Block chain-based enterprise data processing method and system

Abstract

The invention provides an enterprise data processing method and system based on a block chain, and relates to a data processing technology, and the method comprises the steps that a block server analyzes enterprise attributes in a block node to generate an authority relation chain corresponding to the block node, the authority relation chain is a one-way attribute, and the authority is set from low to high; after receiving a data acquisition request of a first block node, the block server determines a first relation chain, a first decomposition mode and a first calculation mode based on the data acquisition request; decomposing the first relation chain based on the first decomposition mode to obtain second relation chains, and assembling different second relation chains to generate a fusion data chain; and generating a request instruction, distributing the request instruction to a fusion node in the fusion data chain, and feeding back the request instruction to the block server based on Hash encryption, wherein the block server calculates based on the first calculation mode to obtain a data processing index.

Description

Technical Field

[0001] The present invention relates to data processing technology, and in particular to a blockchain-based enterprise data processing method and system. Background Art

[0002] Blockchain (English name: blockchain or block chain) is a decentralized distributed ledger with block chain storage, non-tamperability, security and reliability. It combines distributed storage, point-to-point transmission, consensus mechanism, cryptography and other technologies to record transactions and information through a growing data block chain to ensure data security and transparency.

[0003] In the data processing process of multiple enterprises, enterprise data is confidential data and needs to be doubly protected in terms of security and authenticity. Blockchain has the above attributes, but the existing technology cannot effectively collect and process data from multiple enterprises based on blockchain. Summary of the Invention

[0004] The embodiments of the present invention provide a blockchain-based enterprise data processing method and system, which can solve the above-mentioned technical problems, effectively collect and process data from multiple enterprises based on blockchain, realize corresponding correlation analysis, and process and process data in the enterprise chain while ensuring the authenticity and validity of the data.

[0005] A first aspect of an embodiment of the present invention provides a blockchain-based enterprise data processing method, comprising: The block server analyzes the enterprise attributes in the block node and generates a permission relationship chain corresponding to the block node. The permission relationship chain is a one-way attribute and the permissions are set from low to high. After receiving the data collection request from the first block node, the block server determines the first relationship chain, the first decomposition method, and the first calculation method based on the data collection request; Decomposing the first relationship chain based on the first decomposition method to obtain a second relationship chain, and assembling different second relationship chains to generate a fused data chain; Generate a request instruction, distribute it to the fusion node in the fusion data chain, and feed it back to the block server based on hash encryption. The block server obtains the data processing indicator after calculation based on the first calculation method.

[0006] Optionally, in a possible implementation of the first aspect, the block server analyzes enterprise attributes within the block node to generate a permission relationship chain corresponding to the block node, where the permission relationship chain is a unidirectional attribute and permissions are arranged from low to high, including: Extract the shareholder information of each block node's enterprise attributes, and generate the shareholder label and node type of each block node based on the shareholder information; Based on shareholder labels and node types, the corresponding block nodes are connected and integrated to generate the corresponding permission relationship chain.

[0007] Optionally, in a possible implementation of the first aspect, extracting shareholder information of the enterprise attributes of each block node and generating a shareholder label for each block node based on the shareholder information includes: The shareholder tags are classified into personal tags and corporate tags. Each block node corresponds to at least one personal tag and / or corporate tag. If it is determined that a block node only has a personal tag and does not correspond to the personal tags of other block nodes, the corresponding block node will be used as the chain termination node; If the personal tag of a block node is determined to correspond to the personal tags of other block nodes, the corresponding block node will be used as the starting node of the chain; If the enterprise tag of a block node is determined to correspond to other block nodes or other block node enterprise tags, the corresponding block node will be used as the starting node of the chain.

[0008] Optionally, in a possible implementation of the first aspect, if it is determined that the block node has only an enterprise tag and does not correspond to the enterprise tags of other block nodes, the corresponding block node is used as the chain termination node; If the enterprise label of a block node is determined to correspond to the enterprise label of another block node, the corresponding block node will be used as the starting node of the chain; If the personal tag of the block node is determined to correspond to the personal tags of other block nodes, the corresponding block node will be used as the starting node of the chain.

[0009] Optionally, in a possible implementation of the first aspect, generating a corresponding authority relationship chain by fusing and connecting corresponding block nodes based on shareholder labels and node types includes: If it is determined that a block node has the same enterprise tag as another block node, the two block nodes are connected, and the chain of block nodes with corresponding enterprise tags is set towards the corresponding block node; If it is determined that the personal tag of a block node is the same as the personal tag of another block node, a personal relay node is generated, and the personal relay node is connected to the two block nodes respectively, and the chains of the two block nodes are directed towards the corresponding personal relay node; After determining that there are other chain starting nodes in front of any chain starting node, the corresponding chain starting node is used as a chain relay node; The chain start node, chain relay node, chain end node and personal label are integrated to obtain the permission relationship chain.

[0010] Optionally, in a possible implementation of the first aspect, the fusing of the chain start node, the chain relay node, the chain end node, and the personal tag to obtain the permission relationship chain includes: Perform secondary connections on the nodes that are connected to each other to obtain the permission relationship chain formed by all directly or indirectly connected nodes; Determine the node corresponding to the personal label in the authority relationship chain to generate a personal relationship node, and extract the share information corresponding to each shareholder under the corresponding chain.

[0011] Optionally, in a possible implementation of the first aspect, after receiving the data collection request from the first block node, the block server determines the first relationship chain, the first decomposition method, and the first calculation method based on the data collection request, including: Upon receiving the data collection request from the first block node, the block server retrieves the permission relationship chain corresponding to the first block node to obtain a first relationship chain; Extracting the attributes of the data collection request to determine the corresponding first decomposition method, where different attributes have preset first decomposition methods; The first calculation method includes a combination of different machine learning algorithms, and different machine learning algorithms have different preset fusion weights.

[0012] Optionally, in a possible implementation of the first aspect, decomposing the first relationship chain based on the first decomposition method to obtain the second relationship chain, and assembling different second relationship chains to generate the fused data chain includes: If it is determined that there are multiple first relationship chains, the first relationship chain with the most nodes is selected as the main chain and the other first relationship chains are selected as slave chains; The chain node corresponding to the first block node in the slave chain is removed as the decomposition point to obtain a decomposition chain, and the other chain nodes connected to the decomposition point in the decomposition chain are connected to the chain node corresponding to the first block node in the main chain to obtain a fused data chain.

[0013] Optionally, in a possible implementation of the first aspect, the generation request instruction is distributed to the fusion node in the fusion data chain and fed back to the block server based on hash encryption, and the block server obtains the data processing indicator after calculation based on the first calculation method, including: Determine the enterprise information of different chain dimensions corresponding to the data collection request, obtain the fused data chain based on the enterprise information of different chain dimensions, decompose and perform hash calculation, and obtain the multi-level hash keys of different fusion nodes; The enterprise information of the fusion nodes at different positions is encrypted based on a multi-level hash key, and each fusion node has different dimensions in the fusion data chain; The block server inputs the enterprise information into a combination of different machine learning algorithms after receiving the enterprise information, and obtains a data processing index through weighted fusion calculation based on different preset fusion weights.

[0014] In a second aspect, the application provides a blockchain-based enterprise data acquisition and processing system, comprising: The generation module is configured to enable the block server to generate a permission relationship chain corresponding to the block node based on enterprise attribute analysis of the block node. The determination module is configured to enable the block server to determine the first relationship chain, the first decomposition method and the first calculation method based on the data acquisition request after receiving the data acquisition request of the first block node. The fusion module is configured to enable the first relationship chain to be decomposed and processed based on the first decomposition method to obtain a second relationship chain, and to enable different second relationship chains to be assembled to generate a fusion data chain. The calculation module is configured to enable the generation request instruction to be distributed to the fusion nodes in the fusion data chain and fed back to the block server based on hash encryption, and to enable the block server to obtain a data processing index based on the first calculation method. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 The flowchart of the blockchain-based enterprise data processing method; Figure 2 The structural diagram of the blockchain-based enterprise data acquisition and processing system. DETAILED DESCRIPTION

[0016] To make the purpose, technical scheme and advantages of the embodiments of the application clearer, the technical scheme in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0017] The terms "first", "second", "third", "fourth" and the like (if any) in the specification and claims of the application and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than that illustrated or described herein.

[0018] It should be understood that in various embodiments of the present invention, the size of the sequence number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0019] It should be understood that in the present invention, "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0020] It should be understood that in the present invention, "multiple" refers to two or more. "And / or" is only a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "Contains A, B and C", "Contains A, B, C" means that A, B, and C are all included, "Contains A, B or C" means that one of A, B, and C is included, and "Contains A, B and / or C" means that any one, any two, or any three of A, B, and C are included.

[0021] It should be understood that, in the present invention, "B corresponding to A," "B corresponding to A," "A corresponds to B," or "B corresponds to A" means that B is associated with A and B can be determined based on A. Determining B based on A does not mean determining B based solely on A; B can also be determined based on A and / or other information. A and B match when the similarity between A and B is greater than or equal to a preset threshold.

[0022] Depending on the context, "if" as used herein may be interpreted as "when" or "when" or "in response to determining" or "in response to detecting."

[0023] The following specific embodiments are used to describe the technical solution of the present invention in detail. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0024] The present invention provides a method for processing enterprise data based on blockchain. Figure 1 Shown, including: The block server analyzes the enterprise attributes within the block node and generates a permission relationship chain corresponding to the block node. The permission relationship chain is a unidirectional attribute and the permissions are set from low to high. The technical solution provided by the present invention will first analyze the enterprise attributes and determine the corresponding permission relationship chain according to the different permissions of the enterprise. It should be noted that the permission relationship chain is a unidirectional attribute and the permissions are set from low to high. For example, if Company A is a shareholder of Company B and Company B is a shareholder of Company C, then Company A has the permission to view Company B's corporate data and Company B has the permission to view Company C's corporate data. At this time, the present invention can obtain the corresponding permission relationship chain, C→B→A.

[0025] In one possible implementation, the block server analyzes the enterprise attributes within the block node and generates a permission relationship chain corresponding to the block node. The permission relationship chain is a unidirectional attribute and permissions are arranged from low to high, including: The shareholder information of each block node enterprise attribute is extracted, and the shareholder label and node type of each block node are generated based on the shareholder information. The present invention obtains the shareholder information of each block node enterprise attribute. Enterprises of different block nodes may have the same shareholder information or different shareholder information.

[0026] The corresponding authority relationship chain is generated by fusing and connecting the corresponding block nodes based on the shareholder tags and node types. The present invention generates the corresponding authority relationship chain by fusing and connecting the corresponding block nodes based on the shareholder tags and node types. The authority relationship chains generated by the shareholder tags of enterprises and natural persons are likely to be different. For example, while Company B is a shareholder of Company C, Company A can also be a shareholder of Company B. If the natural person Zhang San is also a shareholder of Company C, but Zhang San does not have other shareholders, the chain corresponding to Zhang San may be B→Zhang San.

[0027] In one possible implementation, extracting shareholder information of the enterprise attributes of each block node and generating a shareholder tag for each block node based on the shareholder information includes: The shareholder tags are classified into personal tags and corporate tags. Each block node corresponds to at least one personal tag and / or corporate tag. It should be noted that each company has corresponding shareholders, so the present invention will fix the nature of the shareholders for classification.

[0028] If a block node is determined to have only a personal tag and no corresponding personal tags with any other block nodes, the corresponding block node is designated as the chain's termination node. It should be noted that in this case, the corresponding enterprise has only one or more natural individuals as shareholders, but these individuals are not shareholders of other enterprises, so the chain will not continue. Therefore, the present invention designates the corresponding block node as the chain's termination node. This means that the shareholder structure corresponding to this block is relatively simple.

[0029] If the individual tag of a block node matches the individual tags of other block nodes, the corresponding block node is used as the starting node of the chain. In this case, at least one natural person is a shareholder of multiple companies, and the chain may continue to extend. In this case, the present invention will use the corresponding block node as the starting node of the chain, and through the corresponding natural person, a combined chain will be formed between multiple companies.

[0030] If the enterprise tag of a block node matches that of another block node or another block node, the corresponding block node is used as the starting node of the chain. At this point, at least one enterprise is associated with the block node as a shareholder, and the chain may continue to extend. Therefore, the present invention uses the corresponding block node as the starting node of the chain.

[0031] In one possible implementation, if a block node is determined to have only a company tag and does not correspond to any other block node's company tags, the corresponding block node is designated as the chain's termination node. In this case, the shareholder company is no longer a shareholder of any other company, and the chain will not continue to extend. Therefore, the present invention designates the corresponding block node as the chain's termination node.

[0032] In one possible implementation, if the enterprise tag of a block node matches the enterprise tag of another block node, the corresponding block node is used as the chain starting node. In this case, the shareholder enterprise is also a shareholder of the block nodes corresponding to the other enterprises, and the chain continues to extend. Therefore, the present invention uses the corresponding block node as the chain starting node.

[0033] In one possible implementation, if a block node's personal tag matches another block node's personal tag, the corresponding block node is used as the chain's starting node. In this case, the natural person who is a shareholder is also a shareholder of the corresponding block node of another enterprise, and the chain continues to extend. Therefore, the present invention uses the corresponding block node as the chain's starting node.

[0034] In one possible implementation, the generating of the corresponding authority relationship chain by fusing and connecting corresponding block nodes based on shareholder labels and node types includes: If a block node is determined to have the same corporate tag as another block node, the two block nodes are connected, with the chain of block nodes with corresponding corporate tags arranged toward the corresponding block node. In this case, the two different block nodes have the same shareholder, and the present invention will connect the two corresponding block nodes. In this case, the two block nodes are connected based on the shareholder relationship.

[0035] If a block node's personal tag is determined to be the same as another block node's, a personal relay node is generated and connected to both block nodes, with the chains of the two block nodes oriented toward the corresponding personal relay node. At this point, a natural person acting as a shareholder of two companies would align the chains of the two block nodes toward the corresponding personal relay node, thus linking the two blocks based on the natural person.

[0036] After determining that any chain starting node has other chain starting nodes before it, the corresponding chain starting node will be used as a chain relay node. It should be noted that at this time, multiple small chains will be obtained. The small chains may be links between the block nodes of one enterprise and another enterprise, or they may be connected by natural persons to connect the block nodes of multiple different enterprises.

[0037] The chain start node, chain relay node, chain end node, and personal tag are integrated to obtain a permission relationship chain. The present invention will fully integrate the chain start node, chain relay node, chain end node, and personal tag of multiple block nodes with different attributes according to the attributes of each node. As a result, any nodes that directly or indirectly have a certain permission association relationship can be formed within the permission relationship chain provided by the present invention, making subsequent enterprise data analysis and processing more accurate.

[0038] In one possible implementation, the fusion of the chain start node, the chain relay node, the chain end node, and the personal tag to obtain the permission relationship chain includes: The nodes connected to each other are reconnected to obtain the permission relationship chain formed by all directly or indirectly connected nodes. The present invention will reorganize and merge all the scattered chains to obtain the merged permission relationship chain. This method counts all blocks with direct and indirect relationships.

[0039] Determine the node corresponding to the personal tag in the permission relationship chain to generate a personal relationship node, and extract the share information corresponding to each shareholder under the corresponding chain. The present invention will lock the personal node in the permission relationship chain and obtain the personal relationship node, effectively distinguishing all blocks and nodes in this way.

[0040] After receiving a data collection request from a first block node, the block server determines a first relationship chain, a first decomposition method, and a first calculation method based on the data collection request. In the technical solution provided by the present invention, each company or individual corresponds to a block node. After receiving a data collection request from a first block node, the present invention determines the corresponding first relationship chain, first decomposition method, and first calculation method. Different data collection requests and first block nodes result in different first relationship chains, first decomposition methods, and first calculation methods.

[0041] In a possible implementation, after receiving the data collection request of the first block node, the block server determines the first relationship chain, the first decomposition manner, and the first calculation manner based on the data collection request, including: The block server obtains the first relationship chain by calling the permission relationship chain corresponding to the first block node after receiving the data collection request of the first block node. It should be noted that the block server determines the first relationship chain corresponding to the first block node after receiving the data collection request. At this time, the first relationship chain is the chain of all nodes having a direct or indirect relationship with the first block node.

[0042] The attribute of the data collection request is extracted to determine the corresponding first decomposition manner. Different attributes have a preset first decomposition manner. Different attributes of the data collection request have different decomposition manners. For example, when the data collection request is for financial data collection, it can only collect information corresponding to the block of the direct or indirect shareholder at this time, and the decomposition is performed in this manner. For example, when the data collection request is for questionnaire collection, it can collect all non-natural person enterprises at this time, and another decomposition manner is used. The decomposition manner corresponding to the attribute of the data collection request can be preset for each decomposition manner.

[0043] The first calculation manner includes a combination of different machine learning algorithms, and different machine learning algorithms have different preset fusion weights. In the technical manner provided by the present application, there are also different combination manners. For example, when predicting the credit risk of an enterprise, the present application pre-trains multiple groups of machine learning algorithms, and then performs credit risk prediction according to the corresponding machine learning algorithms.

[0044] The first relationship chain is decomposed and processed based on the first decomposition manner to obtain a second relationship chain. Different second relationship chains are assembled to generate a fusion data chain. It should be noted that the first relationship chain is decomposed and processed to obtain the second relationship chain. For example, the first relationship chain is D→C→B→A→Zhang San. At this time, the second relationship chain is obtained by counting the companies with B as a direct or indirect shareholder. D→C→B→A→Zhang San is split into D→C→B.

[0045] The request instruction is generated and distributed to the fusion nodes in the fusion data chain and is fed back to the block server based on the hash encryption. The block server calculates the data processing index based on the first calculation manner. According to the technical solution provided by the present application, the request instruction is distributed to the fusion nodes in the fusion data chain, and then each fusion node feeds back to the block server based on the hash encryption for comprehensive calculation. For example, credit risk prediction. At this time, the feedback of the fusion node may be related financial data, operating data, etc. for credit risk assessment.

[0046] In a possible implementation, decomposing the first relationship chain based on the first decomposition method to obtain the second relationship chain, and assembling different second relationship chains to generate the fused data chain, includes: If it is determined that there are multiple first relationship chains, the first relationship chain with the most nodes is selected as the master chain and the other first relationship chains are selected as slave chains. The technical solution provided by the present invention decomposes the master-slave relationship of the chains according to the number of nodes, so that the data processing volume can be reduced when the chains are reorganized later.

[0047] The chain node corresponding to the first block node in the slave chain is removed as the decomposition point to obtain a decomposition chain, and the other chain nodes connected to the decomposition point in the decomposition chain are connected to the chain node corresponding to the first block node in the main chain to obtain a fused data chain.

[0048] In one possible implementation, the generation request instruction is distributed to the fusion node in the fusion data chain and fed back to the block server based on hash encryption. The block server obtains the data processing indicator after calculation based on the first calculation method, including: Determine the enterprise information of different chain dimensions corresponding to the data collection request, obtain the fusion data chain based on the enterprise information of different chain dimensions, decompose and perform hash calculation, and obtain the multi-level hash key of different fusion nodes. The technical solution provided by the present invention first generates corresponding chain slots according to the number of nodes in the chain, and generates a multidimensional code in the chain slot according to the order of each slot. The multidimensional code includes the node level dimension and the node number dimension. For example, there are three levels in a chain, then the level dimensions are A, B, and C respectively. The number of nodes in each level chain is 1, 1, and 2 respectively. Then the node number dimensions are 1, 1, and 2 respectively. The corresponding multidimensional code extracts each A1, B1, and C2. The present invention extracts the characters corresponding to the shareholder labels corresponding to the chain slots and fills the values ​​in each corresponding chain slot. For example, if the three shareholders are ABC respectively, the underlying characters used to generate the hash key are A1AB1BC2C. The present invention performs the first step of hash calculation on the underlying characters to obtain the first subkey. At this time, the first subkeys of all fusion nodes in the fusion data chain are the same.

[0049] In order to ensure the security and confidentiality of the data, the present invention will further extract the chain time and current time of the block node corresponding to each chain slot, split and reorganize the chain time and current time respectively to obtain another underlying character, and the splitting method is to cross-set the chain time in front and the current time in the back.

[0050] When cross setting is performed, the application generates a first time slot group corresponding to the uplink time and a second time slot group corresponding to the current time, and respectively sorts them, selects the slots with the same sequence number in the first time slot group and the second time slot group, reorganizes them in the manner of the uplink time first and the current time second to obtain a combined slot, and obtains a second sub-key through second hash calculation in the combined slot. The first sub-key is in front and the second sub-key is in back to generate a final hash key.

[0051] Based on the multi-level hash key, the enterprise information of the fusion nodes at different positions is encrypted, and each fusion node has different dimensions in the fusion data chain. The application encrypts the enterprise information of the fusion nodes at different positions according to the generated hash key, which is sent by the server based on the plug-in to the corresponding fusion node, thereby ensuring the encryption of the data.

[0052] Through the above multi-dimensional data acquisition, the randomness of the key is ensured, and the data traceability is achieved, greatly improving the key generation efficiency and effective feedback traceability, and ensuring the authenticity of the data.

[0053] After receiving the enterprise information, the block server inputs it into the combination of different machine learning algorithms, and calculates the data processing index based on different preset fusion weights. When performing machine learning algorithm calculation, different algorithms can be used for calculation, such as calculating the liabilities and revenues of enterprises. The single model can be used to calculate the liabilities and revenues of enterprises in the prior art, and the application will not be described in detail. After the single algorithm is calculated, the corresponding weight can be configured for each algorithm manually to obtain the fused prediction result.

[0054] Linear Regression: Linear regression is a supervised learning algorithm used to establish a linear relationship between variables. It predicts the value of one or more continuous variables by fitting the best line. Linear regression is widely used in sales trends, market forecasting, and economic analysis.

[0055] Logistic Regression: Logistic regression is a binary classification algorithm used to predict the probability of a discrete output variable. It is commonly used in risk assessment, marketing, and medical disease prediction.

[0056] Decision Trees: Decision trees are a tree-based supervised learning algorithm used for classification and regression problems. It divides the data set into different categories or values through a series of judgment nodes. Decision trees are easy to understand and interpret, and are commonly used in customer segmentation, fraud detection, and recommendation systems.

[0057] Random Forests: Random Forests is an ensemble learning algorithm that combines the predictions of multiple decision trees to improve accuracy and stability. It is suitable for applications such as large datasets, feature selection, and anomaly detection.

[0058] Support Vector Machines (SVM): A support vector machine is a supervised learning algorithm for classification and regression problems. It separates samples of different classes by constructing a hyperplane in the feature space. SVMs are widely used in fields such as text classification, image recognition, and bioinformatics.

[0059] K-Nearest Neighbors (KNN): The K-Nearest Neighbors algorithm is an instance-based supervised learning algorithm used for classification and regression problems. It makes predictions by calculating the distance between an example and its K nearest neighbors. KNN is commonly used in tasks such as recommender systems, image recognition, and pattern recognition.

[0060] Principal Component Analysis (PCA): PCA is an unsupervised learning algorithm used to reduce data dimensionality and extract key features. It projects the original data onto a new set of orthogonal variables through a linear transformation. PCA is widely used in data visualization, feature extraction, and anomaly detection.

[0061] Clustering: Clustering is an unsupervised learning algorithm used to group objects in a dataset into clusters with similar characteristics. Common clustering algorithms include K-means clustering and hierarchical clustering. Clustering is widely used in market segmentation, social network analysis, and image segmentation.

[0062] In practice, the weight of a particular machine learning algorithm in the overall dataset is also affected by other machine learning algorithms. Therefore, considering the differences in the weight of each machine learning algorithm in the overall dataset and the connections between them, a weight-optimized multi-machine learning combination strategy is proposed.

[0063] The technical solution provided by the present invention can first combine various machine learning algorithms into an algorithm matrix. According to the characteristics of each algorithm in the algorithm matrix, all algorithm vectors in the algorithm matrix are traversed. When traversing each algorithm vector, the classification accuracy of the algorithm matrix excluding the algorithm vector is calculated. The overall classification effect of the algorithm matrix can be regarded as 1, and each algorithm vector in the algorithm matrix can be regarded as a separate individual. After excluding each individual machine learning algorithm, the overall classification effect minus the classification effect of the remaining algorithms to form the algorithm matrix is ​​the weight of the machine learning algorithm. Therefore, this strategy uses the value of 1-accuracy as the weight of each machine learning algorithm.

[0064] In order to implement the enterprise data processing method based on blockchain of the present invention, the present invention also provides an enterprise data collection and processing system based on blockchain, such as Figure 2 Shown, including: A generation module, configured to enable the block server to analyze the enterprise attributes in the block node and generate a permission relationship chain corresponding to the block node. The permission relationship chain is a unidirectional attribute and the permissions are set from low to high; a determination module, configured to enable the block server to determine the first relationship chain, the first decomposition method, and the first calculation method based on the data collection request after receiving the data collection request from the first block node; a fusion module, configured to decompose the first relationship chain based on the first decomposition method to obtain a second relationship chain, and assemble different second relationship chains to generate a fused data chain; The calculation module is used to distribute the generated request instruction to the fusion node in the fusion data chain and feed it back to the block server based on hash encryption. The block server obtains the data processing indicator after calculation based on the first calculation method.

[0065] The present invention also provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the methods provided in the various embodiments described above.

[0066] The storage medium may be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of a computer program from one location to another. A computer storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, a storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium may also be an integral part of the processor. The processor and the storage medium may be located in an application-specific integrated circuit (ASIC). In addition, the ASIC may be located in a user device. Of course, the processor and the storage medium may also exist as discrete components in a communication device. The storage medium may be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0067] The present invention also provides a program product, which includes execution instructions stored in a storage medium. At least one processor of a device can read the execution instructions from the storage medium, and at least one processor executes the execution instructions so that the device implements the methods provided in the various embodiments described above.

[0068] In the above-mentioned terminal or server embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASICs), etc. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly executed by a hardware processor or by a combination of hardware and software modules within the processor.

[0069] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. The enterprise data processing method based on blockchain is characterized by: include: The block server analyzes the enterprise attributes in the block node and generates a permission relationship chain corresponding to the block node. The permission relationship chain is a one-way attribute and the permissions are set from low to high. After receiving the data collection request from the first block node, the block server determines the first relationship chain, the first decomposition method, and the first calculation method based on the data collection request; Decomposing the first relationship chain based on the first decomposition method to obtain a second relationship chain, and assembling different second relationship chains to generate a fused data chain; Generate a request instruction, distribute it to the fusion node in the fusion data chain, and feed it back to the block server based on hash encryption. The block server obtains the data processing indicator after calculation based on the first calculation method.

2. The enterprise data processing method based on blockchain according to claim 1 is characterized in that: The block server analyzes the enterprise attributes in the block node and generates a permission relationship chain corresponding to the block node. The permission relationship chain is a one-way attribute and the permissions are set from low to high, including: Extract the shareholder information of each block node's enterprise attributes, and generate the shareholder label and node type of each block node based on the shareholder information; Based on shareholder labels and node types, the corresponding block nodes are connected and integrated to generate the corresponding permission relationship chain.

3. The enterprise data processing method based on blockchain according to claim 1, characterized in that: The step of extracting shareholder information of the enterprise attributes of each block node and generating a shareholder label of each block node according to the shareholder information includes: The shareholder tags are classified into personal tags and corporate tags. Each block node corresponds to at least one personal tag and / or corporate tag. If it is determined that a block node only has a personal tag and does not correspond to the personal tags of other block nodes, the corresponding block node will be used as the chain termination node; If the personal tag of a block node is determined to correspond to the personal tags of other block nodes, the corresponding block node will be used as the starting node of the chain; If the enterprise tag of a block node is determined to correspond to other block nodes or other block node enterprise tags, the corresponding block node will be used as the starting node of the chain.

4. The enterprise data processing method based on blockchain according to claim 2, characterized in that: If it is determined that a block node only has an enterprise tag and does not correspond to the enterprise tags of other block nodes, the corresponding block node will be used as the chain termination node; If the enterprise label of a block node is determined to correspond to the enterprise label of another block node, the corresponding block node will be used as the starting node of the chain; If the personal tag of the block node is determined to correspond to the personal tags of other block nodes, the corresponding block node will be used as the starting node of the chain.

5. The enterprise data processing method based on blockchain according to any one of claims 3 or 4, characterized in that: The generation of the corresponding authority relationship chain by fusing and connecting the corresponding block nodes based on the shareholder labels and node types includes: If it is determined that a block node has the same enterprise tag as another block node, the two block nodes are connected, and the chain of block nodes with corresponding enterprise tags is set towards the corresponding block node; If it is determined that the personal tag of a block node is the same as the personal tag of another block node, a personal relay node is generated, and the personal relay node is connected to the two block nodes respectively, and the chains of the two block nodes are directed towards the corresponding personal relay node; After determining that there are other chain starting nodes in front of any chain starting node, the corresponding chain starting node is used as a chain relay node; The chain start node, chain relay node, chain end node and personal label are integrated to obtain the permission relationship chain.

6. The enterprise data processing method based on blockchain according to claim 5 is characterized in that: The permission relationship chain is obtained by fusing the chain start node, chain relay node, chain end node and personal label, including: Perform secondary connections on the nodes that are connected to each other to obtain the permission relationship chain formed by all directly or indirectly connected nodes; Determine the node corresponding to the personal label in the authority relationship chain to generate a personal relationship node, and extract the share information corresponding to each shareholder under the corresponding chain.

7. The enterprise data processing method based on blockchain according to claim 1, characterized in that: After receiving the data collection request from the first block node, the block server determines a first relationship chain, a first decomposition method, and a first calculation method based on the data collection request, including: Upon receiving the data collection request from the first block node, the block server retrieves the permission relationship chain corresponding to the first block node to obtain a first relationship chain; Extracting the attributes of the data collection request to determine the corresponding first decomposition method, where different attributes have preset first decomposition methods; The first calculation method includes a combination of different machine learning algorithms, and different machine learning algorithms have different preset fusion weights.

8. The enterprise data processing method based on blockchain according to claim 7 is characterized in that: The decomposing the first relationship chain based on the first decomposition method to obtain the second relationship chain, and assembling different second relationship chains to generate a fused data chain, including: If it is determined that there are multiple first relationship chains, the first relationship chain with the most nodes is selected as the main chain and the other first relationship chains are selected as slave chains; The chain node corresponding to the first block node in the slave chain is removed as the decomposition point to obtain a decomposition chain, and the other chain nodes connected to the decomposition point in the decomposition chain are connected to the chain node corresponding to the first block node in the main chain to obtain a fused data chain.

9. The enterprise data processing method based on blockchain according to claim 8, characterized in that: The generation request instruction is distributed to the fusion node in the fusion data chain and fed back to the block server based on hash encryption. The block server obtains the data processing index after calculation based on the first calculation method, including: Determine the enterprise information of different chain dimensions corresponding to the data collection request, obtain the fused data chain based on the enterprise information of different chain dimensions, decompose and perform hash calculation, and obtain the multi-level hash keys of different fusion nodes; Encrypt enterprise information of fusion nodes at different locations based on multi-level hash keys. Each fusion node has different dimensions within the fusion data chain. After receiving the enterprise information, the block server inputs it into a combination of different machine learning algorithms, and obtains data processing indicators through weighted fusion calculation based on different preset fusion weights.

10. The enterprise data collection and processing system based on blockchain is characterized by: include: A generation module, configured to enable the block server to analyze the enterprise attributes in the block node and generate a permission relationship chain corresponding to the block node. The permission relationship chain is a unidirectional attribute and the permissions are set from low to high; a determination module, configured to enable the block server to determine the first relationship chain, the first decomposition method, and the first calculation method based on the data collection request after receiving the data collection request from the first block node; a fusion module, configured to decompose the first relationship chain based on the first decomposition method to obtain a second relationship chain, and assemble different second relationship chains to generate a fused data chain; The calculation module is used to distribute the generated request instruction to the fusion node in the fusion data chain and feed it back to the block server based on hash encryption. The block server obtains the data processing indicator after calculation based on the first calculation method.

Images