A blockchain feedback-based integral intelligent management method and system

Abstract

This invention discloses a blockchain-based intelligent points management method and system, relating to the field of blockchain technology. The method includes acquiring full data of platform transaction orders, points distribution data, service fee collection data, and user identification data to generate a multi-source original business dataset. Based on this dataset, correlation analysis is performed to generate a key business dataset. This invention achieves accurate mining of points correlation features and efficient filtering of key business data through Pearson correlation coefficient calculation and correlation topology graph construction. It achieves low-cost storage of full business data and hash binding of core data through off-chain distributed storage and hash generation. It achieves tamper-proof data storage and full traceability through consortium blockchain consensus verification and on-chain writing of business blocks. Finally, it achieves multi-dimensional and efficient indexing of points order information through combined query condition filtering and conditional hash mapping, improving points query efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of blockchain, and particularly to an intelligent management method and system for points based on blockchain feedback. Background Art

[0002] Points management is the core business link of membership services, consumer e-commerce, platform operations, and digital rights systems. The authenticity, traceability, and security of points issuance, cancellation, transfer, accounting, and query directly relate to the platform's credibility, user rights protection, and transaction compliance. Blockchain technology has the characteristics of decentralization, immutability, consensus credibility, and full-process traceability. Intelligent management of points based on blockchain feedback can achieve reliable digital evidence storage and closed-loop feedback of all business data such as transaction orders, points changes, and service fee accounting, prevent points data tampering, forgery, and reconciliation distortion from the source, and at the same time enable rapid verification, accurate traceability, and efficient query of points data, which has important engineering and application value for protecting the legitimate rights and interests of users and merchants, improving the efficiency of platform operation management, meeting regulatory and auditing requirements, and promoting the development of the points business ecosystem towards credibility, intelligence, and standardization.

[0003] Currently, there are still problems in points management technology, such as the inability to accurately analyze order data, the inability to accurately identify key data, redundant data occupying a large amount of storage resources, seriously affecting processing and query efficiency, adopting a centralized storage or blindly full-chain-on mode, the data on and off the chain being disjointed, lacking a secure binding and efficient indexing mechanism, high storage costs and difficult data retrieval, business data not being stored with blockchain consensus, core data such as points changes and transaction accounting being easily tampered with, no reliable digital evidence and auditing basis, difficult traceability and evidence collection for rights disputes, a single and fixed query method, the query of blockchain being a pull mode rather than the index query mode of a database. To find the record of a specific order number, the smart contract usually needs to traverse all relevant events, which is catastrophic when the data volume is large and greatly affects the query efficiency of points. Summary of the Invention

[0004] To address the aforementioned technical problems, this paper provides a blockchain-based intelligent points management method and system. This solution resolves the issues raised in the background, such as the inability to accurately analyze order data, the inability to accurately identify key data, redundant data consuming significant storage resources and severely impacting processing and query efficiency, the use of centralized storage or blind full-scale on-chain storage leading to data disconnect between on-chain and off-chain, the lack of secure binding and efficient indexing mechanisms, high storage costs and difficulty in data retrieval, the lack of blockchain consensus-based notarization of business data, the vulnerability of core data such as points changes and transaction calculations to tampering, the lack of reliable notarization and auditing evidence, the difficulty in tracing and obtaining evidence for rights disputes, and the single, rigid query method. Blockchain queries are pull-based rather than database index-based; to find a record with a specific order number, smart contracts typically need to traverse all relevant events, which is disastrous with large datasets and significantly impacts points query efficiency.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A blockchain-based intelligent points management method includes: Acquire full data of platform transaction orders, points distribution data, service fee collection data, and user identification data to generate multi-source raw business datasets; Based on the correlation analysis of multi-source original business datasets, key business datasets are generated. Based on the multi-source original business dataset and key business dataset, and distinguishing data storage, data address binding information is obtained. The data address binding information includes the off-chain data index address and the business evidence hash value. Based on the key business dataset, after reaching a consensus on the blockchain, the data is stored on the blockchain and the blockchain storage information is obtained. A query index table is constructed based on the blockchain storage information and data address binding information; Based on the user-input combined query conditions, the system matches the transaction hash value and the off-chain data index address from the query index table. The corresponding business evidence hash value is retrieved from the blockchain based on the transaction hash value, and then hash-verified with the data corresponding to the off-chain data index address to realize the query of order points information.

[0006] Preferably, the step of acquiring full data of platform transaction orders, points distribution data, service fee collection data, and user identification data to generate a multi-source original business dataset specifically includes: Based on the platform's open interface, obtain the order data synchronization authorization certificate and establish a data transmission channel; According to the data transmission channel, the platform collects all transaction order data in real time. The platform transaction order data includes order number information, transaction amount data, product code information, transaction time information, user unique identifier information, and merchant identifier information. Based on the preset service fee calculation rules, service fee data is obtained using the full data of platform transaction orders. The service fee data includes platform service fee rate parameters and actual service fee received data. Based on the automatic points distribution rules, points distribution data is obtained using the full data of platform transaction orders. The points distribution data includes the points distribution base, distribution ratio, points change value, and points account identification data. Based on the user's unique identifier, obtain user identity data, points account status data, and historical transaction data; Using the order number as the unique primary key, the platform's full transaction order data, service fee data, and points distribution data are initially concatenated to generate concatenated business data; Based on the data collection timestamp, a unique time identifier is added to each piece of spliced ​​business data to distinguish multiple data synchronization requests for the same order; Based on the splicing of business data and time stamps, abnormal data such as duplicate collections and unconfirmed data that have exceeded the time limit are removed to generate cleaned business data; Based on the cleaned business data, it is grouped and aggregated by order number to ensure that the same order corresponds to a unique set of business data combinations. Based on the grouped and aggregated business data, a multi-source original business dataset is generated.

[0007] Preferably, the step of performing correlation analysis based on multi-source original business datasets to generate key business datasets specifically includes: Based on the multi-source original business dataset, extract the data dimension set, including order number, transaction amount, transaction time, user identifier, service fee amount, and points change value; By using the numerical change of integrals as the target correlation variable and each data dimension as a feature variable, a variable correlation analysis matrix is ​​constructed. Based on the Pearson correlation coefficient formula, the correlation coefficient between each feature variable and the integral change value is calculated; Based on the need for data correlation analysis, a correlation threshold is set; Based on the correlation coefficients and correlation thresholds corresponding to the feature variables, obtain the associated feature variables; Based on the multi-source original business dataset, the corresponding related business data is extracted based on the related feature variables; Based on the related business data, the data is reordered by order number to generate a sequence of key business data. Based on key business data sequences, these are marked as core related data; Based on core related data, generate key business datasets.

[0008] Preferably, obtaining the associated feature variables based on the correlation coefficients and correlation thresholds corresponding to the feature variables specifically includes: Based on the set of data dimensions, each data dimension is used as a topology node to obtain the data topology node set; Based on the correlation coefficients corresponding to the feature variables, the maximum correlation coefficient is taken as the basic correlation coefficient, and the feature variable corresponding to the maximum correlation coefficient is taken as the initial feature variable; Based on the multi-source original business dataset, any two feature variables are used as a feature variable group; The multi-source original business dataset is traversed, and the feature correlation coefficients corresponding to the two feature variables in each feature variable group are obtained based on the Pearson correlation coefficient. Based on the correlation threshold, the feature variable group is screened, and the feature variable group whose correlation coefficient exceeds the correlation threshold is used as the benchmark feature variable group. Based on the baseline feature variable set, construct the association edge between two feature variables in the baseline feature variable set and obtain the association edge information; Construct an associated topology graph based on the data topology node set and associated edge information; The topology node corresponding to the initial feature variable is taken as the starting topology node, and the topology node corresponding to the target association variable is taken as the ending topology node. Based on the associated topology graph, a topology transmission chain is obtained using the starting topology node and the ending topology node as the basis. The topology transmission chain represents the topology node route from the starting topology node to the ending topology node through associated edges. The product of the feature correlation coefficients of adjacent topological nodes in each topological transit chain is used as the baseline correlation coefficient of that topological transit chain. The difference between the baseline correlation coefficient and the basic correlation coefficient is used as the correlation propagation bias coefficient; The topology transit chain corresponding to the minimum value of the correlation transit deviation coefficient is taken as the topology transit feature chain; The feature variables corresponding to the topology nodes in the topology transitive feature chain are used as associated feature variables.

[0009] Preferably, the step of obtaining data address binding information based on data storage differentiation from multi-source original business datasets and key business datasets specifically includes: Based on the multi-source original business dataset, it is divided and encapsulated according to the order number to generate independent data blocks; Based on IPFS distributed storage, upload data blocks to off-chain storage nodes and obtain the node storage return value; Based on the node storage return value, a unique index address for off-chain data is generated, namely the IPFS hash address; Based on the key business dataset, the core related data is concatenated into a plaintext business string; Based on the SHA-256 hash algorithm, a business evidence hash value is generated by calculating the plaintext business string. Based on the business evidence hash value, verify the uniqueness of the hash value to ensure that the same order corresponds to a unique hash value and there is no hash collision; A two-way binding relationship is established based on the off-chain data index address and the business evidence hash value to generate data address binding information.

[0010] Preferably, the step of storing the key business dataset on the blockchain after reaching a consensus and obtaining the blockchain storage information specifically includes: Based on key business datasets and business evidence hash values, assemble the blockchain on-chain data body; Based on the on-chain data body, calculate the overall data hash value and generate block header verification information; The on-chain data body and block header verification information are encapsulated into a business block awaiting consensus. Based on the blockchain consensus mechanism, obtain the information of the consortium chain nodes in the blockchain, and broadcast the business blocks to be reached for consensus to all consortium chain nodes. Based on the node verification rules corresponding to the consortium blockchain nodes, independently verify the integrity of the on-chain data body, the legality of the hash value, and the uniqueness of the order number; Based on the node verification results, the ratio of the number of nodes that pass verification to the total number of nodes in the consortium blockchain is used as the consensus pass rate. The consensus status of a block is determined based on the consensus pass rate. If the consensus of a block is determined to be passed, the data uploaded to the chain is confirmed to be legal and valid. Based on the business block passed by consensus, an on-chain write operation is performed to obtain blockchain storage information, which includes block height, transaction hash value, and on-chain evidence storage address.

[0011] Preferably, the step of constructing a query index table based on blockchain storage information and data address binding information specifically includes: Based on the data dimension set, the query dimensions are determined according to the order number, user ID, transaction amount, transaction time, points change value, and service fee amount; Combine the order number with any one query dimension, and the user identifier with any two query dimensions to generate basic combination information for query conditions; Based on the query condition combination information and the topology transfer chain, the topology transfer chain containing the topology nodes corresponding to the query dimensions in each basic combination of query conditions is taken as the query topology transfer chain for that query condition combination. Retrieve the topology transit sub-chains between topology nodes corresponding to the query dimension in the query topology transit chain; The product of the feature correlation coefficients corresponding to adjacent topological nodes in the topological transitive subchain is taken as the combined correlation coefficient of the topological transitive subchain. The difference between the feature correlation coefficient corresponding to the query dimension in the basic combination of query conditions and the combination correlation coefficient is used as the combination deviation coefficient; Based on the combination deviation coefficient of the topology transit subchain corresponding to the basic combination of query conditions, the combination identification degree of the basic combination of query conditions is obtained. Based on the combination recognition, the basic combinations of query conditions are filtered to obtain a set of combined query conditions; Based on the set of combined query conditions, data address binding information, and transaction hash value, the combined query conditions are bound to the off-chain data index address, business notarization hash value, and transaction hash value to establish a condition hash mapping relationship; A query index table is generated based on the conditional hash mapping relationship.

[0012] Furthermore, a blockchain-based intelligent points management system is proposed to implement the intelligent management method described above, including: The main control module is used to store key business datasets on the blockchain after consensus, obtain blockchain storage information, obtain data address binding information based on data storage differentiation based on multi-source original business datasets and key business datasets, and construct a query index table based on blockchain storage information and data address binding information. The information acquisition module is used to obtain order data synchronization authorization credentials based on the platform's open interface, establish a data transmission channel, collect all transaction order data on the platform in real time according to the data transmission channel, obtain user identity data, points account status data and historical transaction association data based on the user's unique identifier information, and obtain the combined query conditions input by the user. The topology analysis module is used to calculate the correlation coefficient between each feature variable and the integral change value based on the Pearson correlation coefficient formula, and to obtain the associated feature variables based on the correlation coefficient and correlation threshold corresponding to the feature variables. The display module interacts with the main control module and is used to output and display multi-source original business datasets, key business datasets, data address binding information, blockchain storage information, and query index tables.

[0013] Optionally, the main control module specifically includes: The control unit is used to obtain data address binding information based on data storage differentiation according to the multi-source original business dataset and key business dataset, and to construct a query index table based on the blockchain storage information and data address binding information; An information receiving unit interacts with an information acquisition module and a topology analysis module to receive data and transmit it to a control unit. A blockchain storage unit is used to store key business datasets on the blockchain after consensus and to obtain blockchain storage information.

[0014] Optionally, the information acquisition module specifically includes: The first acquisition unit is used to acquire order data synchronization authorization credentials based on the platform's open interface, establish a data transmission channel, and collect all transaction order data of the platform in real time according to the data transmission channel. The second acquisition unit is used to acquire user identity data, points account status data and historical transaction association data based on the user's unique identifier information, and to acquire the combined query conditions input by the user.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention proposes a blockchain-based intelligent points management method and system. Through Pearson correlation coefficient calculation and association topology graph construction, it achieves accurate mining of points association features and efficient screening of key business data. Through off-chain distributed storage and hash generation, it achieves low-cost storage of all business data and hash binding of core data, providing storage support for trusted blockchain evidence preservation. Through consortium blockchain consensus verification and on-chain writing of business blocks, it achieves tamper-proof evidence preservation and full traceability of key points business data, providing reliable assurance for compliance auditing of points management. Through combined query condition filtering and conditional hash mapping construction, it achieves multi-dimensional and efficient indexing of points order information, improving points query efficiency. Attached Figure Description

[0016] Figure 1 This is a flowchart of a blockchain-based intelligent points management method proposed in this invention. Figure 2 This is a flowchart illustrating the process of acquiring multi-source original business datasets in this invention. Figure 3 This is a flowchart illustrating the process of acquiring key business datasets in this invention. Figure 4 This is a flowchart illustrating the process of obtaining data address binding information in this invention. Figure 5 This is a block diagram of an intelligent points management system based on blockchain feedback proposed in this invention. Detailed Implementation

[0017] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0018] Reference Figure 1 - Figure 4As shown in the figure, an embodiment of the present invention provides a point-based intelligent management method based on blockchain feedback, comprising: Acquire full data of platform transaction orders, points distribution data, service fee collection data, and user identification data to generate multi-source raw business datasets; Specifically, this involves acquiring full data on platform transaction orders, points distribution data, service fee collection data, and user identification data to generate a multi-source raw business dataset, including: Based on the platform's open interface, obtain the order data synchronization authorization certificate and establish a data transmission channel; According to the data transmission channel, the platform collects all transaction order data in real time. The platform transaction order data includes order number information, transaction amount data, product code information, transaction time information, user unique identifier information, and merchant identifier information. Based on the preset service fee calculation rules, service fee data is obtained using the full data of platform transaction orders. The service fee data includes platform service fee rate parameters and actual service fee received data. Based on the automatic points distribution rules, points distribution data is obtained using the full data of platform transaction orders. The points distribution data includes the points distribution base, distribution ratio, points change value, and points account identification data. Based on the user's unique identifier, obtain user identity data, points account status data, and historical transaction data; Using the order number as the unique primary key, the platform's full transaction order data, service fee data, and points distribution data are initially concatenated to generate concatenated business data; Based on the data collection timestamp, a unique time identifier is added to each piece of spliced ​​business data to distinguish multiple data synchronization requests for the same order; Based on the splicing of business data and time stamps, abnormal data such as duplicate collections and unconfirmed data that have exceeded the time limit are removed to generate cleaned business data; Based on the cleaned business data, it is grouped and aggregated by order number to ensure that the same order corresponds to a unique set of business data combinations. Based on the grouped and aggregated business data, a multi-source original business dataset is generated.

[0019] This solution establishes a data transmission channel based on the platform's open interfaces and authorization credentials, enabling the legal and real-time collection of comprehensive business data, including orders, points, service fees, and user identities. This solves the problems of fragmented, delayed, and insufficiently compliant data collection. Using the order number as the unique primary key, it achieves unified association of multi-source data, integrating transaction, service fee, points, and user data around the order number to form a complete business data chain. Adding a collection timestamp to each data entry effectively distinguishes multiple synchronization requests for the same order, providing a clear basis for subsequent abnormal data removal, avoiding duplicate data processing, and automatically removing duplicate collections and unconfirmed timeout data. This significantly improves the purity, accuracy, and usability of the dataset, preventing invalid data from interfering with subsequent analysis and evidence preservation. The cleaned data is grouped and aggregated by order number, ensuring that one order corresponds to only one complete set of business data, eliminating data redundancy and matching errors, and guaranteeing rigorous data logic.

[0020] In this embodiment, the service fee calculation rules and the automatic points distribution rules can be set according to requirements.

[0021] Based on the correlation analysis of multi-source original business datasets, key business datasets are generated. Specifically, correlation analysis is performed on multi-source original business datasets to generate key business datasets, including: Based on the multi-source original business dataset, extract the data dimension set, including order number, transaction amount, transaction time, user identifier, service fee amount, and points change value; By using the numerical change of integrals as the target correlation variable and each data dimension as a feature variable, a variable correlation analysis matrix is ​​constructed. Based on the Pearson correlation coefficient formula, the correlation coefficient between each feature variable and the integral change value is calculated; Based on the need for data correlation analysis, a correlation threshold is set; Based on the correlation coefficients and correlation thresholds corresponding to the feature variables, obtain the associated feature variables; Based on the multi-source original business dataset, the corresponding related business data is extracted based on the related feature variables; Based on the related business data, the data is reordered by order number to generate a sequence of key business data. Based on key business data sequences, these are marked as core related data; Based on core related data, generate key business datasets.

[0022] This solution extracts core data dimensions such as orders, transactions, users, and service fees. A correlation analysis matrix is ​​constructed using points changes as the target variable to quantitatively represent the relationship between business dimensions and points changes. The core influencing factors of points changes are identified, and the Pearson correlation coefficient is used to calculate the correlation between each feature variable and the points change value, objectively quantifying the strength of the correlation between data dimensions. Highly correlated feature variables are selected based on correlation thresholds, and redundant dimensions unrelated to points changes are eliminated, significantly reducing data volume and improving the efficiency of subsequent data processing, storage, and querying. Related business data is reordered by order number to ensure a regular data sequence and that each order corresponds to a unique associated data, eliminating the risk of data corruption and ensuring the protection of critical data. Consistency and traceability are ensured by outputting a critical business dataset containing only core related data. This provides accurate and efficient data support for subsequent off-chain storage, hash generation, blockchain on-chain processing, and query index construction, improving the efficiency of the entire points management process. It locks in business data strongly correlated with points changes, enabling subsequent evidence storage, query, and verification processes to focus on core business information and ensuring the accuracy and relevance of intelligent points management. It is understood that the Pearson correlation coefficient formula is a well-known formula, so it will not be described in detail here. Following the internationally accepted statistical classification standard for Pearson correlation coefficient, the correlation threshold is 0.85, which belongs to the strong linear correlation range and can accurately identify characteristic variables that have a stable quantitative correlation with points changes.

[0023] Specifically, based on the correlation coefficients and correlation thresholds corresponding to the feature variables, associated feature variables are obtained, including: Based on the set of data dimensions, each data dimension is used as a topology node to obtain the data topology node set; Based on the correlation coefficients corresponding to the feature variables, the maximum correlation coefficient is taken as the basic correlation coefficient, and the feature variable corresponding to the maximum correlation coefficient is taken as the initial feature variable; Based on the multi-source original business dataset, any two feature variables are used as a feature variable group; The multi-source original business dataset is traversed, and the feature correlation coefficients corresponding to the two feature variables in each feature variable group are obtained based on the Pearson correlation coefficient. Based on the correlation threshold, the feature variable group is screened, and the feature variable group whose correlation coefficient exceeds the correlation threshold is used as the benchmark feature variable group. Based on the baseline feature variable set, construct the association edge between two feature variables in the baseline feature variable set and obtain the association edge information; Construct an associated topology graph based on the data topology node set and associated edge information; The topology node corresponding to the initial feature variable is taken as the starting topology node, and the topology node corresponding to the target association variable is taken as the ending topology node. Based on the associated topology graph, a topology transmission chain is obtained using the starting topology node and the ending topology node as the basis. The topology transmission chain represents the topology node route from the starting topology node to the ending topology node through associated edges. The product of the feature correlation coefficients of adjacent topological nodes in each topological transit chain is used as the baseline correlation coefficient of that topological transit chain. The difference between the baseline correlation coefficient and the basic correlation coefficient is used as the correlation propagation bias coefficient; The topology transit chain corresponding to the minimum value of the correlation transit deviation coefficient is taken as the topology transit feature chain; The feature variables corresponding to the topology nodes in the topology transitive feature chain are used as associated feature variables.

[0024] This solution models data dimensions using topological methods, transforming business features into topological nodes and related edges. This achieves a structured and computable expression of integral relationships, providing a standardized analytical framework for in-depth feature correlation mining. The solution anchors the starting feature with the highest correlation coefficient, accurately identifying the core dimension most relevant to integral changes. This ensures that correlation analysis starts from key features, improving mining efficiency and directionality. It iterates through feature groups, calculates correlations, and uses threshold filtering to automatically identify strongly correlated feature combinations and eliminate weakly correlated redundant combinations, significantly reducing the scope of invalid analysis and improving the efficiency of correlated feature extraction. A correlation topology graph is constructed, presenting the correlation transmission paths between dimensions, clearly restoring the inherent transmission logic of business data and integral changes. The deviation coefficient of the topology transmission chain is quantified, and the optimal transmission chain is selected based on the minimum deviation, ensuring that the extracted correlated features are most representative and have the lowest transmission error. Based on the optimal topology chain, correlated features are determined, achieving accurate selection of integral-related features. This significantly improves the purity and effectiveness of key business data, building a solid foundation of high-quality features for subsequent hash storage, on-chaining, and indexing.

[0025] Based on the multi-source original business dataset and key business dataset, and distinguishing data storage, data address binding information is obtained. The data address binding information includes the off-chain data index address and the business evidence hash value. Specifically, based on the multi-source original business dataset and the key business dataset, and distinguishing them by data storage, data address binding information is obtained, including: Based on the multi-source original business dataset, it is divided and encapsulated according to the order number to generate independent data blocks; Based on IPFS distributed storage, upload data blocks to off-chain storage nodes and obtain the node storage return value; Based on the node storage return value, a unique index address for off-chain data is generated, namely the IPFS hash address; Based on the key business dataset, the core related data is concatenated into a plaintext business string; Based on the SHA-256 hash algorithm, a business evidence hash value is generated by calculating the plaintext business string. Based on the business evidence hash value, verify the uniqueness of the hash value to ensure that the same order corresponds to a unique hash value and there is no hash collision; A two-way binding relationship is established based on the off-chain data index address and the business evidence hash value to generate data address binding information.

[0026] This solution encapsulates data in blocks according to order numbers, enabling modular and independent storage of multi-source business data. This solves the problems of disorganized centralized storage and difficult retrieval of full data, improving the regularity of data management. IPFS distributed storage is used to generate unique off-chain data index addresses, significantly reducing on-chain storage pressure and costs while ensuring secure storage of all business data. A unique business evidence hash is generated using the SHA-256 algorithm, and hash uniqueness verification is completed to ensure that critical business data is tamper-proof and conflict-free, providing accurate hash basis for trusted blockchain evidence storage. A two-way binding between the off-chain index address and the business evidence hash is established, achieving precise association between off-chain full data and on-chain hash evidence storage. This establishes a collaborative link between data storage and trusted verification. A differentiated storage mode of "full data off-chain storage, key data hash on-chain" is adopted, balancing data integrity and efficient blockchain evidence storage needs. This adapts to the lightweight and highly reliable storage requirements of points management business, forming standardized data address binding information. This provides a unified association basis for subsequent blockchain on-chain, query index construction, and hash verification queries, improving the standardization and traceability of the entire points management process.

[0027] Based on the key business dataset, after reaching a consensus on the blockchain, the data is stored on the blockchain and the blockchain storage information is obtained. Specifically, based on key business datasets, after reaching a consensus on the blockchain, the data is stored on the blockchain to obtain blockchain storage information, including: Based on key business datasets and business evidence hash values, assemble the blockchain on-chain data body; Based on the on-chain data body, calculate the overall data hash value and generate block header verification information; The on-chain data body and block header verification information are encapsulated into a business block awaiting consensus. Based on the blockchain consensus mechanism, obtain the information of the consortium chain nodes in the blockchain, and broadcast the business blocks to be reached for consensus to all consortium chain nodes. Based on the node verification rules corresponding to the consortium blockchain nodes, independently verify the integrity of the on-chain data body, the legality of the hash value, and the uniqueness of the order number; Based on the node verification results, the ratio of the number of nodes that pass verification to the total number of nodes in the consortium blockchain is used as the consensus pass rate. The consensus status of a block is determined based on the consensus pass rate. If the consensus of a block is determined to be passed, the data uploaded to the chain is confirmed to be legal and valid. Based on the business block passed by consensus, an on-chain write operation is performed to obtain blockchain storage information, which includes block height, transaction hash value, and on-chain evidence storage address.

[0028] This solution assembles the on-chain data and generates block header verification information, achieving complete encapsulation and verifiability of the on-chain data. This provides a basic verification basis for data tamper-proofing. The consortium blockchain broadcasts data across the entire network and multiple nodes independently verify the data, simultaneously verifying data integrity, hash validity, and order number uniqueness. This eliminates data forgery and duplicate on-chain issues from the source. Block validity is determined based on consensus pass rate. The decentralized consortium blockchain consensus mechanism ensures the credibility of on-chain data, avoiding single-point manipulation and data tampering risks. After consensus is passed, on-chain writing is performed to permanently store key points business data, generating blockchain storage information including block height, transaction hash, and storage address. This ensures data immutability and full traceability, establishing a credible on-chain storage benchmark. This provides authoritative and credible on-chain data support for subsequent query index construction, on-chain and off-chain hash verification, and points rights traceability auditing. The entire process is standardized and consensus-based on-chain, with an objective, unified, and reproducible process, meeting the compliance and credible storage requirements of points management business.

[0029] In this embodiment, when determining the block consensus status, if the consensus pass rate... If the consensus is passed, it can be understood that the blockchain adopts a consortium blockchain architecture, and all nodes are trusted nodes (third-party platforms, points operators, regulatory nodes, and cooperative institutions), and the consensus mechanism is Practical Byzantine Fault Tolerance (PBFT). The theoretical fault tolerance limit of the PBFT algorithm is: the proportion of faulty / malicious nodes ≤ 1 / 3 (≈33.3%), therefore, a consensus pass rate is set. That is, the percentage of faulty / offline nodes. To avoid the invalidation of evidence due to node failure or malicious tampering, the overall data hash value uses the SHA-256 hash algorithm.

[0030] A query index table is constructed based on the blockchain storage information and data address binding information; Specifically, a query index table is constructed based on blockchain storage information and data address binding information, including: Based on the data dimension set, the query dimensions are determined according to the order number, user ID, transaction amount, transaction time, points change value, and service fee amount; Combine the order number with any one query dimension, and the user identifier with any two query dimensions to generate basic combination information for query conditions; Based on the query condition combination information and the topology transfer chain, the topology transfer chain containing the topology nodes corresponding to the query dimensions in each basic combination of query conditions is taken as the query topology transfer chain for that query condition combination. Retrieve the topology transit sub-chains between topology nodes corresponding to the query dimension in the query topology transit chain; The product of the feature correlation coefficients corresponding to adjacent topological nodes in the topological transitive subchain is taken as the combined correlation coefficient of the topological transitive subchain. The difference between the feature correlation coefficient corresponding to the query dimension in the basic combination of query conditions and the combination correlation coefficient is used as the combination deviation coefficient; Based on the combination deviation coefficient of the topology transit subchain corresponding to the basic combination of query conditions, the combination identification degree of the basic combination of query conditions is obtained. Based on the combination recognition, the basic combinations of query conditions are filtered to obtain a set of combined query conditions; Based on the set of combined query conditions, data address binding information, and transaction hash value, the combined query conditions are bound to the off-chain data index address, business notarization hash value, and transaction hash value to establish a condition hash mapping relationship; A query index table is generated based on the conditional hash mapping relationship.

[0031] This solution determines query dimensions based on core business data, comprehensively covering key retrieval elements such as orders, users, points, and service fees. It provides a standardized, multi-dimensional retrieval foundation for points queries, generating differentiated query combinations based on order number and user identifier to adapt to different users' retrieval habits and business scenarios. This significantly improves the flexibility and adaptability of query methods. Combined with a topological transmission chain, query combinations are quantitatively evaluated. The combination identification degree is accurately calculated through combination correlation coefficients and deviation coefficients, automatically eliminating redundant and inefficient query conditions. A two-way mapping between query conditions and hash data is constructed, precisely binding combined query conditions to off-chain index addresses, business evidence hashes, and transaction hashes. This establishes a link between retrieval and evidence storage, generating a structured query index table. This enables rapid matching and location of combined query conditions, significantly improving the retrieval efficiency and response speed of order points information. Quantitatively filtering high-identity query combinations ensures the rationality and accuracy of query logic, avoids invalid searches, and enhances the convenience and intelligence of points queries.

[0032] In this embodiment, the formula for calculating the combined recognition score is as follows: In the formula, To enhance recognizability, This is the combination deviation coefficient. This refers to the number of query dimensions included in the basic combination of query conditions. Indicates the first The distinctiveness weight value of each query dimension. In the topological transitive subchain, the first The feature correlation coefficients corresponding to the associated edges The total number of associated edges in the topological transitive subchain. This represents the combined correlation coefficient of the topological transitive subchains.

[0033] The ratio of the number of topology transit chains containing the topology nodes corresponding to the query dimension to the total number of topology transit chains is used as the benchmark weight of the query dimension. The baseline weights of all query dimensions are normalized to obtain the identification weight value of each query dimension. If the combination identification of the basic combination of query conditions is lower than 0.85, the combination identification is insufficient, which may lead to multiple orders matching the same hash value and make it impossible to accurately locate the target points change data. If it is higher than 0.85, the basic combination of query conditions can accurately cover the query combination, and the basic combination of query conditions can be used as the query condition combination.

[0034] Based on the user-input combined query conditions, the system matches the transaction hash value and the off-chain data index address from the query index table. The corresponding business evidence hash value is retrieved from the blockchain based on the transaction hash value, and then hash-verified with the data corresponding to the off-chain data index address to realize the query of order points information.

[0035] Reference Figure 5 As shown, further, combining the above-mentioned blockchain-based intelligent points management method, a blockchain-based intelligent points management system is proposed, including: The main control module is used to store key business datasets on the blockchain after consensus, obtain blockchain storage information, obtain data address binding information based on data storage differentiation based on multi-source original business datasets and key business datasets, and construct a query index table based on blockchain storage information and data address binding information. The information acquisition module is used to obtain order data synchronization authorization credentials based on the platform's open interface, establish a data transmission channel, collect all transaction order data on the platform in real time according to the data transmission channel, obtain user identity data, points account status data and historical transaction association data based on the user's unique identifier information, and obtain the combined query conditions input by the user. The topology analysis module is used to calculate the correlation coefficient between each feature variable and the integral change value based on the Pearson correlation coefficient formula, and to obtain the associated feature variables based on the correlation coefficient and correlation threshold corresponding to the feature variables. The display module interacts with the main control module and is used to output and display multi-source original business datasets, key business datasets, data address binding information, blockchain storage information, and query index tables.

[0036] The main control module specifically includes: The control unit is used to obtain data address binding information based on data storage differentiation according to the multi-source original business dataset and key business dataset, and to construct a query index table based on the blockchain storage information and data address binding information; An information receiving unit interacts with an information acquisition module and a topology analysis module to receive data and transmit it to a control unit. A blockchain storage unit is used to store key business datasets on the blockchain after consensus and to obtain blockchain storage information.

[0037] The information acquisition module specifically includes: The first acquisition unit is used to acquire order data synchronization authorization credentials based on the platform's open interface, establish a data transmission channel, and collect all transaction order data of the platform in real time according to the data transmission channel. The second acquisition unit is used to acquire user identity data, points account status data and historical transaction association data based on the user's unique identifier information, and to acquire the combined query conditions input by the user.

[0038] In summary, the advantages of this invention are as follows: Through Pearson correlation coefficient calculation and association topology graph construction, accurate mining of integral association features and efficient screening of key business data are achieved; through off-chain distributed storage and hash generation, low-cost storage of all business data and hash binding of core data are achieved, providing storage support for trusted blockchain evidence storage; through consortium blockchain consensus verification and on-chain writing of business blocks, tamper-proof evidence storage and full traceability of key integral business data are achieved, providing trusted assurance for compliance auditing of integral management; and through combined query condition filtering and conditional hash mapping construction, multi-dimensional and efficient indexing of integral order information is achieved, improving integral query efficiency.

[0039] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A points-based intelligent management method based on blockchain feedback, characterized in that, include: Acquire full data of platform transaction orders, points distribution data, service fee collection data, and user identification data to generate multi-source raw business datasets; Based on the correlation analysis of multi-source original business datasets, key business datasets are generated. Based on the multi-source original business dataset and key business dataset, and distinguishing data storage, data address binding information is obtained. The data address binding information includes the off-chain data index address and the business evidence hash value. Based on the key business dataset, after reaching a consensus on the blockchain, the data is stored on the blockchain and the blockchain storage information is obtained. A query index table is constructed based on the blockchain storage information and data address binding information; Based on the user-input combined query conditions, the system matches the transaction hash value and the off-chain data index address from the query index table. The corresponding business evidence hash value is retrieved from the blockchain based on the transaction hash value, and then hash-verified with the data corresponding to the off-chain data index address to realize the query of order points information.

2. The intelligent points management method based on blockchain feedback according to claim 1, characterized in that, The process of acquiring full data of platform transaction orders, points distribution data, service fee collection data, and user identification data to generate a multi-source raw business dataset specifically includes: Based on the platform's open interface, obtain the order data synchronization authorization certificate and establish a data transmission channel; According to the data transmission channel, the platform collects all transaction order data in real time. The platform transaction order data includes order number information, transaction amount data, product code information, transaction time information, user unique identifier information, and merchant identifier information. Based on the preset service fee calculation rules, service fee data is obtained using the full data of platform transaction orders. The service fee data includes platform service fee rate parameters and actual service fee received data. Based on the automatic points distribution rules, points distribution data is obtained using the full data of platform transaction orders. The points distribution data includes the points distribution base, distribution ratio, points change value, and points account identification data. Based on the user's unique identifier, obtain user identity data, points account status data, and historical transaction data; Using the order number as the unique primary key, the platform's full transaction order data, service fee data, and points distribution data are initially concatenated to generate concatenated business data; Based on the data collection timestamp, a unique time identifier is added to each piece of spliced ​​business data to distinguish multiple data synchronization requests for the same order; Based on the splicing of business data and time stamps, abnormal data such as duplicate collections and unconfirmed data that have exceeded the time limit are removed to generate cleaned business data; Based on the cleaned business data, it is grouped and aggregated by order number to ensure that the same order corresponds to a unique set of business data combinations. Based on the grouped and aggregated business data, a multi-source original business dataset is generated.

3. The intelligent points management method based on blockchain feedback according to claim 2, characterized in that, The process of performing correlation analysis based on multi-source original business datasets to generate key business datasets specifically includes: Based on the multi-source original business dataset, extract the data dimension set, including order number, transaction amount, transaction time, user identifier, service fee amount, and points change value; By using the numerical change of integrals as the target correlation variable and each data dimension as a feature variable, a variable correlation analysis matrix is ​​constructed. Based on the Pearson correlation coefficient formula, the correlation coefficient between each feature variable and the integral change value is calculated; Based on the need for data correlation analysis, a correlation threshold is set; Based on the correlation coefficients and correlation thresholds corresponding to the feature variables, obtain the associated feature variables; Based on the multi-source original business dataset, the corresponding related business data is extracted based on the related feature variables; Based on the related business data, the data is reordered by order number to generate a sequence of key business data. Based on key business data sequences, these are marked as core related data; Based on core related data, generate key business datasets.

4. The intelligent points management method based on blockchain feedback according to claim 3, characterized in that, The step of obtaining the associated feature variables based on the correlation coefficients and correlation thresholds corresponding to the feature variables specifically includes: Based on the set of data dimensions, each data dimension is used as a topology node to obtain the data topology node set; Based on the correlation coefficients corresponding to the feature variables, the maximum correlation coefficient is taken as the basic correlation coefficient, and the feature variable corresponding to the maximum correlation coefficient is taken as the initial feature variable; Based on the multi-source original business dataset, any two feature variables are used as a feature variable group; The multi-source original business dataset is traversed, and the feature correlation coefficients corresponding to the two feature variables in each feature variable group are obtained based on the Pearson correlation coefficient. Based on the correlation threshold, the feature variable group is screened, and the feature variable group whose correlation coefficient exceeds the correlation threshold is used as the benchmark feature variable group. Based on the baseline feature variable set, construct the association edge between two feature variables in the baseline feature variable set and obtain the association edge information; Construct an associated topology graph based on the data topology node set and associated edge information; The topology node corresponding to the initial feature variable is taken as the starting topology node, and the topology node corresponding to the target association variable is taken as the ending topology node. Based on the associated topology graph, a topology transmission chain is obtained using the starting topology node and the ending topology node as the basis. The topology transmission chain represents the topology node route from the starting topology node to the ending topology node through associated edges. The product of the feature correlation coefficients of adjacent topological nodes in each topological transit chain is used as the baseline correlation coefficient of that topological transit chain. The difference between the baseline correlation coefficient and the basic correlation coefficient is used as the correlation propagation bias coefficient; The topology transit chain corresponding to the minimum value of the correlation transit deviation coefficient is taken as the topology transit feature chain; The feature variables corresponding to the topology nodes in the topology transitive feature chain are used as associated feature variables.

5. The intelligent points management method based on blockchain feedback according to claim 4, characterized in that, The step of obtaining data address binding information based on multi-source original business datasets and key business datasets, distinguishing them by data storage, specifically includes: Based on the multi-source original business dataset, it is divided and encapsulated according to the order number to generate independent data blocks; Based on IPFS distributed storage, upload data blocks to off-chain storage nodes and obtain the node storage return value; Based on the node storage return value, a unique index address for off-chain data is generated, namely the IPFS hash address; Based on the key business dataset, the core related data is concatenated into a plaintext business string; Based on the SHA-256 hash algorithm, a business evidence hash value is generated by calculating the plaintext business string. Based on the business evidence hash value, verify the uniqueness of the hash value to ensure that the same order corresponds to a unique hash value and there is no hash collision; A two-way binding relationship is established based on the off-chain data index address and the business evidence hash value to generate data address binding information.

6. The intelligent points management method based on blockchain feedback according to claim 5, characterized in that, The process of uploading and storing key business datasets on the blockchain after consensus is achieved, and obtaining blockchain storage information, specifically includes: Based on key business datasets and business evidence hash values, assemble the blockchain on-chain data body; Based on the on-chain data body, calculate the overall data hash value and generate block header verification information; The on-chain data body and block header verification information are encapsulated into a business block awaiting consensus. Based on the blockchain consensus mechanism, obtain the information of the consortium chain nodes in the blockchain, and broadcast the business blocks to be reached for consensus to all consortium chain nodes. Based on the node verification rules corresponding to the consortium blockchain nodes, independently verify the integrity of the on-chain data body, the legality of the hash value, and the uniqueness of the order number; Based on the node verification results, the ratio of the number of nodes that pass verification to the total number of nodes in the consortium blockchain is used as the consensus pass rate. The consensus status of a block is determined based on the consensus pass rate. If the consensus of a block is determined to be passed, the data uploaded to the chain is confirmed to be legal and valid. Based on the business block passed by consensus, an on-chain write operation is performed to obtain blockchain storage information, which includes block height, transaction hash value, and on-chain evidence storage address.

7. The intelligent points management method based on blockchain feedback according to claim 6, characterized in that, The step of constructing a query index table based on blockchain storage information and data address binding information specifically includes: Based on the data dimension set, the query dimensions are determined according to the order number, user ID, transaction amount, transaction time, points change value, and service fee amount; Combine the order number with any one query dimension, and the user identifier with any two query dimensions to generate basic combination information for query conditions; Based on the query condition combination information and the topology transfer chain, the topology transfer chain containing the topology nodes corresponding to the query dimensions in each basic combination of query conditions is taken as the query topology transfer chain for that query condition combination. Retrieve the topology transit sub-chains between topology nodes corresponding to the query dimension in the query topology transit chain; The product of the feature correlation coefficients corresponding to adjacent topological nodes in the topological transitive subchain is taken as the combined correlation coefficient of the topological transitive subchain. The difference between the feature correlation coefficient corresponding to the query dimension in the basic combination of query conditions and the combination correlation coefficient is used as the combination deviation coefficient; Based on the combination deviation coefficient of the topology transit subchain corresponding to the basic combination of query conditions, the combination identification degree of the basic combination of query conditions is obtained. Based on the combination recognition, the basic combinations of query conditions are filtered to obtain a set of combined query conditions; Based on the set of combined query conditions, data address binding information, and transaction hash value, the combined query conditions are bound to the off-chain data index address, business notarization hash value, and transaction hash value to establish a condition hash mapping relationship; A query index table is generated based on the conditional hash mapping relationship.

8. A points-based intelligent management system based on blockchain feedback, used to implement the intelligent management method as described in any one of claims 1-7, characterized in that, include: The main control module is used to store key business datasets on the blockchain after consensus, obtain blockchain storage information, obtain data address binding information based on data storage differentiation based on multi-source original business datasets and key business datasets, and construct a query index table based on blockchain storage information and data address binding information. The information acquisition module is used to obtain order data synchronization authorization credentials based on the platform's open interface, establish a data transmission channel, collect all transaction order data on the platform in real time according to the data transmission channel, obtain user identity data, points account status data and historical transaction association data based on the user's unique identifier information, and obtain the combined query conditions input by the user. The topology analysis module is used to calculate the correlation coefficient between each feature variable and the integral change value based on the Pearson correlation coefficient formula, and to obtain the associated feature variables based on the correlation coefficient and correlation threshold corresponding to the feature variables. The display module interacts with the main control module and is used to output and display multi-source original business datasets, key business datasets, data address binding information, blockchain storage information, and query index tables.

9. A points-based intelligent management system based on blockchain feedback as described in claim 8, characterized in that, The main control module specifically includes: The control unit is used to obtain data address binding information based on data storage differentiation according to the multi-source original business dataset and key business dataset, and to construct a query index table based on the blockchain storage information and data address binding information; An information receiving unit interacts with an information acquisition module and a topology analysis module to receive data and transmit it to a control unit. A blockchain storage unit is used to store key business datasets on the blockchain after consensus and to obtain blockchain storage information.

10. A points-based intelligent management system based on blockchain feedback as described in claim 8, characterized in that, The information acquisition module specifically includes: The first acquisition unit is used to acquire order data synchronization authorization credentials based on the platform's open interface, establish a data transmission channel, and collect all transaction order data of the platform in real time according to the data transmission channel. The second acquisition unit is used to acquire user identity data, points account status data and historical transaction association data based on the user's unique identifier information, and to acquire the combined query conditions input by the user.

Images