Methods, apparatus, electronic devices, and readable storage media for processing clearing messages.

Abstract

This application discloses a method, apparatus, electronic device, and readable storage medium for processing clearing messages, applicable to the field of financial technology. The method includes: receiving a clearing message, wherein the clearing message includes at least transaction information and remarks data corresponding to a transaction to be cleared; inputting the remarks data from the clearing message into a target clearing model for identification, obtaining an identification result for the remarks data; based on the identification information related to the transaction to be cleared included in the identification result, searching a target transaction record matching the identification result from a transaction number table; determining the matching result between the target transaction record and the clearing message based on the identification result and the transaction information in the clearing message; and updating the clearing status corresponding to the clearing message to a clearing completed status in response to the matching result indicating a successful match between the target transaction record and the clearing message. This application solves the technical problem of low processing efficiency for clearing messages.

Description

Technical Field

[0001] This application relates to the field of financial technology, and more specifically, to a method, apparatus, electronic device, and readable storage medium for processing clearing messages. Background Technology

[0002] Currently, in the financial market, transaction clearing and settlement involves the actual transfer and confirmation of funds between the transacting parties, making it a crucial step. With increasing transaction volume and complexity, traditional clearing and settlement methods rely heavily on precise matching of the date, amount, account number, and remarks in the reported data. This is particularly problematic when clearing reports involve summarizing payments from multiple transactions. Since the remarks depend on the clearing sender, inaccurate remarks pose a significant challenge to automated settlement. Therefore, there is a technical problem of low efficiency in clearing and settlement.

[0003] There is currently no effective solution to the technical problem of low efficiency in clearing and write-off processes in related technologies. Summary of the Invention

[0004] The main objective of this application is to provide a method, apparatus, electronic device, and readable storage medium for processing clearing messages, so as to solve the technical problem of low processing efficiency of clearing messages in related technologies.

[0005] To achieve the above objectives, according to one aspect of this application, a method for processing settlement messages is provided. The method includes: receiving a settlement message, wherein the settlement message includes at least transaction information and remark data corresponding to a transaction to be reconciled, the remark data being unstructured descriptive information of the transaction to be reconciled; inputting the remark data in the settlement message into a target reconciliation model for identification, obtaining an identification result of the remark data, wherein the target reconciliation model is obtained by pre-training an initial reconciliation model using remark data samples, and the identification result includes at least identification information related to the transaction to be reconciled; based on the identification information related to the transaction to be reconciled included in the identification result, searching a target transaction record matching the identification result from a transaction number table, wherein the transaction number table stores at least one transaction record; based on the identification result and the transaction information in the settlement message, determining a matching result between the target transaction record and the settlement message, wherein the matching result indicates the degree of matching between the target transaction record and the settlement message; and updating the reconciliation status corresponding to the settlement message to a reconciliation completed status in response to the matching result indicating a successful match between the target transaction record and the settlement message.

[0006] Optionally, the identification information related to the transaction to be written off includes at least the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off. Based on the identification information related to the transaction to be written off included in the identification result, the target transaction record matching the identification result is searched from the transaction number table. This includes: constructing query conditions based on the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off, wherein the query conditions include at least one of the following: the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off; and searching for the target transaction record matching the identification result in the transaction number table based on the query conditions.

[0007] Optionally, based on the identification results and the transaction information in the clearing message, the matching result between the target transaction record and the clearing message is determined, including: extracting the key elements corresponding to the transaction to be cancelled from the identification results and the transaction information in the clearing message, and extracting the key elements corresponding to the target transaction record from the target transaction record, wherein the key elements include at least the transaction currency, transaction date and transaction amount; matching the key elements corresponding to the transaction to be cancelled with the key elements corresponding to the target transaction record to obtain the matching result.

[0008] Optionally, the method for processing the clearing message further includes: comparing the degree of matching between the clearing message and the target transaction record with a matching threshold to obtain a comparison result; and determining that the target transaction record and the clearing message are successfully matched in response to the comparison result indicating that the degree of matching is greater than the matching threshold.

[0009] Optionally, the method for processing settlement messages further includes: obtaining postscript data samples from historical settlement messages; labeling the identification information related to the transaction samples to be reconciled in the postscript data samples to obtain label data; dividing the labeled postscript data samples into training set and test set; training the initial reconciliation model using the training set to obtain the trained initial reconciliation model; verifying the recognition accuracy of the trained initial reconciliation model using the test set to obtain the verification result; and determining the trained initial reconciliation model as the target reconciliation model in response to the verification result indicating that the recognition accuracy is greater than the accuracy threshold.

[0010] Optionally, the method for processing the settlement message further includes: adjusting the model parameters of the trained initial reconciliation model in response to the recognition accuracy not being greater than the accuracy threshold; and continuing to train the adjusted initial reconciliation model using the training set until the recognition accuracy of the trained initial reconciliation model is greater than the accuracy threshold.

[0011] To achieve the above objectives, according to another aspect of this application, a processing apparatus for clearing messages is provided. The apparatus includes: a receiving unit for receiving a clearing message, wherein the clearing message includes at least transaction information and remark data corresponding to the transaction to be cleared, and the remark data is unstructured descriptive information of the transaction to be cleared; an identification unit for inputting the remark data in the clearing message into a target clearing model for identification, and obtaining the identification result of the remark data, wherein the target clearing model is obtained by pre-training an initial clearing model using remark data samples, and the identification result includes at least identification information related to the transaction to be cleared; a searching unit for searching a target transaction record that matches the identification result from a transaction number table based on the identification information related to the transaction to be cleared included in the identification result, wherein the transaction number table stores at least one transaction record; a determining unit for determining the matching result between the target transaction record and the clearing message based on the identification result and the transaction information in the clearing message, wherein the matching result is used to indicate the degree of matching between the target transaction record and the clearing message; and an updating unit for updating the clearing status corresponding to the clearing message to the clearing completion status in response to the matching result indicating that the target transaction record and the clearing message are successfully matched.

[0012] Optionally, the search unit is also used to: construct query conditions based on the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off, wherein the query conditions include at least one of the following information: the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off; and query the target transaction record that matches the identification result in the transaction number table based on the query conditions.

[0013] Optionally, the determining unit is further configured to: extract key elements corresponding to the transaction to be cancelled from the identification results and transaction information in the clearing message, and extract key elements corresponding to the target transaction record from the target transaction record, wherein the key elements include at least the transaction currency, transaction date and transaction amount; and match the key elements corresponding to the transaction to be cancelled with the key elements corresponding to the target transaction record to obtain a matching result.

[0014] Optionally, the clearing message processing device is further configured to: compare the degree of matching between the clearing message and the target transaction record with a matching threshold to obtain a comparison result; and, in response to the comparison result indicating that the degree of matching is greater than the matching threshold, determine that the target transaction record and the clearing message are successfully matched.

[0015] Optionally, the clearing message processing device is further configured to: obtain postscript data samples from historical clearing messages; annotate the identification information related to the transaction samples to be reconciled in the postscript data samples to obtain label data; divide the annotated postscript data samples into a training set and a test set; train the initial reconciliation model using the training set to obtain the trained initial reconciliation model; verify the recognition accuracy of the trained initial reconciliation model using the test set; and, in response to the recognition accuracy being greater than the accuracy threshold, determine the trained initial reconciliation model as the target reconciliation model.

[0016] Optionally, the clearing message processing device is further configured to: adjust the model parameters of the trained initial reconciliation model in response to the recognition accuracy not being greater than the accuracy threshold; and continue to train the adjusted initial reconciliation model using the training set until the recognition accuracy of the trained initial reconciliation model is greater than the accuracy threshold.

[0017] To achieve the above objectives, according to another aspect of this application, an electronic device is provided. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0018] To achieve the above objectives, according to another aspect of this application, a computer-readable storage medium is provided, which stores a plurality of instructions adapted for loading by a processor and executing the steps in any of the above method embodiments.

[0019] To achieve the above objectives, according to another aspect of this application, a computer program product is provided, which includes a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0020] In this embodiment, a settlement message is received, wherein the settlement message includes at least transaction information and remarks data corresponding to the transaction to be reconciled, and the remarks data is unstructured descriptive information of the transaction to be reconciled; the remarks data in the settlement message is input into a target reconciliation model for identification, and the identification result of the remarks data is obtained, wherein the target reconciliation model is obtained by pre-training an initial reconciliation model using remarks data samples, and the identification result includes at least identification information related to the transaction to be reconciled; based on the identification information related to the transaction to be reconciled included in the identification result, a target transaction record matching the identification result is searched from a transaction number table, wherein the transaction number table stores at least one transaction record; based on the identification result and the transaction information in the settlement message, a matching result between the target transaction record and the settlement message is determined, wherein the matching result is used to indicate the degree of matching between the target transaction record and the settlement message; in response to the matching result indicating that the target transaction record and the settlement message are successfully matched, the reconciliation status corresponding to the settlement message is updated to the reconciliation completed status. In other words, in this embodiment of the application, by inputting the remarks data in the clearing message into the target reconciliation model for automatic identification, the identification information related to the transaction to be reconciled can be accurately extracted from the free-format remarks data. This improves the accuracy of matching with transaction records in the transaction number table, reduces errors and omissions that may be introduced by traditional manual verification, achieves efficient processing of unstructured descriptive information, reduces the need for manual parsing, greatly improves the automation level and reconciliation efficiency of financial market transaction clearing message processing, and thus solves the technical problem of low reconciliation efficiency of clearing messages. Attached Figure Description

[0021] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0022] Figure 1 This is a hardware structure block diagram of a computer terminal for implementing a method for processing clearing messages according to an embodiment of this application;

[0023] Figure 2 This is a flowchart of a method for processing a clearing message according to an embodiment of this application;

[0024] Figure 3 This is a flowchart of another method for processing clearing messages according to an embodiment of this application;

[0025] Figure 4 This is a schematic diagram of a clearing message processing apparatus according to an embodiment of this application;

[0026] Figure 5 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] It should be noted that the information collected in this application (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of this data all comply with relevant laws, regulations, and standards, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding access points are provided for users to choose to authorize or refuse. For example, interfaces are set up between this system and relevant users or organizations, providing users with corresponding access points to choose to agree to or refuse automated decision-making results; if the user chooses to refuse, the process proceeds to the expert decision-making stage.

[0030] Example 1

[0031] According to an embodiment of this application, a method for processing clearing messages is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0032] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1This is a hardware structure block diagram of a computer terminal for implementing a method for processing clearing messages, according to an embodiment of this application. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor (MCU) or a field-programmable gate array (FPGA), etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O), a Universal Serial Bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0033] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0034] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the clearing message processing method in this embodiment. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby implementing the above-mentioned clearing message processing method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0035] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0036] The display can be, for example, a touchscreen liquid crystal display (LCD), which allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0037] Under the aforementioned operating environment, this application provides the following: Figure 2 The method for processing clearing messages is shown. Figure 2 This is a flowchart of the clearing message processing method according to Embodiment 1 of this application.

[0038] Step S201: Create a database execution object.

[0039] In the technical solution provided in step S201 of this application, the clearing message includes at least the transaction information corresponding to the transaction to be cleared and the remarks data, which is unstructured descriptive information of the transaction to be cleared. The transaction information corresponding to the transaction to be cleared is structured information, containing key data such as the transaction date, time, transaction amount, currency, clearing counterparty information, and transaction account. This information is usually the basis for automatic reading and processing by the electronic clearing system. Remarks data differs from transaction information; it is unstructured descriptive information and may contain additional information about the transaction, such as the transaction background, nature of the transaction, specific transaction number, name of the clearing counterparty, and description of the transaction product. However, because remarks data is filled out by different institutions or individuals, its format and content style vary greatly, making it difficult for traditional automated systems to directly parse and utilize it. This is merely an example and does not limit the specific content included in the transaction information and remarks data.

[0040] In this embodiment, when one or more financial transactions meet the settlement conditions, the financial institution or clearing system receives a clearing message. These messages are electronic documents used by the parties to the transaction or their agents to execute transaction confirmation and fund transfer. They contain detailed transaction information, such as the transaction date, amount, currency, account information of both parties, and additional remarks.

[0041] Step S202: Input the remarks data in the settlement message into the target write-off model for identification, and obtain the identification result of the remarks data.

[0042] In the technical solution provided in step S202 of this application, the target reconciliation model is obtained by training the initial reconciliation model in advance using postscript data samples. That is, the target reconciliation model is obtained by training and optimizing the initial reconciliation model based on a large number of labeled postscript data samples. In other words, the target reconciliation model has learned how to identify key transaction information from the postscript data.

[0043] In this embodiment, after receiving the settlement message, the postscript data in the settlement message can be input into the target write-off model for identification to obtain the identification result of the postscript data, thereby improving the identification accuracy of relevant information in the postscript data.

[0044] Optionally, during the training phase of the target reconciliation model, the initial reconciliation model learns the correlation between the appendix data samples and the labeled transaction identification information (such as transaction number, clearing counterparty, product code, etc.), gradually optimizes the parameters, improves the accuracy of identification, and then, after the identification accuracy of the initial reconciliation model meets the conditions, the initial reconciliation model is determined as the target reconciliation model for subsequent transaction matching and reconciliation confirmation.

[0045] Step S203: Based on the identification information related to the transaction to be reconciled included in the identification result, search for the target transaction record that matches the identification result from the transaction number table.

[0046] In the technical solution provided by step S203 of this application, the transaction number table stores at least one transaction record.

[0047] In this embodiment, after obtaining the identification result of the remarks data in the clearing message, the target transaction record that matches the identification result can be found from the transaction number table based on the identification information related to the transaction to be cleared included in the identification result.

[0048] For example, the identification results may include a series of identifying information related to the transactions to be written off, such as the transaction number, the name of the clearing counterparty, the specific bond name or abbreviation, and the principal repayment amount. The transaction number table is a database storing all current or historical transaction records. Each transaction record includes a unique transaction number and all information associated with that transaction, such as the specific bond name or abbreviation, the principal repayment amount, the transaction date, the transaction amount, participant information, and the product type. Based on the identifying information related to the transactions to be written off extracted from the identification results, the transaction number table can be automatically queried to retrieve the transaction record matching the identified information.

[0049] In this step, since the transaction number table stores at least one transaction record, and each transaction record includes specific transaction information, the identification results of the remarks data included in the settlement message of the transaction to be rewritten can be used to quickly retrieve the transaction record that matches the transaction to be rewritten from the transaction number, thereby improving the rewriting efficiency of the transaction to be rewritten.

[0050] Step S204: Based on the identification results and the transaction information in the clearing message, determine the matching result between the target transaction record and the clearing message.

[0051] In the technical solution provided by step S204 of this application, the matching result is used to indicate the degree of matching between the target transaction record and the settlement message.

[0052] In this embodiment, after obtaining the target transaction record, the identification result and the transaction information in the settlement message can be matched with the corresponding information stored in the target transaction record to obtain a matching result. The degree of matching between the target transaction record and the settlement message is determined based on the matching result, and then it is determined whether the target transaction record corresponds to the transaction to be written off based on the degree of matching.

[0053] For example, matching involves comparing multiple dimensions, including but not limited to amount, currency, transaction date, information of both parties, and transaction number. If the degree of matching for each piece of information is greater than the matching threshold, it indicates that the settlement message and the target transaction record are successfully matched, and the target transaction record can be identified as the transaction record corresponding to the transaction to be written off. Conversely, if there is inconsistency, it indicates that the degree of matching between the settlement message and the target transaction record is low. In this case, further manual confirmation or re-invoking the large model for in-depth analysis is required.

[0054] In this step, by comparing the identification results of the transaction information and remarks data corresponding to the clearing message with the corresponding information in the target transaction record from multiple dimensions, it can be confirmed whether the target transaction record is the transaction record corresponding to the transaction to be cleared, thereby improving the accuracy of clearing and clearing.

[0055] Step S205: In response to the matching result indicating that the target transaction record and the settlement message are successfully matched, the write-off status corresponding to the settlement message is updated to the write-off completed status.

[0056] In the technical solution provided by step S205 of this application, after obtaining the matching result through step S204, if the matching result indicates that the target transaction record and the clearing message are successfully matched, the write-off status corresponding to the clearing message is updated from "pending write-off status" to "write-off completed status".

[0057] In this step, after the clearing message is successfully matched with the target transaction record, the write-off status of the clearing message is updated to the write-off completed status. This avoids repeated write-offs of the same clearing message. Moreover, when the write-off status of the clearing message is in the write-off completed status, it can ensure that the transaction is quickly confirmed and settled, reducing the risk of transaction delays.

[0058] In steps S201 to S205 above, by inputting the remarks data in the clearing message into the target reconciliation model for automatic identification, the identification information related to the transaction to be reconciled can be accurately extracted from the free-format remarks data. This improves the accuracy of matching with transaction records in the transaction number table, reduces errors and omissions that may be introduced by traditional manual verification, achieves efficient processing of unstructured descriptive information, reduces the need for manual parsing, greatly improves the automation level and reconciliation efficiency of financial market transaction clearing message processing, and thus solves the technical problem of low reconciliation efficiency of clearing messages.

[0059] The embodiments of the present invention will now be described in detail with reference to the steps described above.

[0060] As an optional implementation, the identification information related to the transaction to be written off includes at least the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off. Step S203 involves searching for a target transaction record that matches the identification result from the transaction number table based on the identification information related to the transaction to be written off included in the identification result. This includes: constructing query conditions based on the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off. The query conditions include at least one of the following: the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off. Based on the query conditions, the target transaction record that matches the identification result is searched in the transaction number table.

[0061] In this embodiment, after identifying the remarks data in the clearing message and obtaining the identification results, query conditions can be constructed based on the identification information of the transaction to be cleared, the identification information of the clearing object, and the identification information of the transaction product corresponding to the transaction to be cleared, which are included in the identification results. These query conditions are used to find transaction records that match these identification information in the transaction number table.

[0062] Optionally, the transaction number table is a database used by financial institutions to record each transaction. It stores all structured information related to the transaction, including the transaction number, clearing object information, and transaction product type, and is used to confirm the consistency between the clearing message and the back-end transaction records. Based on the constructed query conditions, the transaction number table can be used to search for transaction records that match the identification information of the clearing message, and then the found transaction records can be identified as the target transaction records.

[0063] In this step, unstructured information identified by the large model and structured transaction information in the clearing message are used to construct query conditions and locate the target transaction record. This achieves intelligent matching between the clearing message and the transaction record, thereby improving clearing efficiency, reducing operational risks, and laying the foundation for subsequent transaction confirmation and status updates.

[0064] As an optional implementation, step S204, based on the identification result and the transaction information in the clearing message, determines the matching result between the target transaction record and the clearing message, including: extracting the key elements corresponding to the transaction to be cancelled from the identification result and the transaction information in the clearing message, and extracting the key elements corresponding to the target transaction record from the target transaction record, wherein the key elements include at least the transaction currency, transaction date and transaction amount; matching the key elements corresponding to the transaction to be cancelled with the key elements corresponding to the target transaction record to obtain the matching result.

[0065] In this embodiment, after obtaining the identification result of the postscript data, key elements can be extracted from the identification result and the transaction information in the settlement message, such as the transaction currency, transaction date and transaction amount. This is only an example and does not limit the specific content of the key elements.

[0066] Optionally, while extracting key elements from the identification results and transaction information in the clearing message, corresponding key elements can also be extracted from the target transaction record. Then, the key elements of the transactions to be cleared extracted from the identification results and clearing message are compared one-to-one with the key elements in the target transaction record to determine whether the target transaction record corresponds to the clearing message. If all key elements match perfectly, the matching result is successful; if any element does not match, the matching fails, requiring additional manual review or further in-depth analysis using a large model.

[0067] In this step, by matching the identification results and key elements extracted from the clearing message with the key elements in the target transaction record, the verification and confirmation of the clearing message in the financial market can be completed automatically and accurately, which greatly improves the efficiency and accuracy of the clearing process.

[0068] As an optional implementation, the method for processing the clearing message further includes: comparing the degree of matching between the clearing message and the target transaction record with a matching threshold to obtain a comparison result; and determining that the target transaction record and the clearing message are successfully matched in response to the comparison result indicating that the degree of matching is greater than the matching threshold.

[0069] In this embodiment, after obtaining the matching result between the clearing message and the target transaction record, since the matching result indicates the degree of matching between the clearing message and the target transaction record, the degree of matching can be compared with a matching threshold to obtain a comparison result. Then, based on the comparison result, it is determined whether the target transaction record and the clearing message are successfully matched.

[0070] For example, if the comparison result indicates that the match between the clearing message and the target transaction record is greater than the matching threshold, then the target transaction record and the clearing message are successfully matched. If the comparison result indicates that the match between the clearing message and the target transaction record is not greater than the matching threshold, then the match between the target transaction record and the clearing message is determined to be unsuccessful.

[0071] In this step, using the matching threshold as a standard to measure the degree of matching between the clearing message and the target transaction record can ensure that the target transaction record is the transaction record corresponding to the clearing message.

[0072] The training process of the target reimbursement model will be introduced next.

[0073] As an optional implementation, the method for processing the settlement message further includes: obtaining postscript data samples from historical settlement messages; labeling the identification information related to the transaction samples to be reconciled in the postscript data samples to obtain label data; dividing the labeled postscript data samples into a training set and a test set; training the initial reconciliation model using the training set to obtain the trained initial reconciliation model; verifying the recognition accuracy of the trained initial reconciliation model using the test set to obtain a verification result; and determining the trained initial reconciliation model as the target reconciliation model in response to the verification result indicating that the recognition accuracy is greater than the accuracy threshold.

[0074] In this embodiment, samples can be extracted from historical clearing messages of financial institutions. These samples contain various remarks, reflecting the actual diversity and complexity of clearing messages. The selection of historical data should cover different transaction types, clearing counterparties, products, and various possible remarks formats to ensure the comprehensiveness and generalization ability of the model training. The extracted remarks data samples are manually annotated by business experts or system engineers, focusing on highlighting identifying information related to the transactions to be cleared. This identifying information may include transaction number, clearing counterparty name, product code, transaction date, transaction amount, etc. The annotation process forms a set of labeled data to guide the model's learning.

[0075] Optionally, after obtaining the postscript data samples and label data, the labeled postscript data samples can be divided into a training set and a test set. The training set is used to train the initial reconciliation model, enabling the initial reconciliation model to learn how to identify key identification information from the postscripts of clearing messages; the test set is used to verify the performance and accuracy of the initial reconciliation model after training is completed.

[0076] Optionally, test set data can be used to validate the trained initial reconciliation model. The recognition accuracy of the initial reconciliation model is evaluated by comparing its recognition results with manually labeled data. An accuracy threshold is set as the performance evaluation criterion for the initial reconciliation model. If the recognition accuracy verified using the test set is greater than the set threshold, it indicates that the performance of the initial reconciliation model meets the requirements. In this case, the trained initial reconciliation model can be designated as the target reconciliation model. This target reconciliation model can effectively identify and process the remark data in settlement messages, providing support for automatic reconciliation confirmation.

[0077] In this step, data collection, annotation, model training, and model validation ensure that the establishment of the target reconciliation model is based on sufficient historical data and rigorous performance verification. Once the target reconciliation model is determined, it can be applied to automatically process newly received settlement messages, improving settlement efficiency and reducing the risks of manual operation.

[0078] As an optional implementation, the method for processing the settlement message further includes: adjusting the model parameters of the trained initial reconciliation model in response to the recognition accuracy not being greater than the accuracy threshold; and continuing to train the adjusted initial reconciliation model using the training set until the recognition accuracy of the trained initial reconciliation model is greater than the accuracy threshold.

[0079] In this embodiment, if the initial reimbursement model's recognition accuracy on the test set is not greater than the accuracy threshold, it means that the model's current performance has not yet met the business requirements and further optimization is needed. In this case, the model parameters of the trained initial reimbursement model will be adjusted automatically or manually by the developers. Parameter adjustments may include, but are not limited to, modifying the learning rate, increasing the number of neural network layers, and optimizing feature selection, in order to improve the model's recognition ability and accuracy.

[0080] Optionally, the model can be trained again using the adjusted model parameters and training set data. This process may require multiple iterations, gradually improving the model's recognition accuracy through continuous training and parameter fine-tuning. After each model training, the recognition accuracy of the initial reconciliation model is re-verified using test set data to check the effectiveness of the adjustments and whether the initial reconciliation model is close to or has reached the preset accuracy threshold. When the recognition accuracy of the trained initial reconciliation model finally exceeds the accuracy threshold, the system confirms that the model training is complete, and the model is considered the target reconciliation model, ready for use in the actual clearing message processing flow.

[0081] In this step, a closed-loop mechanism for model training and validation ensures that the target reconciliation model's recognition accuracy meets the high standards required for automatic reconciliation confirmation of financial market clearing messages before being deployed in practical applications. Through continuous adjustment of model parameters and iterative training, the initial reconciliation model can continuously learn and optimize, ultimately acquiring the ability to process complex postscript data and providing financial institutions with efficient and accurate automatic reconciliation confirmation services.

[0082] The following describes in detail another optional implementation method.

[0083] Currently, financial market transactions flow into the back-end system for clearing after completion. The back-end system generates payment and receipt information based on the transaction schedule and processes payments and receipts according to the settlement time. For our receiving scenario, it's necessary to verify each clearing report against the original transaction to ensure the accuracy of transaction funds. Currently, the verification of clearing reports in the financial market relies heavily on precise matching of the report's date, amount, account number, and remarks. This is particularly problematic when clearing reports involve multiple transactions aggregated for payment, and the remarks depend on the clearing sender's input, leading to inaccurate matching. The payer might enter a transaction number, a payer's abbreviation / full name, a bond name or abbreviation, or the principal repayment amount, resulting in unsuccessful matching. Business personnel must manually identify the remarks and match them against our payment and receipt flow. Because the remarks from different counterparties lack a fixed format, the manual workload is significant. Automatic identification and processing are not feasible in this scenario, and reliance on manual confirmation is time-consuming, labor-intensive, and carries operational risks.

[0084] However, this application proposes a method for processing clearing messages. It pre-extracts remarks data samples from historical clearing messages and labels these samples. Based on the labeled remarks data samples, an initial write-off model is trained to obtain a target write-off model. After receiving a clearing message, the remarks data in the clearing message can be input into the target write-off model for recognition, thus obtaining the recognition result of the remarks data in the clearing message. Based on the recognition result of the target write-off model, an external transaction number table is queried to determine the target transaction record corresponding to the clearing message. The matching degree between the target transaction record and the clearing message is further verified. When the matching degree is greater than a matching threshold, the matching is confirmed as successful, and the status of the clearing message is updated to write-off confirmation complete. This achieves automatic write-off confirmation of clearing messages, reduces the input of human resources, improves the processing efficiency and accuracy of transaction matching for clearing messages, reduces operational risks, and achieves efficient automated processing.

[0085] The following section will further describe the processing method of the clearing message in the embodiments of this application.

[0086] Figure 3 This is a flowchart of another method for processing clearing messages according to an embodiment of this application. The method includes the following steps:

[0087] Step S301: Receive the settlement report.

[0088] In this embodiment, the bank or financial institution receives a clearing message that has not yet been confirmed for write-off. These messages may contain information about multiple transactions, and the remarks fields may be incomplete or inconsistently formatted, making it difficult for traditional automatic matching methods to process them accurately.

[0089] Step S302: Report and label the verification data.

[0090] In this embodiment, the financial institution processes the settlement messages that have not been fully cleared, mainly by tagging (marking) key information in the messages. This includes identifying and extracting information such as transaction date, amount, account number, and remarks, in order to prepare data for subsequent model training.

[0091] Step S303: Large model construction, training, and optimization.

[0092] In this embodiment, a deep learning model (large model) is constructed and trained based on the labeled data from step S302. This model is designed to understand the postscript content of clearing messages, even if the postscript information is ambiguous or non-standard. Through training on a large amount of data, the model can learn patterns of different clearing postscripts, improving the accuracy of recognition and matching.

[0093] Step S304, Large Model Interface.

[0094] In this embodiment, the trained large model is encapsulated into a callable interface for use by other components of the financial system. This means that any system receiving a clearing message can use this interface to pass the message data to the large model for intelligent analysis and processing.

[0095] Step S305: Execute the large model strategy.

[0096] In this embodiment, the large model performs in-depth analysis of the remarks in the clearing message to identify key information related to the transaction (such as transaction number, counterparty name, bond name, etc.). The model's identification results will be used to guide the subsequent transaction matching process.

[0097] Step S306: Confirm whether the reported reconciliation data has been reconciled.

[0098] In this embodiment, the system matches the settlement message with the actual transaction record. If the match is successful, confirming that the settlement message is completely consistent with a certain transaction, it indicates that the smart reconciliation confirmation has been completed, and step S307 is executed. If the match fails, the entire process ends.

[0099] Step S307: The status of the reported reconciliation data is set to reconciliation completed.

[0100] In this embodiment, if the clearing message is successfully cleared, the status of the clearing message is set to clearing complete. Then, the entire process ends.

[0101] In steps S301 to S307 above, the system automatically processes settlement reports that cannot be resolved by conventional means through large-scale model construction, report identification, strategy execution, transaction matching, and automatic reconciliation confirmation. This reduces human resource input, improves the efficiency and accuracy of back-end settlement report transaction matching, and realizes the function of efficient and automated processing of settlement report reconciliation and automatic reconciliation confirmation matching, saving labor costs and reducing operational risks.

[0102] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0103] Example 2

[0104] This application also provides a clearing message processing apparatus. It should be noted that the clearing message processing apparatus of this application can be used to execute the clearing message processing method provided in this application. The clearing message processing apparatus provided in this application is described below.

[0105] According to an embodiment of this application, an apparatus for implementing the above-described method for processing clearing messages is also provided, such as... Figure 4 As shown, the clearing message processing device 400 includes: a receiving unit 401, an identification unit 402, a searching unit 403, a determining unit 404, and an updating unit 405.

[0106] The receiving unit 401 is used to receive a clearing message, wherein the clearing message includes at least the transaction information and remarks data corresponding to the transaction to be cleared, and the remarks data is unstructured description information of the transaction to be cleared.

[0107] The identification unit 402 is used to input the postscript data in the clearing message into the target write-off model for identification, and obtain the identification result of the postscript data. The target write-off model is obtained by training the initial write-off model in advance using postscript data samples. The identification result includes at least the identification information related to the transaction to be written off.

[0108] The lookup unit 403 is used to search for a target transaction record that matches the identification result from the transaction number table based on the identification information related to the transaction to be reconciled included in the identification result, wherein the transaction number table stores at least one transaction record.

[0109] The determining unit 404 is used to determine the matching result between the target transaction record and the clearing message based on the identification result and the transaction information in the clearing message, wherein the matching result is used to indicate the degree of matching between the target transaction record and the clearing message.

[0110] Update unit 405 is used to update the write-off status corresponding to the clearing message to the write-off completed status in response to the matching result indicating that the target transaction record and the clearing message are successfully matched.

[0111] Optionally, the lookup unit 403 is further configured to: construct query conditions based on the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off, wherein the query conditions include at least one of the following: the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off; and, based on the query conditions, query the transaction number table for the target transaction record that matches the identification result.

[0112] Optionally, the determining unit 404 is further configured to: extract key elements corresponding to the transaction to be cancelled from the identification result and the transaction information in the clearing message, and extract key elements corresponding to the target transaction record from the target transaction record, wherein the key elements include at least the transaction currency, transaction date and transaction amount; and match the key elements corresponding to the transaction to be cancelled with the key elements corresponding to the target transaction record to obtain a matching result.

[0113] Optionally, the clearing message processing device 400 is further configured to: compare the degree of matching between the clearing message and the target transaction record with a matching threshold to obtain a comparison result; and, in response to the comparison result indicating that the degree of matching is greater than the matching threshold, determine that the target transaction record and the clearing message are successfully matched.

[0114] Optionally, the clearing message processing device 400 is further configured to: obtain postscript data samples from historical clearing messages; annotate the identification information related to the transaction samples to be reconciled in the postscript data samples to obtain label data; divide the annotated postscript data samples into a training set and a test set; train the initial reconciliation model using the training set to obtain the trained initial reconciliation model; verify the recognition accuracy of the trained initial reconciliation model using the test set to obtain a verification result; and, in response to the verification result indicating that the recognition accuracy is greater than the accuracy threshold, determine the trained initial reconciliation model as the target reconciliation model.

[0115] Optionally, the clearing message processing device 400 is further configured to: adjust the model parameters of the trained initial reconciliation model in response to the recognition accuracy not being greater than the accuracy threshold; and continue to train the adjusted initial reconciliation model using the training set until the recognition accuracy of the trained initial reconciliation model is greater than the accuracy threshold.

[0116] The clearing message processing apparatus provided in this application automatically identifies the postscript data in the clearing message by inputting it into the target reconciliation model. It can accurately extract the identification information related to the transaction to be reconciled from the free-format postscript data, thereby improving the accuracy of matching with transaction records in the transaction number table, reducing errors and omissions that may be introduced by traditional manual verification, realizing efficient processing of unstructured descriptive information, reducing the need for manual parsing, greatly improving the automation level and reconciliation efficiency of clearing message processing in the financial market, and thus solving the technical problem of low reconciliation efficiency of clearing messages.

[0117] It should be noted that the receiving unit 401, identification unit 402, searching unit 404, determining unit 404, and updating unit 405 mentioned above correspond to steps S201 to S205 in Embodiment 1. The five modules and their corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above modules can also be part of the device and run in the computer terminal 10 provided in Embodiment 1.

[0118] Example 3

[0119] Embodiments of this application may provide an electronic device. Figure 5 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 5 As shown, the electronic device may include: one or more ( Figure 5 Only one of the components is shown: processor 502, memory 504, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module, and display.

[0120] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the methods and apparatus in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the above-described methods. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0121] The processor can invoke information and application programs stored in memory via a transmission device to perform the following steps: receiving a settlement message, wherein the settlement message includes at least transaction information and remark data corresponding to the transaction to be reconciled, and the remark data is unstructured descriptive information of the transaction to be reconciled; inputting the remark data in the settlement message into a target reconciliation model for identification, and obtaining the identification result of the remark data, wherein the target reconciliation model is obtained by pre-training an initial reconciliation model using remark data samples, and the identification result includes at least identification information related to the transaction to be reconciled; based on the identification information related to the transaction to be reconciled included in the identification result, searching for a target transaction record that matches the identification result from a transaction number table, wherein the transaction number table stores at least one transaction record; based on the identification result and the transaction information in the settlement message, determining the matching result between the target transaction record and the settlement message, wherein the matching result is used to indicate the degree of matching between the target transaction record and the settlement message; and in response to the matching result indicating that the target transaction record and the settlement message are successfully matched, updating the reconciliation status corresponding to the settlement message to the reconciliation completed status.

[0122] The processor can also invoke information and applications stored in the memory via a transmission device to perform the following steps: constructing query conditions based on the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off, wherein the query conditions include at least one of the following information: the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off; and querying the transaction number table for the target transaction record that matches the identification result based on the query conditions.

[0123] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: extract the key elements corresponding to the transaction to be cancelled from the transaction information in the identification result and the settlement message, and extract the key elements corresponding to the target transaction record from the target transaction record, wherein the key elements include at least the transaction currency, transaction date and transaction amount; match the key elements corresponding to the transaction to be cancelled with the key elements corresponding to the target transaction record to obtain the matching result.

[0124] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: compare the matching degree of the clearing message with the target transaction record with a matching threshold to obtain a comparison result; in response to the comparison result indicating that the matching degree is greater than the matching threshold, determine that the target transaction record and the clearing message are successfully matched.

[0125] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: obtain postscript data samples from historical clearing messages; label the identification information related to the transaction samples to be reconciled in the postscript data samples to obtain label data; divide the labeled postscript data samples into training set and test set; train the initial reconciliation model using the training set to obtain the trained initial reconciliation model; use the test set to verify the recognition accuracy of the trained initial reconciliation model to obtain the verification result; in response to the verification result indicating that the recognition accuracy is greater than the accuracy threshold, determine the trained initial reconciliation model as the target reconciliation model.

[0126] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: in response to the recognition accuracy not being greater than the accuracy threshold, adjust the model parameters of the trained initial verification model; continue to train the adjusted initial verification model using the training set until the recognition accuracy of the trained initial verification model is greater than the accuracy threshold.

[0127] The present application provides a method for processing clearing messages. By inputting the remarks data in the clearing message into the target reconciliation model for automatic identification, it can accurately extract the identification information related to the transaction to be reconciled from the free-format remarks data, thereby improving the accuracy of matching with transaction records in the transaction number table, reducing errors and omissions that may be introduced by traditional manual verification, realizing efficient processing of unstructured descriptive information, reducing the need for manual parsing, greatly improving the automation level and reconciliation efficiency of financial market transaction clearing message processing, and thus solving the technical problem of low reconciliation efficiency of clearing messages.

[0128] Those skilled in the art will understand that Figure 5 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones (such as Android phones, iOS phones, etc.), tablets, PDAs, mobile internet devices (MIDs), personal access devices (PADs), and other terminal devices. Figure 5 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 5 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 5 The different configurations shown.

[0129] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0130] Example 4

[0131] Embodiments of this application also provide a computer-readable storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the clearing message processing method provided in Embodiment 1.

[0132] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0133] This application also provides a computer program product, which, when executed on a data processing device, is adapted to perform the processing method steps of a clearing message.

[0134] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0135] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0136] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0137] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0138] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0139] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0140] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for processing clearing messages, characterized in that, include: Receive a clearing message, wherein the clearing message includes at least transaction information and remarks data corresponding to the transaction to be cleared, and the remarks data is unstructured description information of the transaction to be cleared; The appendix data in the clearing message is input into the target write-off model for identification to obtain the identification result of the appendix data. The target write-off model is obtained by training an initial write-off model in advance using appendix data samples. The identification result includes at least the identification information related to the transaction to be written off. The identification information related to the transaction to be written off includes at least the identification information of the transaction to be written off, the identification information of the clearing object, and the identification information of the transaction product corresponding to the transaction to be written off. Based on the identification information related to the transaction to be reconciled included in the identification result, a target transaction record matching the identification result is searched from the transaction number table, wherein the transaction number table stores at least one transaction record; Based on the identification result and the transaction information in the clearing message, a matching result between the target transaction record and the clearing message is determined, wherein the matching result is used to indicate the degree of matching between the target transaction record and the clearing message; In response to the matching result indicating that the target transaction record and the clearing message are successfully matched, the write-off status corresponding to the clearing message is updated to the write-off completed status. Specifically, based on the identification information related to the transaction to be written off included in the identification result, the process of searching for a target transaction record matching the identification result from the transaction number table includes: constructing query conditions based on the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off, wherein the query conditions include at least one of the following: the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off; and searching for a target transaction record matching the identification result in the transaction number table based on the query conditions. Based on the identification result and the transaction information in the clearing message, determining the matching result between the target transaction record and the clearing message includes: extracting key elements corresponding to the transaction to be cancelled from the identification result and the transaction information in the clearing message, and extracting key elements corresponding to the target transaction record from the target transaction record, wherein the key elements include at least the transaction currency, transaction date, and transaction amount; matching the key elements corresponding to the transaction to be cancelled with the key elements corresponding to the target transaction record to obtain the matching result.

2. The method according to claim 1, characterized in that, The method further includes: The matching degree between the clearing message and the target transaction record is compared with a matching threshold to obtain the comparison result; In response to the comparison result indicating that the matching degree is greater than the matching threshold, it is determined that the target transaction record and the settlement message are successfully matched.

3. The method according to claim 1, characterized in that, The method further includes: Obtain the appendix data sample from historical settlement messages; The identification information related to the transaction sample to be verified in the remarks data sample is labeled to obtain the tag data; The labeled postscript data samples are divided into training and testing sets; The initial reimbursement model is trained using the training set to obtain the trained initial reimbursement model; Using the test set, the recognition accuracy of the trained initial reimbursement model is verified, and the verification results are obtained; In response to the verification result indicating that the recognition accuracy is greater than the accuracy threshold, the trained initial reimbursement model is determined as the target reimbursement model.

4. The method according to claim 3, characterized in that, The method further includes: In response to the recognition accuracy not being greater than the accuracy threshold, the model parameters of the trained initial verification model are adjusted; The adjusted initial verification model is further trained using the training set until the recognition accuracy of the trained initial verification model is greater than the accuracy threshold.

5. A processing apparatus for clearing messages, characterized in that, include: A receiving unit is configured to receive a clearing message, wherein the clearing message includes at least transaction information and remarks data corresponding to the transaction to be cleared, and the remarks data is unstructured description information of the transaction to be cleared; The identification unit is used to input the postscript data in the clearing message into the target write-off model for identification, and obtain the identification result of the postscript data. The target write-off model is obtained by training an initial write-off model in advance using postscript data samples. The identification result includes at least identification information related to the transaction to be written off. The identification information related to the transaction to be written off includes at least the identification information of the transaction to be written off, the identification information of the clearing object, and the identification information of the transaction product corresponding to the transaction to be written off. The lookup unit is used to search for a target transaction record that matches the identification result from a transaction number table based on the identification information related to the transaction to be reconciled included in the identification result, wherein the transaction number table stores at least one transaction record; The determining unit is configured to determine a matching result between the target transaction record and the clearing message based on the identification result and the transaction information in the clearing message, wherein the matching result is used to indicate the degree of matching between the target transaction record and the clearing message; The update unit is configured to update the write-off status corresponding to the clearing message to the write-off completed status in response to the matching result indicating that the target transaction record and the clearing message are successfully matched. The search unit is configured to search for a target transaction record matching the identification result from the transaction number table based on the identification information related to the transaction to be written off included in the identification result through the following steps: constructing query conditions based on the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off, wherein the query conditions include at least one of the following information: the identification information of the transaction to be written off, the identification information of the liquidation object, and the identification information of the transaction product corresponding to the transaction to be written off; and searching for a target transaction record matching the identification result in the transaction number table based on the query conditions. The determining unit is configured to determine the matching result between the target transaction record and the clearing message based on the identification result and the transaction information in the clearing message through the following steps: extracting key elements corresponding to the transaction to be cancelled from the identification result and the transaction information in the clearing message, and extracting key elements corresponding to the target transaction record from the target transaction record, wherein the key elements include at least the transaction currency, transaction date and transaction amount; matching the key elements corresponding to the transaction to be cancelled with the key elements corresponding to the target transaction record to obtain the matching result.

6. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the storage medium is located to perform the method according to any one of claims 1 to 4.

8. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method according to any one of claims 1 to 4.

Images