Delivery service business rollback method and system

Abstract

The invention relates to the technical field of network transaction settlement, in particular to a delivery service business rollback method and system. Through the block chain technology and the encryption transmission protocol, the transparency and non-tampering performance of the fund flow and asset ownership verification process before, during and after delivery are ensured, and the security of transaction data is ensured. By monitoring fund flow and asset transfer of accounts of the buyer and the seller in real time, it is ensured that transactions conform to regulations, and risks caused by system lag or operation errors are reduced. Besides, in combination with a support vector machine technology, deep analysis is performed on long-term transaction behaviors, and potential violation operations, especially long-term accumulated abnormal behaviors, are found in time. Compared with a traditional mechanism, the method has the advantages that continuous violation behaviors can be recognized through historical data backtracking and auditing, and accurate and efficient anomaly detection and compliance review are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of network transaction settlement, and in particular to a method and system for rolling back a delivery service business. Background Art

[0002] The present application relates to the field of network transaction settlement technology, and in particular, to a delivery service business rollback method and system. In the prior art, the authorized patent CN118898482B previously applied by the applicant discloses a network transaction settlement optimization method based on blockchain and big data. The solution can effectively solve the problems of transaction data tampering and transaction information leakage, and has important application value in real-time settlement during the transaction process. However, in the process of applying the solution to the market, the applicant found that the solution technology cannot fully play its delivery business rollback function in specific application scenarios. The technical solution of patent CN118898482B can also be further extended to optimize its delivery business rollback and verification of the ownership of the delivery subject matter. Specifically, patent CN118898482B discloses the specific technical details and establishment process of the network transaction contract, but does not involve how to perform a transaction rollback operation based on the network transaction contract when it is identified that the buyer and seller have tampered with the transaction data before and after the settlement;

[0003] To this end, this application proposes a technical solution that is further explored on the solution of patent CN118898482B. This application aims to extend the technical solution based on patent CN118898482B, expand its functional implementation in the rollback of delivery business, combine blockchain technology, big data analysis and network transaction contracts, realize business rollback, fund freezing, asset transfer and anomaly detection in the delivery process, ensure the security and consistency of delivery business, and better meet the delivery business needs.

[0004] The technical solution of this application is not a modification of patent CN118898482B, but a further expansion and optimization based on it, so that it can realize full-process trusted backtracking in complex delivery scenarios, and work together with patent CN118898482B to enhance the overall technical effect and application value.

[0005] In view of the above problems, it is necessary to propose a delivery service business rollback method and system. Summary of the invention

[0006] The purpose of the present invention is to solve the problems existing in the background technology and to propose a delivery service business rollback method and system.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] In a first aspect, the present invention provides a method for rolling back a delivery service business, which specifically includes the following steps.

[0009] Step 1: Verify funding conditions before delivery;

[0010] After receiving the delivery request, access the blockchain of the buyer and seller involved in this delivery request, and obtain the initial data matrix of the buyer and seller through the encrypted transmission protocol, including the buyer's initial data matrix and the seller's starting data matrix ;

[0011] Among them, x11(t) is the order number; x12(t) is the order time; x13(t) is the product ID; x14(t) is the latest settlement time; x15(t) is the platform commission rate, and x16(t) is the actual settlement amount.

[0012] Among them, x21(t) is the buyer's ID at the current time t, x22(t) is the buyer's settlement account; x23(t) is the balance of the buyer's settlement account; x31(t) is the seller's ID at the current time t; x32(t) is the seller's settlement account; x33(t) is the balance of the seller's settlement account.

[0013] Based on the initial data matrix of the buyer and seller, the transaction information of this delivery request is monitored before delivery to confirm the availability of the buyer and seller's capital accounts.

[0014] By preset formula Calculate the buyer's balance detection value and seller balance detection value .

[0015] The buyer's balance detection value and the seller's balance detection value are recalculated at every preset detection interval. If the buyer's balance detection value is identified Less than 0 or seller balance detection value If it is greater than or equal to 0, it is determined that the fund account balances of the buyer and seller meet the transaction conditions, and a fund condition verification signal for the buyer and seller before delivery is output.

[0016] Step 2: Verify the flow of funds in the settlement;

[0017] During the delivery process, the blockchain accounts of the buyer and seller are accessed to obtain detailed information on each flow of funds and record real-time changes in the buyer's and seller's accounts.

[0018] For the blockchain accounts of the buyer and seller, the formula is: Calculate the buyer's liquidity test value and seller liquidity test value .

[0019] At preset time intervals, the blockchain accounts of the buyer and seller are accessed to obtain changes in their account balances. If a value of The funds are transferred to the account x32(t), and the seller's account is transferred with a value of , and the account from which the funds are transferred is x23(t), then it is determined that the flow of funds in the buyer's and seller's accounts meets the transaction conditions, and a verification signal of the flow of funds in the settlement between the buyer and seller is output.

[0020] Step 3: Verification of asset ownership after delivery;

[0021] After the delivery is completed, verify the ownership of the delivery subject matter of the buyer and seller. Obtain the status information of the delivery subject matter, including the product IDx13(t) and the corresponding ownership change record, and obtain the asset transfer time, buyer account ID and seller account ID involved in the asset transfer in the ownership change record.

[0022] If the following conditions are met at the same time:

[0023] The buyer's account ID involved in the asset transfer is consistent with x21(t);

[0024] The seller account ID involved in the asset transfer is consistent with x31(t);

[0025] The asset transfer time of product IDx13(t) shall not exceed the latest settlement timex14(t).

[0026] Then output the asset ownership verification signal.

[0027] Step 4: Delivery solidification and rollback based on network transaction contracts;

[0028] If the capital condition verification signal, capital flow verification signal and asset ownership verification signal are identified at the same time in a delivery business, the delivery business is judged to be legal and compliant, and the delivery business is marked as legal and solidified.

[0029] As a preferred method of the present invention, an online transaction contract is obtained and a solidification operation is performed on the delivery business marked as legal. The specific process is to write the settled buyer's block and the settled seller's block of the buyer and seller of the delivery business marked as legal into the blockchain to ensure the immutability and transparency of the delivery process.

[0030] As a preferred embodiment of the present invention, each settled buyer block and settled seller block includes a block header and a block body;

[0031] The block header includes a timestamp, a random number Y, a hash value, etc.; the timestamp refers to the total number of seconds from January 1, 1970 to the current time, and the random number Y refers to a random number Y generated between 0 and 1024 at every preset time interval t; the hash value specifically refers to the result of hashing the last starting buyer data matrix or the starting seller data matrix in the contract;

[0032] The block body includes a settled buyer data matrix and a settled seller data matrix.

[0033] As a preferred method of the present invention, the settled buyer block and the settled seller block are connected in series to the buyer's blockchain and the seller's blockchain through hash transformation to ensure that the information of this delivery business is permanently stored on the blockchain and cannot be tampered with or revoked.

[0034] The rollback operation is performed according to the rollback conditions triggered by anomaly detection. If the capital condition verification signal, capital flow verification signal and asset ownership verification signal in the same delivery business cannot be identified at the same time, the transaction rollback operation of this delivery business is performed.

[0035] The specific operation of transaction rollback is to cancel all delivery-related operations, restore the account balances and asset ownership data of the buyer and seller to the state before delivery, and ensure the consistency of system data, including:

[0036] Cancel fund operation: freeze the account balances and ownership of the delivery assets of the buyer and seller users, unfreeze the frozen funds after a preset time interval, and restore the account balances of the buyer and seller to the account balance status before the delivery business.

[0037] Asset recovery: Revoke the transferred assets and transfer the ownership of the delivered assets back to the seller’s blockchain account to ensure that the asset ownership is restored to the seller.

[0038] Data consistency recovery: Restore all transaction data, including transaction amount, delivery quantity and transaction price, based on the status record before delivery.

[0039] Step 5: Logging and backtracking audit;

[0040] After the rollback is completed, the system records all operations and data changes during the rollback process and generates a detailed rollback report. The system notifies the user of the rollback results and provides relevant logs and reports for review.

[0041] Record every step of the rollback process, and record the operation status and data changes of each link in detail to ensure that all operations can be audited and traced.

[0042] The results of the rollback are notified to both the buyer and seller users and the administrator via email and push notifications. A rollback report of the rollback operation is sent to the administrator.

[0043] As a preferred method of the present invention, machine learning and artificial intelligence technologies are used to conduct a retrospective audit of the transaction data in the consolidated delivery business, review the flow of funds and asset transfers, and identify transaction anomalies in historical data. The specific process is: use SVM support vector machine to review the data of the consolidated delivery business, and automatically determine whether the historical transaction records of each user actually violate compliance requirements.

[0044] A unique corresponding SVM support vector machine model is generated for all users, and the input of the SVM support vector machine model is the settled buyer data matrix or the settled seller data matrix generated in each consolidated delivery business.

[0045] The optimization objective function of the support vector machine is: ; where w is the normal vector of the decision plane, representing the hyperplane in the feature space; where b is the bias term, which determines the uniqueness of the decision hyperplane, and C is the regularization parameter, which is used to control the complexity of the support vector machine model and the tolerance of classification errors. is a slack variable used to measure the error of training sample i, where i is the number of the training sample, specifically the sequential number of the settled buyer data matrix and the settled seller data matrix, and N is the total number of samples.

[0046] Let the output function of the support vector machine be: , where x is the input item, specifically the settled buyer data matrix or settled seller data matrix generated in each consolidated delivery business. Where f(x) is the output decision value, f(x) greater than 0 means that the user's long-term trading behavior is compliant; f(x) less than or equal to 0 means that the user's long-term trading behavior is not compliant;

[0047] The training samples of the support vector machine are manually marked. If the settled buyer data matrix generated in a certain consolidated delivery business is judged to be compliant, then the output decision value f(x) is marked as 1; if the settled buyer data matrix generated in a certain consolidated delivery business is judged to be non-compliant, then the output decision value f(x) is marked as -1;

[0048] The manually labeled training samples are input into the optimization objective function of the support vector machine, and the decision plane normal vector w and the bias term b are obtained through training.

[0049] The decision plane normal vector w and bias term b obtained through training are substituted into the output function of the support vector machine as a machine learning model that automatically distinguishes between compliant and non-compliant historical transaction data, and conducts long-term data monitoring of all users.

[0050] If the output decision value f(x) of the output function of the support vector machine of a user is less than 0 in M ​​consecutive transactions, it is determined that the user has long-term illegal trading behavior, and an alarm message is sent to the user and the administrator.

[0051] In a second aspect, the present invention provides a delivery service business rollback system, including a capital condition verification module, a capital flow verification module, an asset ownership verification module, a delivery solidification and rollback module, and an anomaly detection and compliance review module.

[0052] After receiving the delivery request, the fund condition verification module verifies the accounts of the buyer and seller in real time based on blockchain technology to ensure that the funds in the buyer and seller accounts are sufficient and meet the transaction requirements. This module uses an encrypted transmission protocol to obtain the starting data matrix of the buyer and seller from the blockchain. The data includes the settlement accounts and balances of the buyer and seller. If the buyer's balance detection value is less than 0 or the seller's balance detection value is greater than or equal to 0, the transaction conditions are met and a fund verification signal is output.

[0053] The funds flow verification module monitors the flow of funds between the buyer and seller accounts in real time during the delivery process to ensure that the flow of funds is consistent with the expected transaction. By accessing the blockchain data, detailed information on each fund flow is obtained at set time intervals to check whether the funds flow as agreed.

[0054] The asset ownership verification module verifies whether the ownership of the asset has been correctly transferred after the delivery is completed. This module ensures that the asset ownership changes of the buyer and seller accounts comply with the delivery rules by comparing the asset transfer records. Read and compare the status information and ownership change records of the transaction subject matter, including the product ID and the account IDs of the buyer and seller.

[0055] The delivery solidification and rollback module is responsible for solidifying the legal delivery business into the blockchain to ensure the immutability and transparency of the transaction process. The settled buyer's block and seller's block are written into the blockchain to ensure that the transaction data cannot be tampered with. The security and integrity of the transaction data are ensured through the hash value.

[0056] If an abnormal transaction is found, the system will automatically trigger the rollback mechanism to restore the account status and asset ownership of both the buyer and seller. By locating the change vector of the illegal transaction and the rollback target transaction, the state change after the transaction is reversely replayed, thereby restoring the state before the transaction.

[0057] The anomaly detection and compliance review module conducts real-time audits and inspections of transaction data throughout the delivery process through machine learning and artificial intelligence technologies to identify potential transaction anomalies in historical transaction data. This module combines anomaly detection and compliance review models to review fund flow and asset transfer information to ensure that the historical transaction data of all users complies with the prescribed legal and compliance requirements.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. The present invention ensures the transparency and immutability of the capital flow and asset ownership verification process before, during and after delivery based on blockchain technology and encrypted transmission protocols. All transaction data is solidified on the blockchain to ensure that each transaction information is securely recorded, avoiding the risk of human tampering;

[0060] 2. The present invention ensures that the flow of funds and assets meets the expected transaction requirements by real-time monitoring of the flow of funds and changes in asset ownership between the buyer and seller accounts. Each flow of funds and asset transfer is recorded in detail, and continuous verification is implemented during the delivery process to ensure that every link of the transaction complies with regulations. This real-time monitoring mechanism greatly reduces the risks caused by system lags or operational errors;

[0061] 3. The present invention combines machine learning technology, especially support vector machines, to conduct in-depth analysis and monitoring of long-term transaction behaviors in delivery services. By analyzing historical transaction data, the SVM model can promptly detect small and potential violations, especially those that manifest as long-term accumulated abnormal behaviors. Compared with the traditional timestamp-based version chain mechanism, the system of the present invention can not only detect anomalies in real time, but also identify persistent illegal user behaviors through retrospective auditing of historical data, thereby achieving more accurate and efficient anomaly detection and compliance review. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings:

[0063] Figure 1 is a flow chart of the method of the present invention;

[0064] Figure 2 A schematic diagram of a network transaction contract proposed in an embodiment of the present invention;

[0065] Figure 3 A schematic diagram of a settled buyer block and a settled seller block proposed in an embodiment of the present invention;

[0066] Figure 4 It is a system block diagram of the present invention. DETAILED DESCRIPTION

[0067] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0068] See also Figure 1 As shown, a delivery service business rollback method includes the following steps:

[0069] Step 1: Verify funding conditions before delivery;

[0070] After receiving the delivery request, access the blockchain of the buyer and seller involved in this delivery request, and obtain the initial data matrix of the buyer and seller through the encrypted transmission protocol, including the buyer's initial data matrix and the seller's starting data matrix ;

[0071] Among them, x11(t) is the order number; x12(t) is the order time; x13(t) is the product ID; x14(t) is the latest settlement time; x15(t) is the platform commission rate, and x16(t) is the actual settlement amount.

[0072] Among them, x21(t) is the buyer's ID at the current time t, x22(t) is the buyer's settlement account; x23(t) is the balance of the buyer's settlement account; x31(t) is the seller's ID at the current time t; x32(t) is the seller's settlement account; x33(t) is the balance of the seller's settlement account.

[0073] Based on the initial data matrix of the buyer and seller, the transaction information of this delivery request is monitored before delivery to confirm the availability of the buyer and seller's capital accounts.

[0074] By preset formula Calculate the buyer's balance detection value and seller balance detection value .

[0075] The buyer's balance detection value and the seller's balance detection value are recalculated at every preset detection interval. If the buyer's balance detection value is identified Less than 0 or seller balance detection value If it is greater than or equal to 0, it is determined that the fund account balances of the buyer and seller meet the transaction conditions, and a fund condition verification signal for the buyer and seller before delivery is output.

[0076] Step 2: Verify the flow of funds in the settlement;

[0077] During the delivery process, the blockchain accounts of the buyer and seller are accessed to obtain detailed information on each flow of funds and record real-time changes in the buyer's and seller's accounts.

[0078] For the blockchain accounts of the buyer and seller, the formula is: Calculate the buyer's liquidity test value and seller liquidity test value .

[0079] At preset time intervals, the blockchain accounts of the buyer and seller are accessed to obtain changes in their account balances. If a value of The funds are transferred to the account x32(t), and the seller's account is transferred with a value of , and the account from which the funds are transferred is x23(t), then it is determined that the flow of funds in the buyer's and seller's accounts meets the transaction conditions, and a verification signal of the flow of funds in the settlement between the buyer and seller is output.

[0080] Step 3: Verification of asset ownership after delivery;

[0081] After the delivery is completed, verify the ownership of the delivery subject matter of the buyer and seller. Obtain the status information of the delivery subject matter, including the product IDx13(t) and the corresponding ownership change record, and obtain the asset transfer time, buyer account ID and seller account ID involved in the asset transfer in the ownership change record.

[0082] If the following conditions are met at the same time:

[0083] The buyer's account ID involved in the asset transfer is consistent with x21(t);

[0084] The seller account ID involved in the asset transfer is consistent with x31(t);

[0085] The asset transfer time of product IDx13(t) shall not exceed the latest settlement timex14(t).

[0086] Then output the asset ownership verification signal.

[0087] Step 4: Delivery solidification and rollback based on network transaction contracts;

[0088] If the capital condition verification signal, capital flow verification signal and asset ownership verification signal are identified at the same time in a delivery business, the delivery business is judged to be legal and compliant, and the delivery business is marked as legal and solidified.

[0089] See also Figure 2As shown, the network transaction contract is obtained, and the delivery business marked as legal is solidified. The specific process is to write the settled buyer blocks and settled seller blocks of the buyer and seller of the delivery business marked as legal into the blockchain to ensure the immutability and transparency of the delivery process.

[0090] See also Figure 3 As shown, each settled buyer block and settled seller block includes a block header and a block body;

[0091] The block header includes a timestamp, a random number Y, a hash value, etc.; the timestamp refers to the total number of seconds from January 1, 1970 to the current time, and the random number Y refers to a random number Y generated between 0 and 1024 at every preset time interval t; the hash value specifically refers to the result of hashing the last starting buyer data matrix or the starting seller data matrix in the contract;

[0092] The block body includes a settled buyer data matrix and a settled seller data matrix.

[0093] Furthermore, the settled buyer block and the settled seller block are connected in series to the buyer's blockchain and the seller's blockchain through hash transformation to ensure that the information of this delivery business is permanently stored on the blockchain and cannot be tampered with or revoked.

[0094] The rollback operation is performed according to the rollback conditions triggered by anomaly detection. If the capital condition verification signal, capital flow verification signal and asset ownership verification signal in the same delivery business cannot be identified at the same time, the transaction rollback operation of this delivery business is performed.

[0095] The specific operation of transaction rollback is to cancel all delivery-related operations, restore the account balances and asset ownership data of the buyer and seller to the state before the delivery, and ensure the consistency of system data, including:

[0096] Cancel fund operation: freeze the account balances and ownership of the delivery assets of the buyer and seller users, unfreeze the frozen funds after a preset time interval, and restore the account balances of the buyer and seller to the account balance status before the delivery business.

[0097] Asset recovery: Revoke the transferred assets and transfer the ownership of the delivered assets back to the seller’s blockchain account to ensure that the asset ownership is restored to the seller.

[0098] Data consistency recovery: Restore all transaction data, including transaction amount, delivery quantity and transaction price, based on the status record before delivery.

[0099] It should be noted that traditional transaction rollback systems usually use a timestamp-based version chain mechanism, and its core logic is to achieve rollback by recording the transaction data of all accounts under each time slice. When a transaction anomaly is detected, the system needs to traverse the state version in reverse along the time axis and cancel transactions one by one until the most recent consistent state before the error occurs.

[0100] It should be further explained that the transaction rollback method proposed in the present application directly locates illegal transactions and can complete the rollback by only replaying the state change vector after the transaction in reverse, thereby reducing the complexity of the rollback operation.

[0101] Step 5: Logging and backtracking audit;

[0102] After the rollback is completed, the system records all operations and data changes during the rollback process and generates a detailed rollback report. The system notifies the user of the rollback results and provides relevant logs and reports for review.

[0103] Record every step of the rollback process, and record the operation status and data changes of each link in detail to ensure that all operations can be audited and traced.

[0104] The results of the rollback are notified to both the buyer and seller users and the administrator via email and push notifications. A rollback report of the rollback operation is sent to the administrator.

[0105] Furthermore, machine learning and artificial intelligence technologies are used to conduct retrospective audits on transaction data in consolidated delivery businesses, review fund flows and asset transfers, and identify transaction anomalies in historical data. The specific process is: SVM support vector machine is used to review data on consolidated delivery businesses, and automatically determine whether each user's historical transaction records actually violate compliance requirements.

[0106] A unique corresponding SVM support vector machine model is generated for all users, and the input of the SVM support vector machine model is the settled buyer data matrix or the settled seller data matrix generated in each consolidated delivery business.

[0107] The optimization objective function of the support vector machine is: ; where w is the normal vector of the decision plane, representing the hyperplane in the feature space; where b is the bias term, which determines the uniqueness of the decision hyperplane, and C is the regularization parameter, which is used to control the complexity of the support vector machine model and the tolerance of classification errors. is a slack variable used to measure the error of training sample i, where i is the number of the training sample, specifically the sequential number of the settled buyer data matrix and the settled seller data matrix, and N is the total number of samples.

[0108] Let the output function of the support vector machine be: , where x is the input item, specifically the settled buyer data matrix or settled seller data matrix generated in each consolidated delivery business. Where f(x) is the output decision value, f(x) greater than 0 means that the user's long-term trading behavior is compliant; f(x) less than or equal to 0 means that the user's long-term trading behavior is not compliant;

[0109] The training samples of the support vector machine are manually marked. If the settled buyer data matrix generated in a certain consolidated delivery business is judged to be compliant, then the output decision value f(x) is marked as 1; if the settled buyer data matrix generated in a certain consolidated delivery business is judged to be non-compliant, then the output decision value f(x) is marked as -1;

[0110] The manually labeled training samples are input into the optimization objective function of the support vector machine, and the decision plane normal vector w and the bias term b are obtained through training.

[0111] The decision plane normal vector w and bias term b obtained through training are substituted into the output function of the support vector machine as a machine learning model that automatically distinguishes between compliant and non-compliant historical transaction data, and conducts long-term data monitoring of all users.

[0112] If the output decision value f(x) of the output function of the support vector machine of a user is less than 0 in M ​​consecutive transactions, it is determined that the user has long-term illegal trading behavior, and an alarm message is sent to the user and the administrator.

[0113] It should be noted that in the delivery service business, some illegal delivery businesses, especially those with potential violations, often manifest as long-term accumulated anomalies, and their violations are not detected at the beginning of the transaction, but gradually emerge in the process of continuous capital flow and asset changes. These illegal transactions are lagging and long-term, and simply relying on monitoring before and after the transaction is not enough to detect them in time. The introduction of support vector machines can combine machine learning algorithms for long-term monitoring and historical transaction data analysis to capture subtle but long-standing violations. For example, some illegal users may gradually manipulate capital accounts to make small fund transfers and irregular transactions. These transactions do not immediately violate the capital account balance requirements or asset transfer rules after each delivery, and therefore cannot be discovered through conventional real-time transaction monitoring.

[0114] See also Figure 4 As shown, a delivery service business rollback system includes a capital condition verification module, a capital flow verification module, an asset ownership verification module, a delivery solidification and rollback module, and an anomaly detection and compliance review module.

[0115] After receiving the delivery request, the fund condition verification module verifies the accounts of the buyer and seller in real time based on blockchain technology to ensure that the funds in the buyer and seller accounts are sufficient and meet the transaction requirements. This module uses an encrypted transmission protocol to obtain the starting data matrix of the buyer and seller from the blockchain. The data includes the settlement accounts and balances of the buyer and seller. If the buyer's balance detection value is less than 0 or the seller's balance detection value is greater than or equal to 0, the transaction conditions are met, and the fund verification signal is output according to the verification result.

[0116] The funds flow verification module monitors the flow of funds between the buyer and seller accounts in real time during the delivery process to ensure that the flow of funds is consistent with the expected transaction. By accessing the blockchain data, it obtains detailed information on each fund flow at set time intervals, checks whether the funds flow as agreed, and outputs a funds flow verification signal based on the verification results.

[0117] The asset ownership verification module verifies whether the ownership of the asset has been correctly transferred after the delivery is completed. This module ensures that the asset ownership changes of the buyer and seller accounts comply with the delivery rules by comparing the asset transfer records. It reads and compares the status information and ownership change records of the transaction subject matter, including the product ID and the account IDs of the buyer and seller, and outputs the asset ownership verification signal based on the verification results.

[0118] The delivery solidification and rollback module is responsible for solidifying the legal delivery business into the blockchain to ensure the immutability and transparency of the transaction process. The settled buyer's block and seller's block are written into the blockchain to ensure that the transaction data cannot be tampered with. The security and integrity of the transaction data are ensured through the hash value.

[0119] If an abnormal transaction is found, the system will automatically trigger the rollback mechanism to restore the account status and asset ownership of both the buyer and seller. By locating the change vector of the illegal transaction and the rollback target transaction, the state change after the transaction is reversely replayed, thereby restoring the state before the transaction.

[0120] The anomaly detection and compliance review module conducts real-time audits and inspections of transaction data throughout the delivery process through machine learning and artificial intelligence technologies to identify potential transaction anomalies in historical transaction data. This module combines anomaly detection and compliance review models to review fund flow and asset transfer information to ensure that the historical transaction data of all users complies with the prescribed legal and compliance requirements.

[0121] It should be understood that the terms “include” and “comprising” used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0122] It should also be understood that the terms used in this disclosure are only for the purpose of describing specific embodiments and are not intended to limit the disclosure. As used in this disclosure and the claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and the claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes these combinations;

[0123] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A delivery service business rollback method, characterized in that: The following steps are involved: Step 1: Verification of funding conditions before delivery; After receiving the delivery request, the blockchains of the buyer and seller are accessed through the encrypted transmission protocol to obtain the initial data matrices of the buyer and seller; based on the initial data matrices of the buyer and seller, the availability of the fund accounts of the buyer and seller is verified to confirm whether the fund balances of the buyer and seller meet the transaction conditions; Step 2: Verify the flow of funds in the delivery; During the delivery process, the blockchain accounts of the buyer and seller are accessed to monitor and record the flow of funds in real time; the fund detection value is calculated based on the flow of funds to verify whether the funds flow in the expected direction and confirm whether the fund flow of the buyer and seller meets the conditions; Step 3: Verification of asset ownership after delivery; After the delivery is completed, obtain the status information and ownership change records of the delivery object to confirm whether the asset ownership is correctly transferred. Through asset transfer analysis, ensure that the asset transfer between the buyer and the seller complies with the transaction agreement; Step 4: Settlement and rollback based on online trading contracts; If the verification results of capital conditions, capital flow and asset ownership are all passed, the delivery transaction will be marked as legal, and the delivery business will be solidified to ensure that the transaction data cannot be tampered with; otherwise, the delivery business will be rolled back to restore the transaction data, funds and delivery assets to the state before the transaction; Step 5: Logging and backtracking audit; Record all operations and data changes during the rollback process, generate a detailed rollback report, and provide rollback information notifications; conduct in-depth analysis and monitoring of long-term transaction behaviors in delivery services through support vector machine models.

2. A method for rolling back a delivery service according to claim 1, characterized in that: The specific process of verifying the availability of the buyer and seller's funding accounts based on the buyer and seller's starting data matrix is ​​as follows: Get the starting data matrix of the buyer and seller, including the buyer's starting data matrix and the seller's starting data matrix ; Among them, x11(t) is the order number; x12(t) is the order time; x13(t) is the product ID; x14(t) is the latest settlement time; x15(t) is the platform commission rate, and x16(t) is the actual settlement amount; Among them, x21(t) is the buyer's ID at the current time t, x22(t) is the buyer's settlement account; x23(t) is the balance of the buyer's settlement account; x31(t) is the seller's ID at the current time t; x32(t) is the seller's settlement account; x33(t) is the balance of the seller's settlement account; Based on the initial data matrix of the buyer and seller, the transaction information of this delivery request is monitored before delivery to confirm the availability of the buyer and seller's capital accounts; By preset formula Calculate the buyer's balance detection value and seller balance detection value ; The buyer's balance detection value and the seller's balance detection value are recalculated at every preset detection interval. If the buyer's balance detection value is identified Less than 0 or seller balance detection value If it is greater than or equal to 0, it is determined that the fund account balances of the buyer and seller meet the transaction conditions, and a fund condition verification signal for the buyer and seller before delivery is output.

3. A method for rolling back a delivery service according to claim 1, characterized in that: The specific process of real-time monitoring and recording of fund flows is as follows: During the delivery process, the blockchain accounts of the buyer and seller are accessed to obtain detailed information on each fund flow and record the real-time changes in the buyer and seller accounts; For the blockchain accounts of the buyer and seller, the formula is: Calculate the buyer's liquidity test value and seller liquidity test value ; At preset time intervals, the blockchain accounts of the buyer and seller are accessed to obtain changes in their account balances. If a value of The funds are transferred to the account x32(t), and the seller's account is transferred with a value of , and the account from which the funds are transferred is x23(t), then it is determined that the flow of funds in the buyer's and seller's accounts meets the transaction conditions, and a verification signal of the flow of funds in the settlement between the buyer and seller is output.

4. A method for rolling back a delivery service according to claim 1, characterized in that: The specific process of ensuring that the asset transfer between the buyer and the seller complies with the transaction agreement through asset transfer analysis is as follows: After the delivery is completed, verify the ownership of the delivery subject matter of the buyer and seller; obtain the status information of the delivery subject matter, including the commodity IDx13(t) and the corresponding ownership change record, and obtain the asset transfer time, the buyer's account ID and the seller's account ID involved in the asset transfer in the ownership change record; If the following conditions are met at the same time: The buyer's account ID involved in the asset transfer is consistent with x21(t); The seller account ID involved in the asset transfer is consistent with x31(t); The asset transfer time of commodity IDx13(t) shall not exceed the latest settlement timex14(t); Then output the asset ownership verification signal.

5. A method for rolling back a delivery service according to claim 1, characterized in that: The specific process of solidifying the delivery business is as follows: If the capital condition verification signal, capital flow verification signal and asset ownership verification signal in a delivery business are identified at the same time, the delivery business is judged to be legal and compliant, and the delivery business is marked as legal, and the delivery business is solidified; Obtain the online transaction contract and solidify the delivery business marked as legal. The specific process is to write the settled buyer's block and the settled seller's block of the buyer and seller of the delivery business marked as legal into the blockchain to ensure the immutability and transparency of the delivery process; Among them, each settled buyer block and settled seller block contains a block header and a block body; The block header includes a timestamp, a random number Y, a hash value, etc.; the timestamp refers to the total number of seconds from January 1, 1970 to the current time, and the random number Y refers to a random number Y generated between 0 and 1024 at every preset time interval t; the hash value specifically refers to the result of hashing the last starting buyer data matrix or the starting seller data matrix in the contract; The block body includes a settled buyer data matrix and a settled seller data matrix; The settled buyer block and the settled seller block are connected in series to the buyer's blockchain and the seller's blockchain through hash transformation to ensure that the information of this delivery business is permanently stored on the blockchain and cannot be tampered with or revoked.

6. A method for rolling back a settlement service according to claim 1, characterized in that: The specific process of rolling back the delivery business is as follows: If the capital condition verification signal, capital flow verification signal and asset ownership verification signal in the same delivery business cannot be identified at the same time, the transaction rollback operation of this delivery business will be executed; Transaction rollback operations include: Cancel fund operation: freeze the account balances and the ownership of the delivery assets of the buyer and seller users, unfreeze the frozen funds after a preset time interval, and restore the account balances of the buyer and seller to the account balance status before the delivery business; Asset recovery: Revoke the transferred assets and transfer the ownership of the delivered assets back to the seller’s blockchain account to ensure that the ownership of the assets is restored to the seller; Data consistency recovery: Restore all transaction data, including transaction amount, delivery quantity and transaction price, based on the status record before delivery.

7. A method for rolling back a settlement service according to claim 1, characterized in that: The specific process of generating a detailed rollback report and notifying rollback information is as follows: Record every step of the rollback process, and record the operation status and data changes of each link in detail to ensure that all operations can be audited and traced; The results of the rollback are notified to both the buyer and seller users and the administrator via email and push notifications; a rollback report of the rollback operation is sent to the administrator.

8. A method for rolling back a delivery service according to claim 1, characterized in that: The specific process of in-depth analysis and monitoring of long-term transaction behaviors in delivery services through the support vector machine model is as follows: Generate a unique corresponding SVM support vector machine model for all users, and let the input of the SVM support vector machine model be the settled buyer data matrix or the settled seller data matrix generated in each consolidated delivery business; The optimization objective function of the support vector machine is: ; where w is the normal vector of the decision plane, representing the hyperplane in the feature space; where b is the bias term, which determines the uniqueness of the decision hyperplane, and C is the regularization parameter, which is used to control the complexity of the support vector machine model and the tolerance of classification errors. is a slack variable used to measure the error of training sample i, where i is the number of the training sample, specifically the sequential number of the settled buyer data matrix and the settled seller data matrix, and N is the total number of samples; Let the output function of the support vector machine be: , where x is the input item, specifically the settled buyer data matrix or settled seller data matrix generated in each consolidated delivery business; where f(x) is the output decision value, f(x) greater than 0 means that the user's long-term trading behavior is compliant; f(x) less than or equal to 0 means that the user's long-term trading behavior is not compliant; The training samples of the support vector machine are manually marked. If the settled buyer data matrix generated in a certain consolidated delivery business is judged to be compliant, then the output decision value f(x) is marked as 1; if the settled buyer data matrix generated in a certain consolidated delivery business is judged to be non-compliant, then the output decision value f(x) is marked as -1; The manually labeled training samples are input into the optimization objective function of the support vector machine, and the decision plane normal vector w and the bias term b are obtained through training; Substitute the trained decision plane normal vector w and bias term b into the output function of the support vector machine as a machine learning model that automatically distinguishes between compliant and non-compliant historical transaction data, and conducts long-term data monitoring of all users; If the output decision value f(x) of the output function of the support vector machine of a user is less than 0 in M ​​consecutive transactions, it is determined that the user has long-term illegal trading behavior, and an alarm message is sent to the user and the administrator.

9. A delivery service business rollback system, including a capital condition verification module, a capital flow verification module, an asset ownership verification module, a delivery solidification and rollback module and an anomaly detection and compliance review module, characterized in that A method for rolling back a delivery service business used to implement any one of claims 1-8; After receiving the delivery request, the fund condition verification module verifies the accounts of the buyer and seller in real time based on blockchain technology to ensure that the funds in the buyer and seller accounts are sufficient and meet the transaction requirements; it uses an encrypted transmission protocol to obtain the initial data matrix of the buyer and seller from the blockchain, including the settlement accounts and balances of the buyer and seller, and performs calculation verification, and outputs a fund verification signal based on the verification results; The funds flow verification module monitors the flow of funds between the buyer and seller accounts in real time during the delivery process; by accessing blockchain data, it obtains detailed information on each fund flow at set time intervals, checks whether the funds flow as agreed, and performs calculation verification, and outputs a funds flow verification signal based on the verification results; The asset ownership verification module verifies whether the ownership of the asset has been correctly transferred after the delivery is completed; Ensure that the asset ownership changes of the buyer and seller accounts comply with the delivery rules by comparing the asset transfer records; read and compare the status information and ownership change records of the transaction subject and verify them, and output the asset ownership verification signal based on the verification results; The delivery solidification and rollback module is responsible for solidifying the legal delivery business into the blockchain; writing the settled buyer's block and seller's block into the blockchain; and ensuring the security and integrity of transaction data through hash values; If any transaction anomaly is found, the rollback mechanism will be automatically triggered to restore the account status and asset ownership of the buyer and seller; by locating the change vector of the illegal transaction and the rollback target transaction, the state changes after the transaction are reversely replayed to restore the state before the transaction; during the entire delivery process, the anomaly detection and compliance review module uses machine learning and artificial intelligence technology to conduct real-time audits and inspections on transaction data to identify potential transaction anomalies in historical transaction data.

Images