Integrated supervision platform based on distributed block chain

Abstract

The invention belongs to the technical field of data supervision, and discloses an integrated supervision platform based on a distributed block chain, and the platform comprises the steps: carrying out the identity authentication of a participant in a target supervision process, obtaining a corresponding participation type, and constructing a corresponding distributed block chain; the participation category comprises a supervision subject, a supervised object and a third-party mechanism; carrying out data acquisition on financial data related to a supervised object to obtain corresponding financial information; performing data processing on the acquired financial information to obtain corresponding data to be supervised; constructing a corresponding storage task based on the to-be-supervised data; uploading and storing the obtained to-be-supervised data into the constructed distributed block chain based on the constructed storage task; according to the method, the to-be-supervised data stored in the supervised object in the distributed block chain is looked up to obtain the corresponding lookup result, and the corresponding supervision result is generated based on the corresponding lookup result, so that the accuracy and efficiency of supervision are improved.

Description

Technical Field

[0001] The present invention relates to the field of data supervision technology, and more specifically, to an integrated supervision platform based on distributed blockchain. Background Art

[0002] The patent application with the publication number CN106777923A discloses a hospital information supervision platform and supervision method. The hospital information supervision platform includes a distributed blockchain grid constructed by several data sources and one or more data application parties as supervision nodes, and also includes a blockchain key generation module, a distributed storage module and a perception prompt module. The blockchain key generation module uses blockchain technology to generate corresponding blockchain keys for the medical key information of each supervision node in the distributed blockchain grid through an encryption algorithm; the distributed storage module stores the blockchain key in all supervision nodes of the distributed blockchain grid; each supervision node communicates with each other in the distributed blockchain grid; if the medical key information of any supervision node is modified, the corresponding blockchain key changes, and other supervision nodes will perceive and prompt. The platform realizes multi-point self-regulation and intelligent prompts, reduces the cost of medical key information security management, and can prevent improper modification of the hospital's medical key information;

[0003] Compared with the existing technology, the traditional financial supervision method faces problems such as data silos, information opacity, and low supervision efficiency. Financial institutions have accumulated a large amount of customer information, transaction records, and risk assessment data in their daily operations. These data are scattered in different systems, forming data silos; this not only increases the difficulty of data management, but also limits the value of data utilization. In addition, due to the lack of an effective data sharing mechanism, it is difficult for regulators to obtain comprehensive market information, resulting in lags and inaccuracies in regulatory decisions. Therefore, how to break the data silos and realize the real-time, accurate, and transparent sharing of financial data has become an urgent problem to be solved. In view of this, the present invention proposes a method to solve the above problems. Summary of the invention

[0004] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solutions: an integrated supervision platform based on distributed blockchain, comprising:

[0005] A blockchain construction module is used to authenticate the identity of the participants in the target regulatory process, obtain the corresponding participation category, and build a corresponding distributed blockchain based on the participation category; the participation category includes the regulatory subject, the regulated object, and the third-party institution;

[0006] The data collection module is used to collect financial data related to the regulated object and obtain corresponding financial information;

[0007] A data processing module is used to process the collected financial information to obtain corresponding data to be supervised; and to construct corresponding storage tasks based on the data to be supervised;

[0008] The data uploading module is used to upload the acquired data to be supervised to the constructed distributed blockchain based on the constructed storage task;

[0009] The data supervision module is used to review the data to be supervised stored by the supervised objects in the distributed blockchain, obtain corresponding review results, generate corresponding supervision results based on the corresponding review results, and take corresponding supervision measures.

[0010] Furthermore, the process of authenticating the participants in the target regulatory process, obtaining the corresponding participation category, and building the corresponding distributed blockchain based on the participation category includes:

[0011] A user terminal is set up to obtain the participants involved in the corresponding supervision process and mark them as participating users. The participating users log in to the user terminal and upload identity information for identity authentication. If the identity authentication fails, a login failure notification is fed back to the participating users. If the identity authentication succeeds, the user terminal obtains the identity information of the corresponding participating users for identity authentication, and divides the corresponding participating users into participation categories based on the identity information, and the participation categories include the supervision subject, the supervised object and the third-party organization; and at the same time, a corresponding identity number is generated based on the identity information of the participating users;

[0012] Furthermore, based on the distributed network, a corresponding distributed blockchain is constructed in combination with the participation categories corresponding to the corresponding participating users. The distributed blockchain consists of a number of supervision nodes, audit nodes and storage nodes.

[0013] Furthermore, the process of collecting financial data related to the regulated entity and obtaining corresponding regulatory information includes:

[0014] The data collection module is provided with a plurality of collection units, and the collection units collect financial data related to the supervised object from a plurality of data sources based on the user terminal to obtain corresponding financial information, and the financial information is composed of a plurality of financial data.

[0015] Furthermore, the process of performing data processing on the collected financial information to obtain corresponding data to be supervised; and constructing corresponding storage tasks based on the data to be supervised includes:

[0016] Reading the collected financial information, and performing data preprocessing on the collected financial information, wherein the preprocessing includes missing value processing and data standardization;

[0017] After the data preprocessing is completed, the collection time corresponding to each financial data in the corresponding financial information is obtained respectively, and the timestamp corresponding to the corresponding financial data is obtained based on the collection time, and the timestamp is converted into a binary code to obtain the corresponding time identification code;

[0018] At the same time, the financial data in the corresponding financial information is read, and characteristic fragments are extracted to obtain corresponding characteristic data fragments, and corresponding financial data identifiers are constructed based on the characteristic data fragments, and data of the corresponding financial information is encrypted to obtain corresponding ciphertext data;

[0019] The financial data identifier corresponding to the corresponding financial data and the corresponding ciphertext data are mutually associated to obtain the corresponding subgroup to be stored

[0020] Counting the identity numbers of the subgroups to be stored and the regulated objects corresponding to all the financial data in the financial information to obtain the corresponding data to be regulated;

[0021] Then, the data capacity corresponding to the corresponding data to be supervised, the financial data identifiers corresponding to each financial data and the identity identifiers of the supervised objects are obtained, and corresponding storage tasks are constructed based on them;

[0022] Furthermore, based on the user terminal, the storage corresponding to the corresponding regulated object is uploaded to the constructed distributed blockchain.

[0023] Furthermore, the process of constructing a corresponding financial data identifier based on the characteristic data segment includes:

[0024] Read the corresponding feature data segments, and assign different levels of weight to each feature data segment, wherein the specific weight depends on the specific content of the feature data segment;

[0025] Read the corresponding characteristic data segment, and randomly generate a set of dynamic data identifiers based on the corresponding characteristic data segment. The dynamic data identifier can be obtained by the following mathematical formula: In the formula, α represents the dynamic data representation corresponding to the feature data segment, which is a digital sequence composed of a group of random numbers; N i N represents the data volume of the i-th feature data segment corresponding to the corresponding financial data; max and N min Respectively represent the maximum data volume and the minimum data volume in the characteristic data segment corresponding to the corresponding financial data, wherein i=1, 2, ..., N0, N0 represents the total number of characteristic data segments;

[0026] The dynamic data identifiers of all feature data segments are read and dimension reduction is performed on them; the dynamic data identifiers after dimension reduction are accumulated according to the assigned weights to obtain the corresponding financial character strings; and a unique financial data identifier is constructed based on the financial character string and the time coding data.

[0027] Furthermore, the process of encrypting the corresponding financial information to obtain the corresponding ciphertext data includes:

[0028] Acquire characteristic information corresponding to the corresponding financial data based on the obtained characteristic data fragments;

[0029] Based on the characteristic information, a bifurcated data tree is constructed for the financial information uploaded by the corresponding regulated object to obtain a corresponding financial data tree;

[0030] Based on the CA-ABE algorithm, the corresponding initial parameter C is generated in combination with the identity information of the corresponding supervised object. The initial parameter C = {e, g, H h , Zp}; G is a cyclic group of prime order P, g is a generator of the group, e is a bilinear map; Zp is a non-negative minimal simplified residue system modulo p, p is a prime number; H h =(0,1,2,3,4,5);

[0031] Then, based on the initial parameters, the corresponding master key ZMY=g is constructed. θ And the public key GMY={e(g,g) θ ,g λ}; where θ and λ are a random function and a random number respectively, and both θ and λ are ∈ Zp;

[0032] Randomly select a random number r∈Zp, and split the corresponding financial data based on the random number r to obtain the secret shard σ_u corresponding to the leaf node in the corresponding financial data tree; and calculate the attributes of the corresponding leaf node. The corresponding mathematical formula is: Where σ_u represents the secret shard corresponding to leaf node u; and The first attribute calculation results and the second attribute calculation results of the leaf node are respectively expressed; that is, the ciphertext field of the leaf node, the ciphertext data corresponding to the corresponding financial data tree is composed of the first attribute calculation results and the second attribute calculation results of all leaf nodes; r_u represents the random number corresponding to the u-th leaf node, and the corresponding ciphertext data is obtained based on the attribute calculation.

[0033] Furthermore, the process of uploading and storing the obtained data to be supervised into the constructed distributed blockchain based on the constructed storage task includes:

[0034] After receiving the corresponding storage task, the distributed blockchain will broadcast the corresponding storage task to all storage nodes based on the deployed smart contract, read the financial data identifier in the corresponding storage task, and perform the identifier verification;

[0035] If the verification fails, indicating that the same financial data identifier exists, a storage failure notification and the reason for the failure are fed back to the user terminal, and then the user terminal can resubmit after verifying the corresponding financial data identifier;

[0036] If the verification is successful, indicating that the corresponding financial data identifier does not exist, a storage node is randomly selected as a temporary node, and other storage nodes are marked as slave nodes;

[0037] The slave node requests a node signature from the corresponding temporary node and attaches its own node signature. After the temporary node obtains the request, it records the node signature of the slave node and feeds back the node signature of the temporary node. Then, the slave node feeds back the node registration information to the temporary node. After receiving the corresponding node registration information, the temporary node returns the registered adjacent node information, otherwise it continues to wait. After the slave node receives the registered adjacent node information, it feeds back confirmation information to the temporary node. After the temporary node receives all the confirmation information, it records the corresponding temporary node as the master node and marks the slave node as the child node.

[0038] Repeat the acquisition process of the corresponding master node until all storage nodes are traversed and all marked master nodes are counted; prioritize all master nodes based on the data capacity in the corresponding storage task and the nearest minimum principle, and mark the corresponding master node as the first shared node, the second shared node, ..., the Kth shared node based on the priority sorting, where K represents the total number of master nodes; and feed back the corresponding storage notification to the user terminal;

[0039] After the user terminal receives the corresponding storage notification, the user terminal uploads the obtained data to be regulated to the corresponding first shared node for storage. The first shared node stores its own node signature and the corresponding data to be regulated in the second shared node, and so on, until it is stored in the Kth shared node; at the same time, a corresponding storage log is generated and broadcast to the corresponding distributed blockchain.

[0040] Furthermore, the process of consulting the data to be supervised stored by the supervised object in the distributed blockchain and obtaining the corresponding consulting results includes:

[0041] If the participating user is a regulatory subject, and the regulatory subject needs to review the historical financial information uploaded by a regulated object, the regulatory subject may generate a corresponding regulatory request based on the user terminal;

[0042] Feedback the corresponding supervision request to the supervision node corresponding to the corresponding supervision subject; the supervision node searches the storage node in the corresponding distributed blockchain based on the supervision request, obtains the corresponding search list, and feeds back the storage node to the corresponding user terminal. The supervision subject can check whether the corresponding search list is a storage node associated with the corresponding supervised object. If not, the supervision subject is reminded to check whether the corresponding supervision request is accurate and conduct a secondary search; if so, the corresponding supervision node feeds back a review notification;

[0043] After receiving the corresponding query notification, the supervisory node obtains the registration information and node signature of the node corresponding to each storage node in the corresponding search list and performs verification. If the verification fails, a node information abnormality notification is fed back to the corresponding supervised object based on the identity number of the supervised object stored in the corresponding storage node; if the verification passes, the corresponding supervisory node sends its own node signature to the storage node. After the storage node receives the corresponding node signature, it feeds back a confirmation notification to the supervisory node and marks the corresponding storage node as a query node;

[0044] The node to be monitored receives confirmation information from all storage nodes in the corresponding monitoring list and obtains the corresponding reference list. Then, based on the data processing process, the data to be monitored stored in the corresponding reference node is reversely processed to obtain the corresponding supervision information.

[0045] Furthermore, the supervisory subject may consult the corresponding supervisory data, and input the corresponding supervisory data as input parameters into a pre-built supervisory model to obtain the corresponding first auxiliary result;

[0046] Furthermore, the regulatory body may generate corresponding monitoring results based on the understanding after reviewing the corresponding regulatory information and the first auxiliary results.

[0047] Further, the process of building a regulatory model includes:

[0048] Obtaining a number of historical financial data and historical monitoring results corresponding to the corresponding financial data, wherein the historical monitoring results are mostly obtained by the monitoring subject through manual evaluation of the corresponding historical financial data;

[0049] Read the obtained historical financial data and perform feature extraction on the data to obtain financial feature information corresponding to the corresponding historical financial information, wherein the financial feature includes several types of financial data features corresponding to the corresponding historical financial data;

[0050] Based on the historical financial information and the corresponding historical monitoring results, the correlation between different financial data features and the corresponding historical monitoring results is obtained, and a corresponding feature-monitoring graph is constructed based on the correlation, and corresponding training data is constructed based on the graph;

[0051] Construct a supervision model, which is based on a convolutional neural network as a skeleton framework; the supervision model includes an input layer, a convolutional layer, a fully connected layer, and an output layer

[0052] Randomly initialize the weight W and bias term b in the corresponding supervision model;

[0053] After initialization is completed, a batch of sample data is randomly selected from the training data to perform iterative training on the corresponding supervision model, and the corresponding sample data participating in the iterative training is forward propagated to obtain the output result corresponding to the output layer. The error value between the output result and the corresponding training sample is recorded based on the supervision loss function, and the cycle is iterated until the preset number of iterations is reached or the corresponding error value is less than the preset error threshold; then the training is stopped, otherwise the training continues;

[0054] Back propagation is performed based on the error value obtained based on the supervision loss function, and the weight of the corresponding supervision model is updated; at the same time, the number of corresponding iterative training and the corresponding error value are recorded. If the corresponding number of iterations does not reach the preset number of iterations or the preset error threshold, a new batch of sample data is selected to continue the cyclic iteration until the supervision model meets the requirements;

[0055] Among them, the regulatory loss function formula corresponding to the regulatory model is: In the formula, Y represents the function value of the current supervision loss function, y n , Respectively represent the predicted value and true value of the output result corresponding to the nth sample data; where n = 1, 2, ..., M, M is the total number of sample data participating in iterative training;

[0056] The formula for weight update is: W1=mu×v-δ×dw+W; wherein W1 is the updated weight, mu represents a set hypervariable, δ represents the learning rate of the supervision model, W represents the weight before update, and V is the set initialization.

[0057] The technical effects and advantages of the integrated supervision platform based on distributed blockchain of the present invention are as follows:

[0058] 1. By building a distributed blockchain, the supervision process is decentralized and the data is transparent; all participants can access relevant supervision data on the blockchain, which improves the efficiency and transparency of supervision;

[0059] 2. The automated nature of distributed blockchain reduces manual intervention and reduces labor costs in the regulatory process. At the same time, since data is automatically stored, verified, and transmitted on the blockchain, the use of paper documents is reduced, reducing storage and logistics costs.

[0060] 3. By building a regulatory model and using historical data for training, the platform can realize intelligent analysis and prediction of financial data, improving the accuracy and timeliness of supervision; it helps to timely identify potential risks and problems, and take corresponding measures to prevent and deal with them. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A schematic diagram of a system invention of an integrated supervision platform based on a distributed blockchain of the present invention;

[0062] Figure 2 This is a flow chart of an integrated supervision platform method based on distributed blockchain of the present invention. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] Example 1

[0065] See also Figure 1 As shown, the integrated supervision platform based on distributed blockchain described in this embodiment includes:

[0066] A blockchain construction module is used to authenticate the identity of the participants in the target regulatory process, obtain the corresponding participation category, and build a corresponding distributed blockchain based on the participation category; the participation category includes the regulatory subject, the regulated object, and the third-party institution;

[0067] The data collection module is used to collect financial data related to the regulated object and obtain corresponding financial information;

[0068] A data processing module is used to process the collected financial information to obtain corresponding data to be supervised; and to construct corresponding storage tasks based on the data to be supervised;

[0069] The data uploading module is used to upload and store the acquired data to be supervised into the constructed distributed blockchain based on the constructed storage task;

[0070] The data supervision module is used to review the data to be supervised stored by the supervised object in the distributed blockchain, obtain the corresponding review results, and generate corresponding supervision results and take corresponding supervision measures based on the corresponding review results;

[0071] The modules are connected to each other via wired and / or wireless means to achieve data transmission between the modules.

[0072] It should be further explained that, in the specific implementation process, the process of authenticating the participants in the target regulatory process, obtaining the corresponding participation category, and building the corresponding distributed blockchain based on the participation category includes:

[0073] A user terminal is set up to obtain the participants involved in the corresponding supervision process and mark them as participating users. The participating users log in to the user terminal and upload identity information for identity authentication. If the identity authentication fails, a login failure notification is fed back to the participating users. If the identity authentication succeeds, the user terminal obtains the identity information of the corresponding participating users for identity authentication, and divides the corresponding participating users into participation categories based on the identity information, and the participation categories include the supervision subject, the supervised object and the third-party organization; and at the same time, a corresponding identity number is generated based on the identity information of the participating users;

[0074] Furthermore, based on the distributed network, a corresponding distributed blockchain is constructed in combination with the participation categories corresponding to the corresponding participating users. The distributed blockchain is composed of a number of supervision nodes, audit nodes and storage nodes; wherein the node corresponding to the supervision subject is the supervision node in the distributed blockchain, the third-party organization corresponds to the audit node, and the supervised object corresponds to the storage node.

[0075] It should be further explained that, in the specific implementation process, the process of collecting financial data involved in the regulated object and obtaining the corresponding regulatory information includes:

[0076] The data collection module is provided with a plurality of collection units, and the collection units collect financial data involved in the regulated object from multiple data sources based on the user terminal to obtain corresponding financial information, and the financial information is composed of a plurality of financial data, and the financial data includes financial transaction information, financial information, etc. involved in the regulated object;

[0077] It should be further explained that, in the specific implementation process, the collected financial information is processed to obtain the corresponding data to be supervised; and the process of constructing the corresponding storage task based on the data to be supervised includes:

[0078] Reading the collected financial information, and performing data preprocessing on the collected financial information, wherein the preprocessing includes missing value processing and data standardization;

[0079] The missing value processing is used to fill in the missing data in the collected Internet big data, and the missing value processing adopts a multiple interpolation algorithm; the data standardization refers to the standardization of the corresponding photovoltaic data according to the preset format standard; it is used to make the collected data more complete and in a unified format;

[0080] After the data preprocessing is completed, the collection time corresponding to each financial data in the corresponding financial information is obtained respectively, and the timestamp corresponding to the corresponding financial data is obtained based on the collection time, and the timestamp is converted into a binary code to obtain the corresponding time identification code;

[0081] At the same time, the financial data in the corresponding financial information is read, and characteristic segments are extracted to obtain corresponding characteristic data segments, and different levels of weights are assigned to each characteristic data segment, wherein the specific weight is determined by the specific content of the characteristic data segment;

[0082] Read the corresponding characteristic data segment, and randomly generate a set of dynamic data identifiers based on the corresponding characteristic data segment. The dynamic data identifier can be obtained by the following mathematical formula: In the formula, α represents the dynamic data representation corresponding to the feature data segment, which is a digital sequence composed of a group of random numbers; N i N represents the data volume of the i-th feature data segment corresponding to the corresponding financial data; max and N min Respectively represent the maximum data volume and the minimum data volume in the characteristic data segments corresponding to the corresponding financial data, wherein i=1, 2, ..., N0, N0>0 and N0 is an integer, and N0 represents the total number of characteristic data segments;

[0083] Read the dynamic data identifiers of all feature data segments and perform dimensionality reduction processing on them, wherein the dimensionality reduction processing refers to reducing the numbers in the corresponding dynamic data identifiers to 0 or 1;

[0084] Further, the normalized dynamic data identifiers are accumulated according to the assigned weights to obtain the corresponding financial character string;

[0085] Furthermore, a unique financial data identifier is constructed based on the financial character string and the time coding data. The financial data identifiers between different financial data can not only reflect whether the two are similar, but also reflect the degree of similarity between the two; at the same time, they can also be used as data identifiers of corresponding financial data;

[0086] Acquire characteristic information corresponding to the corresponding financial data based on the obtained characteristic data fragments, wherein the characteristic information includes financial growth rate, transaction status and other characteristics;

[0087] At the same time, based on the characteristic information, a bifurcated data tree is constructed for the financial information uploaded by the corresponding regulated object to obtain a corresponding financial data tree, which can be expressed as {(prison_id)∠(financial_l,time)∠(trait_t)}, where prison_id represents the identity number of the regulated object; financial_l,time represents the lth financial data uploaded by the regulated object at the corresponding upload time, and trait_t represents the characteristic information corresponding to the lth financial data;

[0088] Based on the CA-ABE algorithm, the corresponding initial parameter C is generated in combination with the identity information of the corresponding supervised object. The initial parameter C={e,g,H h , Zp}; G is a cyclic group of prime order p, g is a generator of the group, e is a bilinear map; Zp is a non-negative minimal simplified residue system modulo p, p is a prime number; H h =(0,1,2,3,4,5);

[0089] Then, based on the initial parameters, the corresponding master key ZMY=g is constructed. θ And the public key GMY={e(g,g) θ ,g λ}; where θ and λ are a random function and a random number respectively, and both θ and λ are ∈Zp; e(g,g) represents the mapping result of the bilinear mapping applied to the generator of the corresponding group;

[0090] Randomly select a random number r∈Zp, and split the corresponding financial data based on the random number r to obtain the secret shard σ_u corresponding to the leaf node in the corresponding financial data tree; and calculate the attributes of the corresponding leaf node. The corresponding mathematical formula is: Where σ_u represents the secret shard corresponding to leaf node u; and Respectively express the first attribute calculation result and the second attribute calculation result of the leaf node; that is, it can be used as the ciphertext field of the leaf node. The ciphertext data corresponding to the corresponding financial data tree is composed of the first attribute calculation result and the second attribute calculation result of all leaf nodes; r_u represents the random number corresponding to the u-th leaf node;

[0091] Then, based on the attribute calculation, corresponding ciphertext data is obtained;

[0092] The financial data identifier corresponding to the corresponding financial data and the corresponding ciphertext data are mutually associated to obtain the corresponding subgroup to be stored

[0093] Counting the identity numbers of the subgroups to be stored and the regulated objects corresponding to all the financial data in the financial information to obtain the corresponding data to be regulated;

[0094] Then, the data capacity corresponding to the corresponding data to be supervised, the financial data identifiers corresponding to each financial data and the identity identifiers of the supervised objects are obtained, and corresponding storage tasks are constructed based on them;

[0095] Furthermore, the storage corresponding to the corresponding supervised object is uploaded to the constructed distributed blockchain based on the user terminal;

[0096] It should be further explained that, in the specific implementation process, the process of uploading and storing the obtained data to be supervised into the constructed distributed blockchain based on the constructed storage task includes:

[0097] After receiving the corresponding storage task, the distributed blockchain will broadcast the corresponding storage task to all storage nodes based on the deployed smart contract, read the financial data identifier in the corresponding storage task, and perform identifier verification on it, and the identifier verification is used to verify whether there is a financial data identifier in the corresponding storage node that is the same as that in the corresponding storage task;

[0098] If the verification fails, indicating that the same financial data identifier exists, a storage failure notification and the reason for the failure are fed back to the user terminal, and then the user terminal can resubmit after verifying the corresponding financial data identifier;

[0099] If the verification is successful, indicating that the corresponding financial data identifier does not exist, a storage node is randomly selected as a temporary node, and other storage nodes are marked as slave nodes;

[0100] Then, the slave node requests the node signature of the corresponding temporary node and attaches its own node signature. After the temporary node obtains the request, it records the node signature of the slave node and feeds back the node signature of the temporary node. Then, the slave node feeds back the node registration information to the temporary node. After receiving the corresponding node registration information, the temporary node returns the registered adjacent node information, otherwise it continues to wait. After the slave node receives the registered adjacent node information, it feeds back confirmation information to the temporary node. After the temporary node receives all the confirmation information, it records the corresponding temporary node as the master node and marks the slave node as the child node.

[0101] Repeat the acquisition process of the corresponding master node until all storage nodes are traversed and all marked master nodes are counted; prioritize all master nodes based on the data capacity in the corresponding storage task and the nearest minimum principle, and mark the corresponding master node as the first shared node, the second shared node, ..., the Kth shared node based on the priority sorting, where K represents the total number of master nodes; and feed back the corresponding storage notification to the user terminal;

[0102] After the user terminal receives the corresponding storage notification, the user terminal uploads the obtained data to be supervised to the corresponding first shared node for storage. The first shared node stores its own node signature and the corresponding data to be supervised in the second shared node, and so on until it is stored in the Kth shared node; at the same time, a corresponding storage log is generated and broadcast to the corresponding distributed blockchain.

[0103] It should be further explained that, in the specific implementation process, the process of consulting the data to be supervised stored by the supervised object in the distributed blockchain, obtaining the corresponding consulting results, and generating the corresponding supervision results based on the corresponding consulting results and taking the corresponding supervision measures includes:

[0104] If the participating user is a regulatory subject, and the regulatory subject needs to review the historical financial information uploaded by a regulated object, the regulatory subject may generate a corresponding regulatory request based on the user terminal, and the regulatory request includes the identity number of the regulatory subject, the identity number of the regulated object and other relevant information;

[0105] Feedback the corresponding supervision request to the supervision node corresponding to the corresponding supervision subject; the supervision node searches the storage nodes in the corresponding distributed blockchain based on the supervision request to obtain a corresponding search list, which includes a number of storage nodes related to the corresponding supervision request, and feeds back the storage node to the corresponding user terminal. The supervision subject can check whether the storage node in the corresponding search list is associated with the corresponding supervised object. If not, the supervision subject is reminded to check whether the corresponding supervision request is accurate and conduct a secondary search; if so, the corresponding supervision node feeds back a review notification;

[0106] After receiving the corresponding inquiry notification, the supervisory node obtains the registration information and node signature of the node corresponding to each storage node in the corresponding search list and performs verification. If the verification fails, based on the identity number of the supervised object stored in the corresponding storage node, a node information abnormality notification is fed back to the corresponding supervised object to remind the corresponding supervised object to conduct information inspection on the uploaded financial information; if the verification passes, the corresponding supervisory node sends its own node signature to the storage node. After the storage node receives the corresponding node signature, it feeds back a confirmation notification to the supervisory node and marks the corresponding storage node as an inquiry node;

[0107] The node to be monitored receives confirmation information from all storage nodes in the corresponding monitoring list and obtains the corresponding reference list. Then, based on the data processing process, the data to be monitored stored in the corresponding reference node is reversely processed to obtain corresponding regulatory information, which is the original financial information corresponding to the data to be monitored after the reverse process of the data processing;

[0108] The supervisory subject may review the corresponding supervisory data and input the corresponding supervisory data as input parameters into a pre-built supervisory model to obtain a corresponding first auxiliary result;

[0109] Furthermore, the supervisory subject may generate corresponding supervision results based on the cognition after reviewing the corresponding supervision information and the first auxiliary result, and the supervision results include normal supervision, abnormal supervision and risk supervision;

[0110] Furthermore, the supervisory body takes corresponding supervisory measures based on the supervisory results; if the supervisory result is no abnormal supervision, no other operations are performed; if the supervisory result is abnormal supervision, the supervisory body marks the abnormal part in the corresponding supervisory information and generates a corresponding supervision notice to urge the corresponding supervised object to correct the abnormal part, and conduct a second supervision after the correction is completed; if the supervisory result is risk supervision, the supervisory body dispatches supervisors to conduct on-site supervision of the corresponding supervised object;

[0111] It should be further explained that in the specific implementation process, the process of building a regulatory model includes:

[0112] Obtaining a number of historical financial data and historical monitoring results corresponding to the corresponding financial data, wherein the historical monitoring results are mostly obtained by the monitoring subject through manual evaluation of the corresponding historical financial data;

[0113] Read the obtained historical financial data and perform feature extraction on the data to obtain financial feature information corresponding to the corresponding historical financial information, wherein the financial feature includes several types of financial data features corresponding to the corresponding historical financial data;

[0114] Based on the historical financial information and the corresponding historical monitoring results, the correlation between different financial data features and the corresponding historical monitoring results is obtained, and a corresponding feature-monitoring graph is constructed based on the correlation, and corresponding training data is constructed based on the graph;

[0115] Construct a supervision model, which is based on a convolutional neural network as a skeleton framework; the supervision model includes an input layer, a convolutional layer, a fully connected layer, and an output layer

[0116] Randomly initialize the weight W and bias term b in the corresponding supervision model;

[0117] After initialization is completed, a batch of sample data is randomly selected from the training data to perform iterative training on the corresponding supervision model, and the corresponding sample data participating in the iterative training is forward propagated to obtain the output result corresponding to the output layer. The error value between the output result and the corresponding training sample is recorded based on the supervision loss function, and the cycle is iterated until the preset number of iterations is reached or the corresponding error value is less than the preset error threshold; then the training is stopped, otherwise the training continues;

[0118] Back propagation is performed based on the error value obtained based on the supervision loss function, and the weight of the corresponding supervision model is updated; at the same time, the number of corresponding iterative training and the corresponding error value are recorded. If the corresponding number of iterations does not reach the preset number of iterations or the preset error threshold, a new batch of sample data is selected to continue the cyclic iteration until the supervision model meets the requirements;

[0119] Among them, the regulatory loss function formula corresponding to the regulatory model is: In the formula, Y represents the function value of the current supervision loss function, y n , Respectively represent the predicted value and true value of the output result corresponding to the nth sample data; where n = 1, 2, ..., M, M is the total number of sample data participating in iterative training;

[0120] The formula for weight update is: W1=mu×v-δ×dw+W; wherein W1 is the updated weight, mu represents a set hypervariable, δ represents the learning rate of the supervision model, W represents the weight before update, and V is the set initialization parameter.

[0121] The present invention has significant beneficial effects in improving regulatory efficiency, enhancing data security and privacy protection, improving data integrity and traceability, reducing regulatory costs, promoting multi-party collaboration and trust, supporting smart contracts and automated supervision, and improving regulatory accuracy and timeliness.

[0122] Example 2

[0123] See also Figure 2 As shown, the part not described in detail in this embodiment is described in Example 1, which provides a working method of an integrated supervision platform based on blockchain, including:

[0124] Step 1: Authenticate the identity of the participants in the target regulatory process, obtain the corresponding participation category, and build the corresponding distributed blockchain based on the participation category; the participation category includes the regulatory subject, the regulated object, and the third-party institution;

[0125] Step 2: Collect financial data related to the regulated entity and obtain relevant financial information;

[0126] Step 3: Process the collected financial information to obtain corresponding data to be regulated; and construct corresponding storage tasks based on the data to be regulated;

[0127] Step 4: Upload and store the acquired data to be supervised into the constructed distributed blockchain based on the constructed storage task;

[0128] Step 5: Review the data to be supervised stored by the supervised object in the distributed blockchain, obtain the corresponding review results, generate corresponding supervision results based on the corresponding review results, and take corresponding supervision measures.

[0129] Example 3

[0130] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the integrated supervision platform based on distributed blockchain provided above is implemented.

[0131] Since the electronic device introduced in this embodiment is an electronic device used to implement an integrated supervision platform based on distributed blockchain in the embodiment of this application, based on the integrated supervision platform based on distributed blockchain introduced in the embodiment of this application, the technical personnel of this field can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application is not described in detail here. As long as the technical personnel of this field implement the electronic device used in the embodiment of this application based on the integrated supervision platform of distributed blockchain, it belongs to the scope of protection of this application.

[0132] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0133] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. An integrated supervision platform based on distributed blockchain, characterized in that: include: The blockchain building module is used to authenticate the identity of the participants in the target regulatory process, obtain the corresponding participation category, and build the corresponding distributed blockchain based on the participation category; The participating categories include regulatory entities, regulated entities and third-party institutions; The data collection module is used to collect financial data related to the regulated object and obtain corresponding financial information; A data processing module is used to process the collected financial information to obtain corresponding data to be supervised; and to construct corresponding storage tasks based on the data to be supervised; The data uploading module is used to upload and store the acquired data to be supervised into the constructed distributed blockchain based on the constructed storage task; The data supervision module is used to review the data to be supervised stored by the supervised objects in the distributed blockchain, obtain corresponding review results, generate corresponding supervision results based on the corresponding review results, and take corresponding supervision measures.

2. According to claim 1, an integrated supervision platform based on distributed blockchain is characterized in that: The process of authenticating the participants in the target regulatory process, obtaining the corresponding participation category, and building the corresponding distributed blockchain based on the participation category includes: A user terminal is set up to obtain the participants involved in the corresponding supervision process and mark them as participating users. The participating users log in to the user terminal and upload identity information for identity authentication. If the identity authentication fails, a login failure notification is fed back to the participating users. If the identity authentication succeeds, the user terminal obtains the identity information of the corresponding participating users for identity authentication, and divides the corresponding participating users into participation categories based on the identity information, and the participation categories include the supervision subject, the supervised object and the third-party organization; and at the same time, a corresponding identity number is generated based on the identity information of the participating users; Furthermore, based on the distributed network, a corresponding distributed blockchain is constructed in combination with the participation categories corresponding to the corresponding participating users. The distributed blockchain consists of a number of supervision nodes, audit nodes and storage nodes.

3. According to claim 2, an integrated supervision platform based on distributed blockchain is characterized in that: The process of collecting financial data related to the regulated entity and obtaining the corresponding regulatory information includes: The data collection module is provided with a plurality of collection units, and the collection units collect financial data related to the supervised object based on the user terminal to obtain corresponding financial information, and the financial information is composed of a plurality of financial data.

4. According to claim 3, an integrated supervision platform based on distributed blockchain is characterized in that: Process the collected financial information to obtain the corresponding data to be regulated; The process of building a corresponding storage task based on the data to be supervised includes: Reading the collected financial information, and performing data preprocessing on the collected financial information, wherein the preprocessing includes missing value processing and data standardization; After the data preprocessing is completed, the financial data in the corresponding financial information is read, and feature fragments are extracted to obtain corresponding feature data fragments, and corresponding financial data identifiers are constructed based on the feature data fragments, and the corresponding financial information is encrypted to obtain corresponding ciphertext data; The financial data identifier and ciphertext data corresponding to the corresponding financial data are correlated to obtain the corresponding subgroup to be stored. Counting the identity numbers of the subgroups to be stored and the regulated objects corresponding to all the financial data in the financial information to obtain the corresponding data to be regulated; Obtain the data capacity corresponding to the corresponding data to be regulated, the financial data identifiers corresponding to each financial data, and the identity identifiers of the regulated objects, and build corresponding storage tasks based on them; Based on the user terminal, the storage corresponding to the corresponding regulated object is uploaded to the constructed distributed blockchain.

5. According to claim 3, an integrated supervision platform based on distributed blockchain is characterized in that: The process of constructing a corresponding financial data identifier based on the characteristic data fragment includes: Respectively obtain the collection time corresponding to each financial data in the corresponding financial information, and based on it, obtain the timestamp corresponding to the corresponding financial data, and convert it into a binary code to obtain a corresponding time identification code; Read the corresponding feature data fragments, and assign different levels of weight to each feature data fragment; Read the corresponding characteristic data fragment, and randomly generate a set of dynamic data identifiers α based on the corresponding characteristic data fragment. The acquisition formula of the dynamic data identifier α is: Where α represents the dynamic data representation corresponding to the feature data segment; N i N represents the data volume of the i-th feature data segment corresponding to the corresponding financial data; max and N min Respectively represent the maximum data volume and the minimum data volume in the characteristic data segment corresponding to the corresponding financial data, wherein i=1, 2, ..., N0, N0 represents the total number of characteristic data segments; The dynamic data identifiers of all feature data segments are read and dimension reduction is performed on them; the dynamic data identifiers after dimension reduction are accumulated according to the assigned weights to obtain the corresponding financial character strings; and a unique financial data identifier is constructed based on the financial character string and the time coding data.

6. According to claim 3, an integrated supervision platform based on distributed blockchain is characterized in that: The process of encrypting the corresponding financial information and obtaining the corresponding ciphertext data includes: Acquire characteristic information corresponding to the corresponding financial data based on the obtained characteristic data fragments; Based on the characteristic information, a bifurcated data tree is constructed for the financial information uploaded by the corresponding regulated object to obtain a corresponding financial data tree; Based on the CA-ABE algorithm, the corresponding initial parameter C is generated in combination with the identity information of the corresponding supervised object. The initial parameter C={e,g,H h , Zp}; G is a cyclic group of prime order P, g is a generator of the group, e is a bilinear map; Zp is a non-negative minimal simplified residue system modulo p, p is a prime number; H h =(0,1,2,3,4,5); Then, based on the initial parameters, the corresponding master key ZMY=g is constructed. θ And the public key GMY={e(g,g) θ ,g λ }; where θ and λ are a random function and a random number respectively, and both θ and λ are ∈Zp; e(g,g) represents the mapping result of the bilinear mapping applied to the generator of the corresponding group; Randomly select a random number r∈Zp, and split the corresponding financial data based on the random number r to obtain the secret shard σ_u corresponding to the leaf node in the corresponding financial data tree; and the mathematical formula for calculating the attributes of the corresponding leaf node is: Where σ_u represents the secret shard corresponding to leaf node u; and The first attribute calculation result and the second attribute calculation result of the leaf node are respectively expressed; based on the attribute calculation, the corresponding ciphertext data is obtained.

7. The integrated supervision platform based on distributed blockchain according to claim 6 is characterized in that: The process of uploading and storing the obtained data to be supervised into the constructed distributed blockchain based on the constructed storage task includes: After receiving the corresponding storage task, the distributed blockchain will broadcast the corresponding storage task to all storage nodes based on the deployed smart contract, read the financial data identifier in the corresponding storage task, and perform the identifier verification; If the verification fails, a storage failure notification and the reason for the failure are fed back to the user terminal, and then the user terminal can resubmit after verifying the corresponding financial data identifier; If the verification is successful, a storage node is randomly selected as a temporary node, and other storage nodes are marked as slave nodes; The slave node requests a node signature from the corresponding temporary node and attaches its own node signature. After the temporary node obtains the request, it records the node signature of the slave node and feeds back the node signature of the temporary node. Then, the slave node feeds back the node registration information to the temporary node. After receiving the corresponding node registration information, the temporary node returns the registered adjacent node information, otherwise it continues to wait. After the slave node receives the registered adjacent node information, it feeds back confirmation information to the temporary node. After the temporary node receives all the confirmation information, it records the corresponding temporary node as the master node and marks the slave node as the child node. Repeat the acquisition process of the corresponding master node until all storage nodes are traversed and all marked master nodes are counted; prioritize all master nodes based on the data capacity in the corresponding storage task and the nearest minimum principle, and mark the corresponding master node as the first shared node, the second shared node, ..., the Kth shared node based on the priority sorting, where K represents the total number of master nodes; and feed back the corresponding storage notification to the user terminal; After the user terminal receives the corresponding storage notification, the user terminal uploads the obtained data to be regulated to the corresponding first shared node for storage. The first shared node stores its own node signature and the corresponding data to be regulated in the second shared node, and so on, until it is stored in the Kth shared node; at the same time, a corresponding storage log is generated and broadcast to the corresponding distributed blockchain.

8. The integrated supervision platform based on distributed blockchain according to claim 6 is characterized in that: The process of reviewing the data to be supervised corresponding to the supervised object includes: If the participating user is a regulatory subject, and the regulatory subject needs to review the historical financial information uploaded by a regulated object, the regulatory subject may generate a corresponding regulatory request based on the user terminal; Feedback the corresponding supervision request to the supervision node corresponding to the corresponding supervision subject; the supervision node searches the storage node in the corresponding distributed blockchain based on the supervision request, obtains the corresponding search list, and feeds back the storage node to the corresponding user terminal. The supervision subject can check whether the corresponding search list is a storage node associated with the corresponding supervised object. If not, the supervision subject is reminded to check whether the corresponding supervision request is accurate and conduct a secondary search; if so, the corresponding supervision node feeds back a review notification; After receiving the corresponding query notification, the supervisory node obtains the registration information and node signature of the node corresponding to each storage node in the corresponding search list and performs verification. If the verification fails, a node information abnormality notification is fed back to the corresponding supervised object based on the identity number of the supervised object stored in the corresponding storage node; if the verification passes, the corresponding supervisory node sends its own node signature to the storage node. After the storage node receives the corresponding node signature, it feeds back a confirmation notification to the supervisory node and marks the corresponding storage node as a query node; The node to be monitored receives confirmation information from all storage nodes in the corresponding monitoring list and obtains the corresponding reference list. Then, based on the data processing process, the data to be monitored stored in the corresponding reference node is parsed to obtain the corresponding monitoring information.

9. The integrated supervision platform based on distributed blockchain according to claim 8 is characterized in that: The supervisory subject may review the corresponding supervisory data and input the corresponding supervisory data as input parameters into a pre-built supervisory model to obtain a corresponding first auxiliary result; Furthermore, the regulatory body may generate corresponding monitoring results based on the understanding after reviewing the corresponding regulatory information and the first auxiliary results.

10. The integrated supervision platform based on distributed blockchain according to claim 8 is characterized in that: The process of building a governance model includes: Obtaining a number of historical financial data and the historical monitoring results corresponding to the corresponding financial data; Read the obtained historical financial data and perform feature extraction on the data to obtain financial feature information corresponding to the corresponding historical financial information, wherein the financial feature includes several types of financial data features corresponding to the corresponding historical financial data; Based on the historical financial information and the corresponding historical monitoring results, the correlation between different financial data features and the corresponding historical monitoring results is obtained, and a corresponding feature-monitoring graph is constructed based on the correlation, and corresponding training data is constructed based on the graph; Construct a supervision model, which is based on a convolutional neural network as a skeleton framework; the supervision model includes an input layer, a convolutional layer, a fully connected layer, and an output layer Randomly initialize the weight W and bias term b in the corresponding supervision model; After initialization is completed, a batch of sample data is randomly selected from the training data to perform iterative training on the corresponding supervision model, and the corresponding sample data participating in the iterative training is forward propagated to obtain the output result corresponding to the output layer. The error value between the output result and the corresponding training sample is recorded based on the supervision loss function, and the cycle is iterated until the preset number of iterations is reached or the corresponding error value is less than the preset error threshold; then the training is stopped, otherwise the training continues; Back propagation is performed based on the error value obtained based on the supervision loss function, and the weight of the corresponding supervision model is updated; at the same time, the number of corresponding iterative training and the corresponding error value are recorded. If the corresponding number of iterations does not reach the preset number of iterations or the preset error threshold, a new batch of sample data is selected to continue the cyclic iteration until the supervision model meets the requirements; Among them, the regulatory loss function formula corresponding to the regulatory model is: In the formula, Y represents the function value of the current supervision loss function, y n , Respectively represent the predicted value and true value of the output result corresponding to the nth sample data; where n = 1, 2, ..., M, M is the total number of sample data participating in iterative training; The formula for weight update is: W1=mu×v-δ×dw+W; wherein W1 is the updated weight, mu represents a set hypervariable, δ represents the learning rate of the supervision model, W represents the weight before update, and V is the set initialization parameter.

Images