analyzed by using any existing deep learning correlation model or correlation analysis algorithm. The correlation information may be a positive correlation or a negative correlation, or may be a nonlinear correlation.

S404, determining a target wind control object corresponding to the missing data, and determining target wind control data corresponding to the target wind control object in the second data source.

Optionally, the data provider may associate the wind control data with the identification information of each wind control object to transmit the wind control data when providing the data source. Therefore, the target wind control data corresponding to the target wind control object in the second data source can be found out based on the identification information of the target wind control object.

S405, the target wind control data is used as the known parameters of the correlation function to predict the correlation data, and a predicted value corresponding to the missing data is obtained.

S406, filling missing data by using the predicted value to obtain a target data source corresponding to the first data source.

And S407, performing risk assessment on each wind control object according to the target data source, other data sources and a preset wind control model.

It can be seen that, with the technical solution provided in this embodiment, by detecting whether a first data source provided by a first data provider for a plurality of wind control objects has missing data, and when the missing data exists, analyzing a correlation function between the first data source and a second data source, determining a predicted value corresponding to the missing data based on the correlation function and the second data source, and filling the missing data with the predicted value, a target data source corresponding to the first data source is obtained. Therefore, when the first data source has a data missing phenomenon, missing data can be predicted and filled through correlation information (represented by a correlation function in the embodiment) among the data sources, so that the first data source is processed into a complete target data source, and complete and accurate data support is provided for subsequent risk assessment. Further, risk assessment is performed on the wind control object according to the target data source, other data sources and the preset wind control model obtained after filling, and because input data (including the target data source and the other data sources) of the wind control model are all complete data, the situation that the risk assessment result is inaccurate due to inaccurate model output caused by data loss can be avoided. Therefore, the accuracy of risk assessment is improved, a series of problems caused by data fluctuation to the wind control system can be solved, and the performance stability of the wind control system based on the block chain is ensured.

Fig. 5 is a schematic flowchart of a method for blockchain-based wind control according to still another embodiment of the present disclosure, as shown in fig. 5, the method is applicable to the blockchain service platform shown in fig. 1, and includes the following steps S502-S508:

and S502, detecting whether a third data source provided by a third data provider and aiming at a plurality of wind control objects exists new data, wherein the third data provider is one or more of the plurality of data providers.

The third data source has new data, which may be a new part of data in the third data source or a new part of the entire third data source. Optionally, if the data sources provided by the multiple data providers form a data set, and each data source is respectively used as a column of data in the data set, for a third data source provided by a third data provider, if new data occurs, the third data source may be a part of the new data added in the column where the third data source is located, or an entire column of data where the third data source is located is added. For the case that the whole third data source is newly added, the third data provider may be newly accessed by the blockchain service platform. That is, when the blockchain service platform newly accesses the third data provider, the blockchain service platform detects that new data exists in a third data source provided by the third data provider and directed to the plurality of wind control objects, and the new data is the entire third data source at this time.

And S504, if the newly added data exist, updating a sample data set corresponding to a first wind control model stored in the block chain service platform according to the newly added data, wherein the sample data set comprises the newly added data, and the first wind control model is used for risk assessment of the wind control object.

The input data of the first wind control model is a data source provided by at least one data provider accessed by the blockchain service platform. The data source comprises wind control data corresponding to one or more wind control objects, such as credit investigation data, payment data, communication payment records and the like. The output data of the first wind control model may have various forms, and the embodiment of the present specification does not limit this. Optionally, the output data (i.e., the risk assessment result) of the first wind control model is a risk assessment score of the wind control subject, and the higher the risk assessment score is, the higher the risk degree of the wind control subject is. Optionally, the output data of the first wind control model is a risk level of the wind control object, such as high level, medium level, low level, and the like. Optionally, the output data of the first wind control model is a probability value that the wind control object belongs to a risk user, and the higher the probability value is, the higher the risk degree of the wind control object is.

In this step, the newly added data may be stored in a preset storage space in the blockchain service platform, where the preset storage space is used to store a sample data set, and the sample data set includes historical sample data and/or newly added data corresponding to each data provider. Optionally, the newly added data may be directly supplemented into the original sample data set, that is, the sample data set includes the historical sample data and the newly added data. Or, two different sample data sets may be separately created, and each sample data set is used for storing the historical sample data and the newly added data, so that the newly added data is stored in the corresponding sample data set, and the newly added data and the historical sample data are separately stored in different sample data sets.

In this embodiment, the sample data set corresponding to the first wind control model is stored in the block chain service platform. In another embodiment, the sample data set may be stored elsewhere, such as in a third party platform, where the third party platform may or may not have access to the blockchain service platform.

And S506, updating the first wind control model according to the sample data set to obtain a second wind control model.

In one embodiment, when the first wind control model is updated according to the sample data set, the following steps may be performed: and retraining the model training according to the sample data set and the model training parameters corresponding to the first wind control model to obtain a second wind control model. When the model is retrained, the second wind control model and the first wind control model should use the same network structure.

In one embodiment, when the first wind control model is updated according to the sample data set, the updated sample data set can be directly utilized and updated on the basis of the original first wind control model, that is, the first wind control model is adjusted by utilizing new sample data to optimize the wind control performance of the first wind control model, and the adjusted model is the second wind control model.

And S508, performing risk assessment on each target wind control object by using the second wind control model.

And the input data and the output data of the second wind control model and the first wind control model are the same. Specifically, the input data of the second wind control model is a data source provided by at least one data provider accessed by the blockchain service platform. The data source comprises wind control data corresponding to one or more wind control objects, such as credit investigation data, payment data, communication payment records and the like. The output data of the first wind control model may have various forms, and the embodiment of the present specification does not limit this. Optionally, the output data (i.e., the risk assessment result) of the second wind control model is a risk assessment score of the wind control subject, and the higher the risk assessment score is, the higher the risk degree of the wind control subject is. Optionally, the output data of the second wind control model is a risk level of the wind control object, such as high level, medium level, low level, etc. Optionally, the output data of the second wind control model is a probability value that the wind control object belongs to the risk user, and the higher the probability value is, the higher the risk degree of the wind control object is.

By adopting the technical scheme of one or more embodiments of the specification, whether new data exist in a third data source provided by a third data provider and aiming at a plurality of wind control objects is detected, and when the new data exist, a sample data set corresponding to a first wind control model is updated according to the new data, and then the first wind control model is updated according to the sample data set, so that a second wind control model is obtained. And performing risk assessment on each target wind control object according to the updated second wind control model, wherein the first wind control model and the second wind control model are both used for performing risk assessment on each wind control object. Therefore, the sample data set corresponding to the first wind control model is stored in the block chain service platform and is based on the non-tampering property of the data in the block chain service platform, so that the accuracy and the safety of the sample data set can be ensured, and the accuracy of updating the wind control model according to the sample data set and the newly added data is ensured. And when a data adding phenomenon occurs in the third data source, the wind control model is updated on line in real time, so that the wind control model can be matched with the updated sample data set in real time, the problem of high delay caused by offline training, updating and online models is solved, the accuracy of a risk evaluation result of the wind control model is ensured, and the wind control model has better performance. In addition, a series of problems caused by data fluctuation to the wind control system can be solved, and the performance stability of the wind control system is ensured.

In one embodiment, whether the third data source has new data or not may be detected when a preset abnormal event occurs during the process of providing the third data source by the third data provider. The preset exception event may include at least one of: accessing a new data provider, receiving a new data source (for example, providing the newly added wind control data of the wind control object to the block chain service platform in the case of newly adding the wind control object), and the like. Whatever the predetermined exception event, the resulting result is a fluctuation in the data source. For example, if a third data provider is newly accessed, or wind control data of a wind control object that is not found before is newly added to a third data source, fluctuation of the third data source may be caused, that is, the wind control data in the third data source suddenly changes, and it may be determined that a preset abnormal event occurs in a process of providing the third data source by the third data provider.

Optionally, the blockchain service platform may send connection detection information to the third data provider according to a preset detection period, where the connection detection information is used to detect whether an abnormality occurs in a process of providing data to the blockchain service platform by the third data provider. For example, the blockchain service platform sends connection detection information to the third data provider once every 24 hours, and if response information of the third data provider to the connection detection information is received within a preset time period, it can be determined that no abnormality occurs in the process of providing data to the blockchain service platform by the third data provider; if the response information of the third data provider to the connection detection information is not received within the preset time, it can be determined that the process of providing data by the third data provider to the block chain service platform is abnormal, and at this time, it is detected whether the third data source has newly added data.

When a preset abnormal event occurs in the process of providing the third data source by the third data provider, whether the third data source has new data or not is detected, so that the data new situation of the third data source caused by the preset abnormal event can be detected in time, and the new data is filled in time when the new data exists, so that a complete target data source is obtained. Therefore, the missing data is filled in time, the data missing problem is solved quickly and efficiently, and a series of problems caused by the missing data to the wind control system are avoided, for example, the system stops running due to data change, and more seriously, the connection between other data providers and the block chain service platform is disconnected, and the data source cannot be provided. Therefore, in the embodiment, even if a preset abnormal event occurs in the process of providing the first data source by the first data provider, the normal operation of the wind control system is not affected, and the stability of the wind control performance of the wind control system is ensured.

In one embodiment, before detecting whether a third data source provided by a third data provider and aiming at a plurality of wind control objects has new data, the blockchain service platform is detected to be accessed to the third data provider, and the third data source provided by the third data provider is received. Optionally, a data correlation is established between the third data source and a data source provided by at least one other data provider accessed on the blockchain service platform.

Wherein, the data sources provided by other data providers can be called other data sources. Establishing data correlations between the third data source and other data sources may be performed as: and performing relevance analysis on the third data source and other data sources to obtain relevance information between the third data source and other data sources, and establishing data relevance between the third data source and other data sources based on the relevance information. The data correlation between data sources may be linear or non-linear. If the correlation is linear, the correlation can be positive correlation or negative correlation; if a non-linear relationship, the non-linear relationship may be characterized by the form of a correlation function.

Alternatively, the correlation information may be analyzed by a deep learning correlation model, which may be any existing model capable of analyzing the correlation between data. Taking analysis of the correlation information between the first data source and the third data source as an example, the input data of the deep learning correlation model may be a plurality of groups of wind control data, and each group of wind control data includes wind control data corresponding to the same wind control object in the first data source and the third data source respectively. For example, the wind control object includes a and B, the wind control data in the first data source and the third data source respectively corresponding to the wind control object a is used as one group of wind control data, and the wind control data in the first data source and the third data source respectively corresponding to the wind control object B is used as the other group of wind control data. And respectively taking the multiple groups of wind control objects as model data to perform correlation analysis to obtain output data, namely correlation information between the first data source and the third data source.

The deep learning correlation model can be stored in a block chain service platform, and the block chain service platform can realize correlation analysis of the first data source and the third data source in an intelligent contract mode. Specifically, a second intelligent contract deployed on the blockchain service platform is called, so that a deep learning relevance model is utilized to perform relevance analysis on the first data source and the third data source, and therefore relevance information between the first data source and the second data source is obtained.

In one embodiment, the correlation information between data sources may be analyzed in advance and stored in the blockchain service platform. Optionally, the correlation information between the first data source and the third data source may be stored in association with provider identification information corresponding to the first data source and/or the third data source, where the provider identification information corresponding to the data source is provider identification information of a data provider for providing the data source, and may be any one or more of a registered name, a registered account, a unique identification code, and the like of the data provider, which can uniquely identify an identity of the data provider. The correlation information between the first data source and the third data source may also be stored in a block of the blockchain service platform, which is dedicated to storing the correlation information, and the correlation information is stored in association with the provider identification information corresponding to the first data source and the provider identification information corresponding to the third data source. In this way, according to the provider identification information corresponding to the data source, the correlation information stored in association with the provider identification information can be searched and acquired from the block.

In this embodiment, the pre-stored correlation information between the data sources may provide data basis for the blockchain service platform when detecting that a data missing phenomenon occurs (that is, when detecting that the data sources provided by one or more data providers have missing data), that is, the correlation information between the data sources is used to predict the missing data. How to predict missing data by using the correlation information between data elements has been described in detail in the above embodiments, and is not described here again.

In one embodiment, updating the first wind control model according to the sample data set may be performed as: and when the first wind control model is detected to meet the preset updating condition, acquiring a sample data set from a preset storage space, and updating the first wind control model according to the acquired sample data set.

The preset storage space is used for storing a sample data set, and the preset updating condition comprises at least one of the following items: the data volume of the newly added data is larger than or equal to a preset threshold, the number of times of updating the sample data set is larger than or equal to a preset number of times, and the service life of the first wind control model is larger than or equal to a preset duration.

For example, the preset updating condition includes that the data volume of the newly added data is greater than or equal to a preset threshold, and the preset threshold is 1 ten thousand data volumes. And when detecting that the data volume of the newly added data is more than or equal to 1 ten thousand, updating the first wind control model based on the updated sample data set.

After a second wind control model is obtained by updating the first wind control model, the second wind control model can be stored in the block chain service platform; or replacing the first wind control model stored in the blockchain service platform with the second wind control model.

In this embodiment, when it is determined that the first wind control model meets the preset update condition, the first wind control model is updated by using the updated sample data set, so that the update timing and the update condition of the first wind control model are normalized. Under the conditions that newly added data are less and cannot reach a certain magnitude, the number of times of updating the sample data set is less, or the service life of the wind control model is short, the effect of the wind control model may not be achieved by updating the wind control model, for example, the model performance of the second wind control model and the model performance of the first wind control model are not greatly different, so that not only resources are wasted when the wind control model is frequently updated, but also the model performance optimization effect brought by updating the wind control model is not obvious. Therefore, compared with a mode of updating the wind control model in real time, the method and the device can reduce the number of times of model updating, avoid unnecessary model updating and ensure that a better model optimization effect can be brought by updating the model each time.

In summary, particular embodiments of the present subject matter have been described. Other embodiments are within the scope of the following claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may be advantageous.

Based on the same idea, the wind control method based on the block chain provided in one or more embodiments of the present specification further provides a wind control device based on the block chain.

Fig. 6 is a schematic block diagram of a block chain-based wind control device according to an embodiment of the present specification. As shown in fig. 6, the apparatus includes:

the first detection module 61 is used for detecting whether missing data exists in a first data source which is provided by a first data provider and aims at a plurality of wind control objects; the first data provider is one or more of a plurality of data providers accessing a blockchain service platform;

if yes, a first determining module 62 obtains associated data of data sources respectively provided by each data provider from the blockchain service platform, and determines a predicted value corresponding to the missing data according to the associated data; the associated data comprises correlation information between the data sources;

a filling module 63, configured to fill the missing data with the predicted value to obtain a target data source corresponding to the first data source;

and the first wind control module 64 is used for carrying out risk assessment on each wind control object according to the target data source, other data sources and a preset wind control model.

In one embodiment, the relevance information includes: correlation information between the first data source and a second data source provided by a second data provider and aiming at the wind control object; the second data provider is a data provider of the plurality of data providers other than the first data provider;

the first determination module 62 includes:

and the first determining unit is used for determining a predicted value corresponding to the missing data according to the correlation information between the first data source and the second data source.

In one embodiment, the apparatus further comprises:

the analysis unit is used for performing correlation analysis on the first data source and the second data source to obtain the correlation information before determining a predicted value corresponding to the missing data according to the correlation data of the data sources respectively provided by the data providers;

the first storage unit is used for storing provider identification information corresponding to the first data provider and the correlation information in a correlation mode to the blockchain service platform;

the first determination module 62 includes:

the first acquisition unit is used for acquiring the correlation information which is stored in association with the provider identification information from the blockchain service platform according to the provider identification information corresponding to the first data provider;

and the second determining unit is used for determining the predicted value corresponding to the missing data according to the acquired correlation information.

In one embodiment, the relevance information includes a relevance function; the analysis unit:

acquiring the non-missing data in the first data source and the second data source;

and calling a first intelligent contract deployed on the blockchain service platform to perform correlation analysis on the non-missing data and the second data source by using a deep learning correlation model to obtain the correlation function between the first data source and the second data source.

In one embodiment, the first determination module 62 includes:

the third determining unit is used for determining a target wind control object corresponding to the missing data; the target wind control object is one or more of the plurality of wind control objects;

the fourth determining unit is used for determining target wind control data corresponding to the target wind control object in the second data source; the second data source comprises wind control data corresponding to each wind control object;

and the prediction unit is used for predicting the related data by taking the target wind control data as the parameter of the related function to obtain the predicted value corresponding to the missing data.

In one embodiment, the first determination module 62 includes:

the calculation unit is used for calculating the mean value of the data which are not missed in each first data source and determining the predicted value corresponding to the missed data according to the mean value; the associated data further includes the mean;

and/or the presence of a gas in the gas,

a fifth determining unit, configured to determine non-missing data adjacent to the missing data in the first data source, and determine the predicted value corresponding to the missing data according to the adjacent non-missing data; the associated data further includes the contiguous non-missing data.

In one embodiment, the first determination module 62 includes:

a sixth determining unit that determines prediction weight values corresponding to the correlation information, the mean value, and non-missing data adjacent to the missing data, respectively;

a seventh determining unit that determines the predicted value corresponding to the missing data according to the prediction weight value, the correlation information, the mean value, and the non-missing data adjacent to the missing data.

In one embodiment, the first detection module 61 comprises:

the first connection detection unit sends connection detection information to the first data provider through the block chain service platform according to a preset detection period; the connection detection information is used for detecting whether the process of providing data to the block chain service platform by the first data provider is abnormal or not;

the first detection unit detects whether the first data source has missing data or not when detecting that a preset abnormal event occurs in the process that the first data provider provides the first data source.

By adopting the device in one or more embodiments of the present specification, whether missing data exists in a first data source provided by a first data provider and directed to a plurality of wind control objects is detected, and when the missing data exists, a predicted value corresponding to the missing data is determined according to associated data of the data sources provided by each data provider, wherein the associated data includes correlation information between the data sources, and then the missing data is filled by using the predicted value, so that a target data source corresponding to the first data source is obtained. Therefore, when the first data source has a data missing phenomenon, missing data can be predicted and filled through correlation information among the data sources, including correlation information between the first data source and other data sources, so that the first data source is processed into a complete target data source, and complete and accurate data support is provided for subsequent risk assessment. And the associated data of each data source is stored in the blockchain service platform, and based on the non-tampering property and the traceability of the data in the blockchain service platform, the accuracy of the associated data can be ensured, so that the accuracy of the target data source predicted according to the associated data is ensured, and the accuracy of risk assessment on the wind control object is ensured. Further, risk assessment is performed on the wind control object according to the target data source, other data sources and the preset wind control model obtained after filling, and because input data (including the target data source and the other data sources) of the wind control model are all complete data, the situation that the risk assessment result is inaccurate due to inaccurate model output caused by data loss can be avoided. Therefore, the accuracy of risk assessment is improved, a series of problems caused by data fluctuation to the wind control system can be solved, and the performance stability of the wind control system based on the block chain is ensured.

Fig. 7 is a schematic block diagram of a block chain-based wind control device according to another embodiment of the present disclosure. As shown in fig. 7, the apparatus includes:

the second detection module 71 is configured to detect whether new data exists in a third data source, provided by a third data provider, for the multiple wind control objects; the third data provider is one or more of a plurality of data providers accessing a blockchain service platform; if yes, updating a sample data set corresponding to a first wind control model stored in the block chain service platform according to the newly added data; the sample data set comprises the newly-added data, and the first wind control model is used for risk assessment of the wind control object;

the second updating module 73 updates the first wind control model according to the sample data set to obtain a second wind control model;

and the second wind control module 74 is used for performing the risk assessment on each target wind control object by using the second wind control model.

In one embodiment, the apparatus further comprises:

the third detection module detects that the blockchain service platform accesses the third data provider before detecting whether the third data source provided by the third data provider and aiming at the plurality of wind control objects has new data;

the receiving module is used for receiving the third data source provided by the third data provider;

and the establishing module is used for establishing data correlation between the third data source and data sources provided by at least one other data provider accessed on the blockchain service platform.

In one embodiment, the first update module 72 includes:

the second storage unit is used for storing the newly added data into a preset storage space in the block chain service platform; the preset storage space is used for storing the sample data set, and the sample data set comprises historical sample data and/or newly added data corresponding to each data provider.

In one embodiment, the second updating module 73 includes:

the second acquisition unit is used for acquiring the sample data set from the preset storage space when the first wind control model is detected to meet a preset updating condition;

the updating unit is used for updating the first wind control model according to the sample data set:

wherein the preset updating condition comprises at least one of the following items: the data volume of the newly added data is greater than or equal to a preset threshold, the number of times of updating the sample data set is greater than or equal to a preset number of times, and the service life of the first wind control model is greater than or equal to a preset length of time.

In one embodiment, the second updating module 73 includes:

the training unit retrains model training according to the sample data set and model training parameters corresponding to the first wind control model to obtain the second wind control model;

the storage or replacement unit is used for storing the second wind control model in the block chain service platform; or replacing the first wind control model stored in the blockchain service platform with the second wind control model.

In one embodiment, the second detection module 71 includes:

the second connection detection unit sends connection detection information to the third data provider according to a preset detection period through the block chain service platform; the connection detection information is used for detecting whether the process of providing data to the block chain service platform by the third data provider is abnormal or not;

and the second detection unit is used for detecting whether newly added data exists in the third data source or not when a preset abnormal event occurs in the process of providing the third data source by the third data provider.

By adopting the device in one or more embodiments of the present specification, whether new data exists in a third data source provided by a third data provider and directed to a plurality of wind control objects is detected, and when the new data exists, a sample data set corresponding to the first wind control model is updated according to the new data, and then the first wind control model is updated according to the sample data set, so as to obtain a second wind control model. And performing risk assessment on each target wind control object according to the updated second wind control model, wherein the first wind control model and the second wind control model are both used for performing risk assessment on each wind control object. Therefore, the sample data set corresponding to the first wind control model is stored in the block chain service platform and is based on the non-tampering property of the data in the block chain service platform, so that the accuracy and the safety of the sample data set can be ensured, and the accuracy of updating the wind control model according to the sample data set and the newly added data is ensured. And when a data adding phenomenon occurs in the third data source, the wind control model is updated on line in real time, so that the wind control model can be matched with the updated sample data set in real time, the problem of high delay caused by offline training, updating and online models is solved, the accuracy of a risk evaluation result of the wind control model is ensured, and the wind control model has better performance. In addition, a series of problems caused by data fluctuation to the wind control system can be solved, and the performance stability of the wind control system is ensured.

It should be understood by those skilled in the art that the above-mentioned block chain based wind control device can be used to implement the above-mentioned block chain based wind control method, and the detailed description thereof should be similar to the above-mentioned method, and in order to avoid complexity, it is not repeated herein.

Based on the same idea, one or more embodiments of the present specification further provide a wind control device based on a block chain, as shown in fig. 8. The chain of blocks based wind control devices may vary significantly due to different configurations or performance and may include one or more processors 801 and memory 802, where one or more stored applications or data may be stored in memory 802. Wherein the memory 802 may be a transient storage or a persistent storage. The application program stored in memory 802 may include one or more modules (not shown), each of which may include a series of computer-executable instructions for a blockchain-based wind control device. Still further, the processor 801 may be configured to communicate with the memory 802 to execute a series of computer-executable instructions in the memory 802 on a blockchain based wind control device. The blockchain-based wind control apparatus may also include one or more power supplies 803, one or more wired or wireless network interfaces 804, one or more input-output interfaces 805, one or more keyboards 806.

In particular, in this embodiment, the wind control device based on the blockchain includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions for the wind control device based on the blockchain, and the one or more programs configured to be executed by the one or more processors include computer-executable instructions for:

detecting whether a first data source provided by a first data provider and aiming at a plurality of wind control objects has missing data or not; the first data provider is one or more of a plurality of data providers accessing a blockchain service platform;

if so, acquiring associated data of data sources respectively provided by each data provider from the block chain service platform, and determining a predicted value corresponding to the missing data according to the associated data; the associated data comprises correlation information between the data sources;

filling the missing data by using the predicted value to obtain a target data source corresponding to the first data source;

and performing risk assessment on each wind control object according to the target data source, other data sources and a preset wind control model.

In particular, in another embodiment, a blockchain based wind control device includes a memory, and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer executable instructions for the blockchain based wind control device, and the one or more programs configured to be executed by the one or more processors include computer executable instructions for:

detecting whether a third data source provided by a third data provider and aiming at a plurality of wind control objects has new data or not; the third data provider is one or more of a plurality of data providers accessing the blockchain service platform;

if yes, updating a sample data set corresponding to a first wind control model stored in the block chain service platform according to the newly added data; the sample data set comprises the newly-added data, and the first wind control model is used for risk assessment of the wind control object;

updating the first wind control model according to the sample data set to obtain a second wind control model;

and performing the risk assessment on each target wind control object by using the second wind control model.

One or more embodiments of the present specification further provide a storage medium, where the storage medium stores one or more computer programs, where the one or more computer programs include instructions, and when the instructions are executed by an electronic device including multiple application programs, the electronic device can execute each process of the above-mentioned partition chain-based wind control method embodiment, and can achieve the same technical effect, and details are not described here to avoid repetition.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the various elements may be implemented in the same one or more software and/or hardware implementations in implementing one or more embodiments of the present description.

One skilled in the art will recognize that one or more embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, one or more embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, one or more embodiments of the present description may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

One or more embodiments of the present specification are described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the specification. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that 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 an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

One or more embodiments of the present description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

The above description is only one or more embodiments of the present disclosure, and is not intended to limit the present disclosure. Various modifications and alterations to one or more embodiments described herein will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of one or more embodiments of the present specification should be included in the scope of claims of one or more embodiments of the present specification.

Claims

1. A block chain-based wind control method comprises the following steps:

2. The method of claim 1, the relevance information comprising: correlation information between the first data source and a second data source for the wind-controlled object provided by a second data provider; the second data provider is a data provider of the plurality of data providers other than the first data provider;

the determining a predicted value corresponding to the missing data according to the associated data of the data source provided by each data provider comprises:

and determining a predicted value corresponding to the missing data according to the correlation information between the first data source and the second data source.

3. The method according to claim 1, before determining the predicted value corresponding to the missing data according to the associated data of the data source provided by each of the data providers, the method further comprises:

performing correlation analysis on the first data source and the second data source to obtain the correlation information;

associating and storing provider identification information corresponding to the first data provider and the correlation information to the blockchain service platform;

the determining a predicted value corresponding to the missing data according to the correlation information between the first data source and the second data source includes:

acquiring the correlation information stored in association with the provider identification information from the blockchain service platform according to the provider identification information corresponding to the first data provider;

and determining the predicted value corresponding to the missing data according to the acquired correlation information.

4. The method of claim 3, the correlation information comprising a correlation function; the performing correlation analysis on the first data source and the second data source to obtain the correlation information includes:

5. The method of claim 4, wherein determining the predicted value corresponding to the missing data according to the correlation information between the first data source and the second data source comprises:

determining a target wind control object corresponding to the missing data; the target wind control object is one or more of the plurality of wind control objects;

determining target wind control data corresponding to the target wind control object in the second data source; the second data source comprises wind control data corresponding to each wind control object;

and predicting relevant data by taking the target wind control data as a parameter of the relevant function to obtain the predicted value corresponding to the missing data.

6. The method according to claim 1, wherein the determining a predicted value corresponding to the missing data according to the associated data of the data source provided by each of the data providers includes:

calculating the mean value of the data which are not missed in each first data source, and determining the predicted value corresponding to the missed data according to the mean value; the associated data further includes the mean;

and/or the presence of a gas in the gas,

determining non-missing data adjacent to the missing data in the first data source, and determining the predicted value corresponding to the missing data according to the adjacent non-missing data; the associated data further includes the contiguous non-missing data.

7. The method of claim 6, wherein determining a predicted value corresponding to the missing data according to the associated data of the data source provided by each data provider comprises:

determining prediction weight values respectively corresponding to the correlation information, the mean value and the non-missing data adjacent to the missing data;

and determining the predicted value corresponding to the missing data according to the predicted weight value, the correlation information, the mean value and the non-missing data adjacent to the missing data.

8. The method of claim 1, the detecting whether there is missing data in a first data source for a plurality of wind-controlled objects provided by a first data provider, comprising:

sending connection detection information to the first data provider according to a preset detection period through the block chain service platform; the connection detection information is used for detecting whether the process of providing data to the block chain service platform by the first data provider is abnormal or not;

when a preset abnormal event is detected to occur in the process that the first data provider provides the first data source, whether the first data source has missing data or not is detected.

9. A block chain-based wind control method comprises the following steps:

detecting whether a third data source provided by a third data provider and aiming at a plurality of wind control objects has new data or not; the third data provider is one or more of a plurality of data providers accessing a blockchain service platform;

10. The method of claim 9, before detecting whether new data exists in a third data source provided by a third data provider for the plurality of wind-controlled objects, further comprising:

detecting that the blockchain service platform accesses the third data provider;

receiving the third data source provided by the third data provider;

establishing data correlation between the third data source and data sources provided by at least one other data provider accessed on the blockchain service platform.

11. The method of claim 9, wherein updating the sample data set corresponding to the first wind control model stored in the blockchain service platform according to the new added data comprises:

storing the newly added data to a preset storage space in the block chain service platform; the preset storage space is used for storing the sample data set, and the sample data set comprises historical sample data and/or newly added data corresponding to each data provider.

12. The method of claim 11, said updating said first wind control model according to said sample data set, comprising:

when the first wind control model is detected to meet a preset updating condition, acquiring the sample data set from the preset storage space;

updating the first wind control model according to the sample data set:

13. The method of claim 9, said updating said first wind control model according to said sample data set, resulting in a second wind control model, comprising:

retraining model training according to the sample data set and model training parameters corresponding to the first wind control model to obtain the second wind control model;

storing the second wind control model in the blockchain service platform; or replacing the first wind control model stored in the blockchain service platform with the second wind control model.

14. The method of claim 9, wherein detecting whether new data exists in a third data source provided by a third data provider for a plurality of wind-controlled objects comprises:

sending connection detection information to the third data provider according to a preset detection period through the block chain service platform; the connection detection information is used for detecting whether the process of providing data to the block chain service platform by the third data provider is abnormal or not;

and when detecting that a preset abnormal event occurs in the process of providing the third data source by the third data provider, detecting whether the third data source has newly added data.

15. A block chain based wind control device comprising:

the first detection module is used for detecting whether missing data exist in a first data source which is provided by a first data provider and aims at a plurality of wind control objects; the first data provider is one or more of a plurality of data providers accessing a blockchain service platform;

if yes, acquiring associated data of data sources respectively provided by each data provider from the block chain service platform, and determining a predicted value corresponding to the missing data according to the associated data; the associated data comprises correlation information between the data sources;

the filling module is used for filling the missing data by using the predicted value to obtain a target data source corresponding to the first data source;

and the first wind control module is used for carrying out risk assessment on each wind control object according to the target data source, other data sources and a preset wind control model.

16. A blockchain based wind control device comprising:

the second detection module is used for detecting whether a third data source provided by a third data provider and aiming at the plurality of wind control objects has new data or not; the third data provider is one or more of a plurality of data providers accessing a blockchain service platform;

the second updating module is used for updating the first wind control model according to the sample data set to obtain a second wind control model;

and the second wind control module is used for carrying out the risk assessment on each target wind control object by utilizing the second wind control model.

17. A blockchain based wind control device comprising a processor and a memory electrically connected to the processor, the memory storing a computer program, the processor being configured to invoke and execute the computer program from the memory to implement:

detecting whether a first data source provided by a first data provider and aiming at a plurality of wind control objects has missing data or not; the first data provider is one or more of a plurality of data providers accessed by the blockchain service platform;

18. A blockchain based wind control device comprising a processor and a memory electrically connected to the processor, the memory storing a computer program, the processor being configured to invoke and execute the computer program from the memory to implement:

detecting whether a third data source provided by a third data provider and aiming at a plurality of wind control objects has new data or not; the third data provider is one or more of a plurality of data providers accessed by the blockchain service platform;

19. A storage medium storing a computer program executable by a processor to implement the following:

if yes, acquiring associated data of data sources respectively provided by the data providers from the block chain service platform, and determining a predicted value corresponding to the missing data according to the associated data; the associated data comprises correlation information between the data sources;

20. A storage medium storing a computer program executable by a processor to implement the following: