Method and device for realizing data version management based on block chain in Xinchuang environment, processor and computer readable storage medium thereof

Abstract

The present application relates to a kind of method for realizing data version management based on block chain in Xinxing environment, comprising the following steps: create a block chain to store data version, calculate the hash value of each block;Create new data version, calculate the hash value of data version, and verify data version;Create decentralized data version control mechanism, and control the data version of each node;Different versions of data are classified and managed by machine learning-based data classification algorithm.The present application also relates to a kind of device for realizing data version management based on block chain in Xinxing environment, processor and its storage medium.The method for realizing data version management based on block chain in Xinxing environment, device, processor and its computer readable storage medium of the present application are used, so that the integrity and accuracy of data version are guaranteed, the history record of data version is completely saved, and the security of data is increased by decentralized data version control mechanism.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data version management, in particular to the field of data version management based on blockchain, and specifically refers to a method and device for realizing data version management based on blockchain in a China-specific environment, a processor and a computer readable storage medium thereof. BACKGROUND

[0002] In the current era of big data and artificial intelligence, data has become one of the most important assets in enterprises and organizations, and the growth rate and scale of data have also shown an explosive growth trend. In order to better manage and apply these data assets, the management of data versions is particularly important. Blockchain is a decentralized distributed database technology. Through encryption algorithms, consensus mechanisms and distributed storage technologies, blockchain can achieve secure, transparent and tamper-proof transaction records, thereby building a decentralized trust mechanism. At present, blockchain has been widely used in digital currency, Internet of Things, supply chain management, e-government and other fields. The core concept of blockchain is "block" and "chain". Each block contains some transaction records and a hash value pointing to the previous block, so that a continuous and tamper-proof blockchain can be constructed. The hash value of each block is calculated from the data in the block and the hash value of the previous block. If the previous block is tampered with, the hash value of the current block will also be affected, thereby invalidating the entire blockchain. Therefore, blockchain can guarantee the integrity and security of data. Blockchain also has an important feature of distributed storage. Each node can store the entire blockchain, and can ensure data consistency through consensus mechanisms. In Bitcoin, the consensus mechanism uses the Proof of Work (PoW) algorithm, which requires nodes to obtain the right to record by calculating complex mathematical problems. However, the PoW algorithm has high computational complexity and consumes a lot of computing resources and energy. Therefore, in recent years, some new consensus algorithms have appeared, such as the Proof of Stake (PoS) algorithm, the Delegated Proof of Stake (DPoS) algorithm, etc. These algorithms can reduce computational complexity and energy consumption while ensuring security. Therefore, blockchain technology is a new and very promising technology that has the characteristics of security, transparency, tamper-proof, decentralization, etc. and plays an important role in the process of data version management.

[0003] The prior art mainly has the following deficiencies:

[0004] (1) The integrity and accuracy of data versions cannot be guaranteed and may be affected by factors such as data tampering and data damage.

[0005] (2) Data version history is not complete enough to provide sufficient support for data traceability and audit.

[0006] (3) The centralized data version control mechanism has the risk of single point of failure, reducing the security of data.

[0007] (4) The scalability of data storage and management is insufficient, making it difficult to cope with the growing demand for data storage.

[0008] (5) The transparency and fairness of data are not guaranteed, and there may be data uncertainty and errors. SUMMARY

[0009] The purpose of the present application is to overcome the above-mentioned shortcomings of the prior art, and to provide a method, device, processor and computer readable storage medium for realizing data version management based on blockchain in a signal creation environment that meets the requirements of integrity, accuracy and refinement.

[0010] In order to achieve the above-mentioned purpose, the method, device, processor and computer readable storage medium for realizing data version management based on blockchain in a signal creation environment according to the present application are as follows:

[0011] The main feature of the method for realizing data version management based on blockchain in a signal creation environment is that the method comprises the following steps:

[0012] (1) Create a blockchain to store data versions and calculate the hash value of each block;

[0013] (2) Create a new data version, calculate the hash value of the data version, and verify the data version;

[0014] (3) Create a decentralized data version control mechanism and control the data version of each node;

[0015] (4) Classify and manage different versions of data based on machine learning data classification algorithms.

[0016] Preferably, in step (1), the hash value of each block is calculated as follows:

[0017] The hash value of each block is calculated according to the following formula:

[0018] H(n) = Hash(data(n) + H(n-1));

[0019] Where H(n) is the hash value of the nth block, data(n) is the data version contained in the nth block, and H(n-1) is the hash value pointing to the previous block.

[0020] Preferably, the step (2) specifically comprises the following steps:

[0021] (2.1) creating a new data version, adding it to the blockchain and creating a new block, and calculating the hash value of the data version;

[0022] (2.2) comparing the hash value of the data version with the hash value in the corresponding block in the blockchain to verify the integrity and accuracy of the data version, if the hash value of the data version is different from the hash value in the corresponding block in the blockchain, the data version has been tampered with.

[0023] Preferably, the step (2.1) of calculating the hash value of the data version specifically comprises:

[0024] The hash value of the data version is calculated according to the following formula:

[0025] h = Hash(data);

[0026] Wherein, h is the hash value of the data version, and data is the content of the data version.

[0027] Preferably, the step (4) specifically comprises the following steps:

[0028] (4.1) data preprocessing for a given data set;

[0029] (4.2) feature extraction for the preprocessed data set;

[0030] (4.3) optimizing the DCNN network through a multi-objective optimization algorithm, and using the improved DCNN network as a feature extraction model for model training;

[0031] (4.4) using machine learning algorithm to train the extracted features;

[0032] (4.5) applying the trained model for data classification, and assigning the classification result to the data version number, and performing blockchain management and storage.

[0033] Preferably, the step (4.2) specifically comprises the following steps:

[0034] (4.2.1) constructing a binary string with length m according to each feature of the data, m is the number of features of the data;

[0035] (4.2.2) calculating the fitness of each individual, using a multi-objective fitness function to evaluate each individual, and calculating the multi-objective fitness of individual i;

[0036] (4.2.3) using fast non-dominated sorting algorithm for multi-objective optimization selection;

[0037] (4.2.4) using a binary partition-based crossover operator, exchanging a certain continuous binary string in individual i and individual j;

[0038] (4.2.5) using a binary partition-based mutation operator, inverting a certain binary string in individual i;

[0039] (4.2.6) merging the individual obtained after the selection, crossover and mutation operations with the current population;

[0040] (4.2.7) repeating (4.2.2) to (4.2.6) until the stop condition is met.

[0041] Preferably, the step (4.2.2) of calculating the multi-objective fitness of individual i specifically comprises:

[0042] The multi-objective fitness of individual i is calculated according to the following formula:

[0043]

[0044] wherein k is the preset maximum number of features, θ is the penalty coefficient, f1 is the classification error, f2 is the number of features, p i represents the number of selected features in individual i.

[0045] Preferably, the step (4.3) specifically comprises the following steps:

[0046] (4.3.1) inputting the training set into the DCNN network and calculating the gradient error;

[0047] (4.3.2) calculating the updated weight;

[0048] (4.3.3) updating each weight value obtained from the training sample and calculating multiple times according to the method of updating the weight;

[0049] (4.3.4) selecting the mean absolute percentage error, mean absolute error and mean square error as the three indicators to evaluate the prediction model;

[0050] (4.3.5) performing multi-objective optimization on the four objective functions, and obtaining the optimal solution according to the Pareto solution set;

[0051] (4.3.6) obtaining the Pareto front.

[0052] The device for realizing data version management based on block chain in a signal creation environment, its main feature is that the device comprises:

[0053] a processor configured to execute computer executable instructions;

[0054] A memory for storing one or more computer executable instructions, which, when executed by the processor, implement the steps of the method for realizing data version management based on a blockchain in a Xingcheng environment.

[0055] The processor for realizing data version management based on a blockchain in a Xingcheng environment, which is mainly characterized in that the processor is configured to execute computer executable instructions, which, when executed by the processor, implement the steps of the method for realizing data version management based on a blockchain in a Xingcheng environment.

[0056] The computer readable storage medium, which is mainly characterized in that a computer program is stored thereon, and the computer program can be executed by a processor to implement the steps of the method for realizing data version management based on a blockchain in a Xingcheng environment.

[0057] The method, device, processor and computer readable storage medium for realizing data version management based on a blockchain in a Xingcheng environment of the application guarantee the integrity and accuracy of data versions, complete the history record of data versions, increase the security of data by using a decentralized data version control mechanism, and have the characteristics of strong scalability, strong transparency and strong manageability. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 The flowchart of the method for realizing data version management based on a blockchain in a Xingcheng environment of the application. DETAILED DESCRIPTION

[0059] In order to more clearly describe the technical content of the application, the following further describes in combination with specific embodiments.

[0060] The method for realizing data version management based on a blockchain in a Xingcheng environment of the application includes the following steps:

[0061] (1) creating a blockchain to store data versions, and calculating the hash value of each block;

[0062] (2) creating a new data version, calculating the hash value of the data version, and verifying the data version;

[0063] (3) creating a decentralized data version control mechanism, and controlling the data version of each node;

[0064] (4) classifying and managing different versions of data by using a machine learning-based data classification algorithm.

[0065] As a preferred embodiment of the application, the step (1) of calculating the hash value of each block specifically includes:

[0066] The hash value of each block is calculated according to the following formula:

[0067] H(n) = Hash(data(n) + H(n-1));

[0068] Wherein, H(n) is the hash value of the nth block, data(n) is the data version contained in the nth block, H(n-1) is the hash value pointing to the previous block.

[0069] As a preferred embodiment of the present application, the step (2) specifically comprises the following steps:

[0070] (2.1) Create a new data version, add it to the blockchain and create a new block, and calculate the hash value of the data version;

[0071] (2.2) Compare the hash value of the data version with the hash value in the corresponding block in the blockchain to verify the integrity and accuracy of the data version, if the hash value of the data version is different from the hash value in the corresponding block in the blockchain, the data version has been tampered with.

[0072] As a preferred embodiment of the present application, the step (2.1) of calculating the hash value of the data version is specifically:

[0073] The hash value of the data version is calculated according to the following formula:

[0074] h = Hash(data);

[0075] Wherein, h is the hash value of the data version, and data is the content of the data version.

[0076] As a preferred embodiment of the present application, the step (4) specifically comprises the following steps:

[0077] (4.1) Data preprocessing is performed on the given data set;

[0078] (4.2) Feature extraction is performed on the preprocessed data set;

[0079] (4.3) The DCNN network is optimized by a multi-objective optimization algorithm, and the improved DCNN network is used as a feature extraction model for model training;

[0080] (4.4) The extracted features are trained using a machine learning algorithm;

[0081] (4.5) The trained model is applied to data classification, and the classification result is assigned to the data version number, and the blockchain management and storage are performed.

[0082] As a preferred embodiment of the present application, the step (4.2) specifically comprises the following steps:

[0083] (4.2.1) Constructing a binary string with a length of m according to each feature of the data, m being the number of features of the data;

[0084] (4.2.2) Calculating the fitness of each individual, using a multi-objective fitness function to evaluate each individual, and calculating the multi-objective fitness of the individual i;

[0085] (4.2.3) Performing multi-objective optimization selection using a fast non-dominated sorting algorithm;

[0086] (4.2.4) Performing a binary partition-based crossover operation to exchange a certain continuous binary string in the individual i and the individual j;

[0087] (4.2.5) Performing a binary partition-based mutation operation to invert a certain binary string in the individual i;

[0088] (4.2.6) Merging the individual obtained after the selection, crossover and mutation operations with the current population;

[0089] (4.2.7) Repeating the execution of (4.2.2) to (4.2.6) until the stop condition is met.

[0090] As a preferred embodiment of the present application, the step (4.2.2) of calculating the multi-objective fitness of the individual i specifically comprises:

[0091] The multi-objective fitness of the individual i is calculated according to the following formula:

[0092]

[0093] wherein k is the preset maximum number of features, θ is the penalty coefficient, f1 is the classification error, f2 is the number of features, p i represents the number of selected features in the individual i.

[0094] As a preferred embodiment of the present application, the step (4.3) specifically comprises the following steps:

[0095] (4.3.1) Inputting the training set into the DCNN network and calculating the gradient error;

[0096] (4.3.2) Calculating the updated weight;

[0097] (4.3.3) Updating each weight value obtained from the training sample and calculating multiple times according to the method of updating the weight;

[0098] (4.3.4) Select the mean absolute percentage error, mean absolute error and mean square error three indicators to evaluate the prediction model;

[0099] (4.3.5) Multi-objective optimization is carried out on the four objective functions, and the optimal solution is obtained according to the Pareto solution set;

[0100] (4.3.6) The Pareto front is obtained.

[0101] The device for realizing the data version management based on the block chain in the Xingcheng environment, wherein the device comprises:

[0102] The processor is configured to execute computer executable instructions.

[0103] The memory stores one or more computer executable instructions, and the computer executable instructions are executed by the processor to realize the steps of the method for realizing the data version management based on the block chain in the Xingcheng environment.

[0104] The processor for realizing the data version management based on the block chain in the Xingcheng environment, wherein the processor is configured to execute computer executable instructions, and the computer executable instructions are executed by the processor to realize the steps of the method for realizing the data version management based on the block chain in the Xingcheng environment.

[0105] The computer readable storage medium of the application, wherein the computer program is stored on the computer readable storage medium, and the computer program can be executed by the processor to realize the steps of the method for realizing the data version management based on the block chain in the Xingcheng environment.

[0106] In the current big data and artificial intelligence era, data has become one of the most important assets in enterprises and organizations, and the growth rate and scale of data also show an explosive growth trend. In order to better manage and apply these data assets, the management of data version is particularly important.

[0107] In a specific embodiment of this invention, a blockchain-based data version management method is proposed for use in a domestically developed information technology environment. This method aims to use blockchain technology to store and verify the integrity and accuracy of data versions. Each data version is stored on the blockchain, and hash values ​​are used to verify the integrity of data in each block. When a new data version is created, it is added to the blockchain, creating a new block. The new block contains a hash value pointing to the previous block, ensuring the entire blockchain is continuous and immutable. Furthermore, this method provides a decentralized data version control mechanism, allowing each node to access the entire blockchain and verify the integrity and accuracy of any data version. In addition, this invention proposes a machine learning-based data classification method. After feature selection, feature extraction, and classification using machine learning, the classification results are assigned a data version number for blockchain management and storage.

[0108] (1) Blockchain creation and storage:

[0109] First, a blockchain needs to be created to store data versions. Each block contains one or more data versions, along with a hash value pointing to the previous block. The hash value of each block is calculated using the following formula:

[0110] H(n)=Hash(data(n)+H(n-1));

[0111] Where H(n) is the hash value of the nth block, data(n) is the data version contained in the nth block, and H(n-1) refers to the hash value of the previous block.

[0112] When calculating the hash value, data(n) and H(n-1) are concatenated and then the hash function is used for calculation. This ensures that if the previous block is tampered with, the hash value of the current block will also be affected, thus invalidating the entire blockchain.

[0113] (2) Data version creation and verification:

[0114] When a new data version is created, it is added to the blockchain, creating a new block. The new block contains a hash value pointing to the previous block, ensuring the entire blockchain is continuous and immutable. To verify the integrity and accuracy of the data version, the hash value is calculated using the following formula:

[0115] h = Hash(data);

[0116] where h is the hash value of the data version, and data is the content of the data version. The h can be compared with the hash value in the corresponding block in the blockchain, thereby verifying the integrity and accuracy of the data version. If h is different from the hash value in the corresponding block in the blockchain, it means that the data version has been tampered with.

[0117] (3) Decentralized data version control mechanism:

[0118] Each node can access the entire blockchain and verify the integrity and accuracy of any data version. If a node finds that a data version is incorrect, it can send a notification to the network so that other nodes can check and correct the error. This decentralized data version control mechanism can ensure the consistency and integrity of the data version, and also prevent data from being tampered with. When multiple nodes modify the same data version, each node will create its own version and add it to the blockchain. Since each block contains a hash value pointing to the previous block, it can ensure that the entire blockchain is continuous and tamper-proof.

[0119] (4) Machine learning-based data classification:

[0120] This step proposes a machine learning-based data classification algorithm that can effectively classify given data and manage different versions of data. The main steps include:

[0121] 1) Data preprocessing:

[0122] For a given data set, some data preprocessing work is needed to make subsequent processing more efficient. Specifically, the original data set is normalized to eliminate the scale difference between different attributes, and noise in the data is removed to reduce interference.

[0123] 2) Feature selection:

[0124] For the preprocessed data set, the most representative features are selected for extraction. In this step, an evolutionary algorithm-based multi-objective optimization algorithm is proposed to select features from the data, which includes the following steps:

[0125] ① Perform initialization operations.

[0126] The data obtained in step two has m features, and each feature is treated as a binary string, i.e., a binary string of length m is constructed, and the elements in the string represent the presence (1) or absence (0) of the feature. The initial population size is N, and each individual (i.e., a binary string of length m) is randomly generated.

[0127] ② Fitness evaluation operation is performed.

[0128] The fitness of each individual is calculated, and a multi-objective fitness function is used to evaluate each individual. The feature selection problem is converted into two objectives:

[0129] Objective 1: Minimize classification error;

[0130] Objective 2: Minimize the number of selected features.

[0131] Let f1 be the classification error, f2 be the number of features, p i represents the number of selected features in individual i, then the multi-objective fitness of individual i is:

[0132]

[0133] where k is the maximum number of features, and θ is the penalty coefficient.

[0134] ③ Selection operation is performed.

[0135] The fast non-dominated sorting algorithm is used for multi-objective optimization selection, and the top N / 2 non-dominated solutions are retained.

[0136] ④ Crossover operation is performed.

[0137] The crossover operator based on binary division is used. The continuous binary string in individual i and individual j is exchanged. The two new individuals after exchange must satisfy the feature number less than or equal to k. The probability of crossover is p c .

[0138] ⑤ Mutation operation is performed.

[0139] The mutation operator based on binary division is used. The binary string in individual i is taken as the inverse. The new individual after mutation must satisfy the feature number less than or equal to k. The probability of mutation is p m .

[0140] ⑥ Update the population.

[0141] The individuals obtained after selection, crossover and mutation operations are merged with the current population, and the top N individuals are retained.

[0142] ⑦ Stop condition.

[0143] Stop condition: repeat steps ②-⑥ until the stop condition is met. Common stop conditions include: reaching the maximum number of iterations, the population fitness no longer changes, reaching the pre-set target value, etc.

[0144] After the algorithm ends, the importance of each feature is evaluated according to the number of times it is selected. The more times it is selected, the more important it is in the feature selection process.

[0145] The innovation of the algorithm mainly lies in the combination of multi-objective optimization algorithm and evolutionary algorithm, the adoption of multi-objective fitness function and roulette selection operator based on fitness value to realize feature selection.

[0146] The innovative improvements of the method are embodied in the following aspects:

[0147] Combination of evolutionary algorithm and multi-objective optimization algorithm: most traditional feature selection algorithms are based on single-objective optimization, while the algorithm introduces the idea of multi-objective optimization, and performs feature selection by balancing the accuracy and diversity of feature subsets.

[0148] Penalty term is introduced: the algorithm introduces a penalty term to constrain the search space, prevent overfitting and improve the generalization performance of the model.

[0149] 3) Feature extraction:

[0150] The feature extraction model adopted by the application is an improved deep convolutional neural network (DCNN) for model training. The application adopts a multi-objective optimization algorithm to optimize the DCNN network, and the specific steps and algorithm improvements are as follows:

[0151] The training set V (V1, V1, …, V i ) is input into the DCNN network, and in the parameters of the DCNN network, the weight is W and the bias is b.

[0152] 1. Calculate the gradient error. The actual output v i ′ The gradient error of the ideal output v i is:

[0153] δ k =v i ′ (1v i ′ )(v i -v i ′ )

[0154] The gradient error of the hidden layer element a and the hidden layer element b is:

[0155]

[0156] Wherein, the connection weight of node a and node b is θ ab , and the value calculated by the activation function of node b is δ b .

[0157] 2. Update the weight. The formula for updating the weight is as follows:

[0158] θ ij = θ ij + Δθ ij , Δθ ij = φ0 i δ j

[0159] where φ is the learning rate, O i is the output of node i, O j and δ j are the output of node j and the recurrent gradient error, respectively.

[0160] 3. Update each weight value obtained from the training sample, and calculate multiple times according to the method of updating the weight. Repeat the above process until the variance of the output error is small enough. The variance of the output error is expressed by the following formula:

[0161] E = ∑ m ∑ n (d mn -O mn )) 2 << ε E = ∑ s (∑ z (d sz -o sz )) 2 << ε

[0162] In the formula, the output node sequence is n, the training sequence is m, d mn is the actual output of the node n of the sample m.

[0163] Therefore, the objective function A is defined as: E = ∑ m ∑ n (d mn -O mn )) 2 << ε E = ∑ s (∑ z (d sz -o sz )) 2 << ε

[0164] Objective functions B, C, and D select the mean absolute percentage error (MAPE), the mean absolute error (MAE), and the root mean square error (RMSE) as three indicators to evaluate the prediction model. The specific formula is as follows:

[0165]

[0166]

[0167]

[0168] where N represents the number of samples, Y i represents the actual observation value of the model, represents the predicted value of the model.

[0169] Among the above several indicators, MAPE represents the average of absolute percentage error, which not only considers the prediction error, but also calculates the ratio of prediction error to observation value, and can be used to measure the pros and cons of the model. MAE represents the average of absolute error, which well avoids the problem of affecting the accuracy of the prediction result due to the mutual offset of errors. RMSE is the square root of mean square error (MSE), which is used to evaluate the deviation between the predicted value and the actual observation value, and is often used to measure the prediction effect of the neural network model.

[0170] Under the same dimension, the four objective functions are multi-objective optimized, and the optimal solution is obtained according to the Pareto solution set.

[0171] The set of Pareto optimal solutions is the Pareto optimal solution set, and its function expression is as follows:

[0172]

[0173] Then, the Pareto frontier is obtained. The Pareto frontier is the set of target values corresponding to the Pareto optimal solution set, and the formula is as follows:

[0174] p f ={F(x)|x∈p s}

[0175] 4) Model training

[0176] For the extracted features, a machine learning algorithm is used for model training. In this algorithm, support vector machine (SVM) algorithm is used as a classifier for model training.

[0177] 5) Apply the trained model to data classification

[0178] After the model training is completed, the trained model is applied to data classification, and the classification result is assigned to the data version number for blockchain management and storage.

[0179] Based on the blockchain version control mechanism, data tampering and damage can be effectively prevented, and the credibility and reliability of the data can be improved. The technical effects of the method proposed in the present application include:

[0180] (1) The integrity and accuracy of the data version are guaranteed. Since each data version is stored on the blockchain, and the hash value is used to verify the integrity of the data in each block, any tampering of the data can be discovered in time.

[0181] (2) The history of data versions is completely saved. Each block contains one or more data versions and a hash value pointing to the previous block, forming a continuous blockchain, and the history of each data version is completely saved, facilitating data version tracing and auditing.

[0182] (3) The decentralized data version control mechanism increases the security of data. Each node can access the entire blockchain and verify the integrity and accuracy of any data version, reducing the risk of single point failure and improving data security.

[0183] (4) Strong scalability. Since the blockchain itself is a decentralized distributed system, this method can easily achieve horizontal expansion, thereby supporting larger-scale data storage and management.

[0184] (5) Strong transparency. The blockchain is a public distributed ledger, and any node can view the data versions and transaction records therein, so this method can improve data transparency and fairness and reduce data uncertainty.

[0185] (6) Strong manageability. Improved machine learning algorithms are used to classify data, further providing category attributes for data management, which is conducive to fine-grained management of data.

[0186] The specific implementation scheme of the embodiment can be referred to the related description in the above embodiment, which will not be repeated here.

[0187] It can be understood that the same or similar parts in the above embodiments can be mutually referred to, and the contents not described in detail in some embodiments can be referred to the same or similar contents in other embodiments.

[0188] It should be noted that in the description of the present application, the terms "first", "second", etc. are only for the purpose of description, and cannot be understood as indicating or implying relative importance. In addition, in the description of the present application, unless otherwise specified, "a plurality of" means at least two.

[0189] Any process or method descriptions in the flowchart or otherwise described herein can be understood as representing code modules, segments, or portions of code that include one or more executable instructions for implementing specific logic functions or other processes, and the scope of preferred embodiments of the present application includes additional implementation in which the functions are performed in different orders, in substantially simultaneous fashion, or in reverse order, depending on the functions involved, as will be understood by those skilled in the art of the embodiments of the present application.

[0190] It should be understood that each part of the present application can be realized by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be realized by software or firmware stored in a memory and executed by a suitable instruction execution device. For example, if realized by hardware, and as in another embodiment, it can be realized by any one or a combination of the following technologies known in the art: discrete logic circuit with logic gate circuit for implementing logic function on data signal, application specific integrated circuit with suitable combination logic gate circuit, programmable gate array (PGA), field programmable gate array (FPGA) and the like.

[0191] Those skilled in the art of the present technology can understand that all or part of the steps carried out by the above-mentioned embodiment method can be completed by a program instructing the relevant hardware, and the corresponding program can be stored in a computer readable storage medium. When the program is executed, it includes one of the steps of the method embodiment or a combination thereof.

[0192] In addition, each functional unit in each embodiment of the present application can be integrated in one processing module, or each unit can exist physically separately, or two or more units can be integrated in one module. The above-mentioned integrated module can be realized in the form of hardware or in the form of software functional module. The integrated module, if realized in the form of software functional module and sold or used as an independent product, can also be stored in a computer readable storage medium.

[0193] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc.

[0194] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0195] The method, device, processor and computer readable storage medium thereof for realizing data version management based on blockchain in the Xingcheng environment of the present application guarantee the integrity and accuracy of the data version, the history record of the data version is completely saved, the decentralized data version control mechanism increases the security of the data, and has the characteristics of strong scalability, strong transparency and strong manageability.

[0196] In this specification, the application has been described with reference to specific embodiments thereof. It is apparent, however, that various modifications and changes can be made thereto without departing from the spirit and scope of the application. Therefore, the description and drawings should be regarded in an illustrative rather than a restrictive sense.

Claims

1. A method for realizing data version management based on a block chain in a Xinxin environment, characterized in that, The method comprises the following steps: (1) creating a blockchain to store data versions, calculating the hash value of each block; (2) creating a new data version, calculating the hash value of the data version, and verifying the data version; (3) creating a decentralized data version control mechanism and controlling the data version of each node; (4) classifying and managing different versions of data through a machine learning-based data classification algorithm. 2.The method for realizing data version management based on a block chain in a Xinwen environment according to claim 1, characterized in that, The step (1) of calculating the hash value of each block is specifically: The hash value of each block is calculated according to the following formula: H(n) = Hash(data(n) + H(n-1)); Where H(n) is the hash value of the nth block, data(n) is the data version contained in the nth block, and H(n-1) is the hash value pointing to the previous block. 3.The method for realizing data version management based on a block chain in a Xinwen environment according to claim 1, characterized in that, The step (2) specifically comprises the following steps: (2.1) creating a new data version, adding it to the blockchain and creating a new block, and calculating the hash value of the data version; (2.2) comparing the hash value of the data version with the hash value in the corresponding block in the blockchain to verify the integrity and accuracy of the data version, if the hash value of the data version is different from the hash value in the corresponding block in the blockchain, the data version has been tampered with. 4.The method for realizing data version management based on a block chain in a Xinwen environment according to claim 3, characterized in that, The step (2.1) of calculating the hash value of the data version is specifically: The hash value of the data version is calculated according to the following formula: h = Hash(data); Where h is the hash value of the data version, and data is the content of the data version. 5.The method for realizing data version management based on a block chain in a Xinwen environment according to claim 1, characterized in that, The step (4) specifically comprises the following steps: (4.1) data preprocessing for a given data set; (4.2) feature extraction for the preprocessed data set; (4.3) optimizing the DCNN network through a multi-objective optimization algorithm, and using the improved DCNN network as a feature extraction model for model training; (4.4) using machine learning algorithms to train the extracted features; (4.5) applying the trained model for data classification, assigning the classification results to the data version number, and performing blockchain management and storage. 6.The method for realizing data version management based on a block chain in a Xinwen environment according to claim 5, characterized in that, The step (4.2) specifically comprises the following steps: (4.2.1) constructing a binary string of length m according to each feature of the data, where m is the number of features of the data; (4.2.2) calculating the fitness of each individual, using a multi-objective fitness function to evaluate each individual, and calculating the multi-objective fitness of individual i; (4.2.3) using a fast non-dominated sorting algorithm for multi-objective optimization selection; (4.2.4) using a binary partition-based crossover operator to exchange a certain continuous binary string in individual i and individual j; (4.2.5) using a binary partition-based mutation operator to invert a certain binary string in individual i; (4.2.6) merging the individuals obtained after selection, crossover and mutation operations with the current population; (4.2.7) repeating (4.2.2) to (4.2.6) until the stopping condition is met.

7. The method for realizing data version management based on a block chain in a Xinwen environment according to claim 6, characterized in that, The step (4.2.2) of calculating the multi-objective fitness of individual i is specifically: The multi-objective fitness of individual i is calculated according to the following formula: where k is the preset maximum number of features, θ is the penalty coefficient, f1 is the classification error, f2 is the number of features, and p i denotes the number of selected features in individual i. 8.The method for realizing data version management based on a block chain in a Xinwen environment according to claim 5, characterized in that, The step (4.3) specifically comprises the following steps: (4.3.1) inputting the training set into the DCNN network and calculating the gradient error; (4.3.2) calculating the updated weight; (4.3.3) updating each weight value obtained from the training sample and calculating multiple times according to the method of updating the weight; (4.3.4) selecting the mean absolute percentage error, the mean absolute error and the mean square error as the three indexes to evaluate the prediction model; (4.3.5) performing multi-objective optimization on the four objective functions and obtaining the optimal solution according to the Pareto solution set; (4.3.6) obtaining the Pareto front.

9. An apparatus for implementing blockchain-based data version management in a credit creation environment, comprising: The device comprises: a processor configured to execute computer executable instructions; a memory storing one or more computer executable instructions, which, when executed by the processor, implement the steps of the method for realizing data version management based on a blockchain in a Xingcheng environment according to any one of claims 1 to 8. 10.A processor for implementing blockchain-based data version management in a credit creation environment, the processor comprising: The processor is configured to execute computer executable instructions, which, when executed by the processor, implement the steps of the method for realizing data version management based on a blockchain in a Xingcheng environment according to any one of claims 1 to 8.

11. A computer readable storage medium, characterized in that, A computer program is stored thereon, which can be executed by a processor to implement the steps of the method for realizing data version management based on a blockchain in a Xingcheng environment according to any one of claims 1 to 8.

Images