Transaction sink matching method and device based on natural language processing, and electronic equipment

Abstract

The invention discloses a transaction remittance path matching method and device based on natural language processing and electronic equipment, and relates to the field of financial science and technology, and the method comprises the steps: detecting whether a financial transaction result contains a transaction transaction order or not under the condition that a new financial transaction result and a transaction abstract are detected, and when the transaction transaction order is not detected, sending the transaction transaction order to a server; starting a confluence path self-matching mechanism, extracting transaction opponent information, transaction financial institution information and benefit object information in the transaction abstract by adopting a natural language processing model to obtain a transaction element set, and matching each transaction element in the transaction element set with a cash confluence path mapping table to obtain a matching result; and selecting the sink path which has the highest matching saturation and accords with the preset business rule as a transaction sink path, and adopting the transaction sink path to complete a transaction liquidation process. According to the method and the device, the technical problem of transaction clearing risk caused by influence on the automatic processing efficiency of the transaction due to incapability of automatically matching the confluence path for the financial transaction without the transaction order in the related technology is solved.

Description

Technical Field

[0001] The present invention relates to the field of financial technology or other related fields, and in particular to a transaction exchange matching method and device based on natural language processing, and an electronic device. Background Art

[0002] In the process of processing financial market transactions, once a transaction is concluded, its subsequent clearing and remittance operations need to be achieved through accurate remittance route matching. In related technologies, the matching of transaction remittance routes depends on the detailed information provided in the transaction transaction form, such as the account line, account number and account name, etc. For those transactions that are not accompanied by a transaction form or the transaction form information is incomplete when the transaction is concluded, if the mapping between the remittance route and the remittance route is not maintained, the remittance route cannot be automatically matched. The existing financial market backend system requires business personnel to manually maintain the remittance route parameters and remittance route mapping in the transaction preprocessing stage for the system to automatically match.

[0003] For transactions without a deal order, if the front-end transaction is completed late and close to the clearing time, the back-end system will not be able to automatically match the exchange route when processing, which will affect the automatic processing efficiency of the transaction and the subsequent clearing process. If it is not processed in time, the clearing window may be missed, and the resulting clearing risk is unpredictable.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present invention provide a transaction remittance matching method and device, and electronic device based on natural language processing, so as to at least solve the technical problem in the related art that for financial transactions without transaction orders, the remittance cannot be automatically matched, which affects the automatic processing efficiency of the transaction and leads to the risk of transaction settlement.

[0006] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a transaction remittance matching method based on natural language processing is provided, including: when a new financial transaction result and a transaction summary are detected, detecting whether the financial transaction result contains a transaction completion order, and when the transaction completion order is not detected, starting the remittance self-matching mechanism; when the remittance self-matching mechanism is started, using a natural language processing model to extract the counterparty information, transaction financial institution information and beneficiary information in the transaction summary to obtain a transaction element set, wherein the natural language processing model is a pre-trained model for parsing summary text, and the transaction financial institution information includes at least: transaction account bank, transaction intermediary bank; matching each transaction element in the transaction element set with the cash remittance mapping table, selecting the remittance with the highest matching saturation and meeting the preset business rules as the transaction remittance, and using the transaction remittance to complete the transaction settlement process.

[0007] Optionally, before detecting a new financial transaction result and a transaction summary, it also includes: when detecting the new financial transaction result, retrieving the front-end transaction information table, the associated system parameter table and the transaction details corresponding to each financial product in the financial transaction result; based on the first type of transaction element set, using a key element sorter to extract the transaction information in the front-end transaction information table to obtain a first transaction information set, wherein the first type of transaction element set includes at least one of the following: counterparty characters, transaction type characters, transaction amount characters, and transaction date characters; based on the second type of transaction element set, using a key element sorter to extract the transaction information in the associated system parameter table to obtain a second transaction information set, wherein the second type of transaction element set includes at least one of the following: account line characters, middle line characters, and account name characters; based on the third type of transaction element set, using a key element sorter to extract the transaction information in the transaction details to obtain a third transaction information set, wherein the third type of transaction element set includes at least one of the following: counterparty abbreviation characters, currency characters, and transaction settlement category characters; based on the first transaction information set, the second transaction information set, and the third transaction information set, generate the transaction summary.

[0008] Optionally, the step of using a natural language processing model to extract counterparty information, transaction financial institution information and beneficiary information in the transaction summary to obtain a set of transaction elements includes: using the natural language processing model to identify the transaction subject element entity in the transaction summary, obtain the transaction account and account name involving the counterparty element, and obtain the counterparty information; using the natural language processing model to identify the institution element entity in the transaction summary, obtain the account bank and intermediary bank of the associated financial institution element, and obtain the transaction financial institution information.

[0009] Optionally, the step of using a natural language processing model to extract counterparty information, transaction financial institution information and beneficiary information in the transaction summary to obtain a transaction element set also includes: detecting whether the counterparty information in the transaction element set contains the customer code SWI FTCODE of the account bank; when it is detected that the counterparty information does not contain the customer code SWI FTCODE of the account bank, generating beneficiary information based on the counterparty's abbreviation information, and storing the beneficiary information in a cash remittance route mapping table; or, when it is detected that the counterparty information contains the customer code SWI FTCODE of the account bank, using the customer code of the account bank as the beneficiary information.

[0010] Optionally, the step of matching each transaction element in the transaction element set in the cash remittance route mapping table, and selecting a remittance route with the highest matching saturation and complying with preset business rules as the transaction remittance route, further includes: when multiple candidate remittance routes are generated by matching each transaction element in the transaction element set with the cash remittance route mapping table, calculating the matching saturation of each candidate remittance route according to the matched element fields in the cash remittance route mapping table; sorting all the matching saturations to obtain a sorting result; and selecting the candidate remittance route with the highest matching saturation and complying with preset business rules in the sorting result as the transaction remittance route.

[0011] Optionally, the steps of using the transaction remittance route to complete the transaction settlement process include: sending the remittance route information of the transaction remittance route to the business end; when the business end has not confirmed the transaction remittance route, receiving the remittance route adjustment instruction transmitted by the business end; re-matching the transaction remittance route according to the transaction-related business type and business restriction information in the remittance route adjustment instruction, obtaining a new transaction remittance route, and sending the new transaction remittance route to the business end; when receiving the remittance route confirmation instruction from the business end, using the transaction remittance route to complete the transaction settlement process.

[0012] Optionally, after detecting whether the financial transaction result includes the transaction completion note, it also includes: when it is detected that the financial transaction result includes the transaction completion note, extracting the transaction account information, transaction financial institution information and transaction party information in the transaction completion note; generating a corresponding transaction remittance route and remittance route mapping based on the transaction account information, the transaction financial institution information and the transaction party information, and using the transaction remittance route to authorize the transaction; when the transaction authorization is obtained, initiating the payment and receipt flow based on the transaction confirmation information, generating a clearing message based on the payment and receipt flow, and sending the clearing message to the transaction counterparty to complete the transaction clearing process.

[0013] According to another aspect of an embodiment of the present invention, a transaction remittance matching device based on natural language processing is also provided, including: a transaction order detection unit, which is used to detect whether a new financial transaction result and a transaction summary contain a transaction transaction order when the transaction transaction order is detected, and to start the remittance self-matching mechanism when the transaction transaction order is not detected; a transaction element extraction unit, which is used to extract the counterparty information, transaction financial institution information and beneficiary information in the transaction summary by using a natural language processing model when the remittance self-matching mechanism is started, so as to obtain a transaction element set, wherein the natural language processing model is a pre-trained model for parsing summary text, and the transaction financial institution information includes at least: a transaction account bank and a transaction intermediary bank; a remittance matching unit, which is used to match each transaction element in the transaction element set with a cash remittance mapping table, select the remittance with the highest matching saturation and in accordance with preset business rules as the transaction remittance, and use the transaction remittance to complete the transaction settlement process.

[0014] Optionally, the transaction route matching device based on natural language processing also includes: an information table retrieval unit, which is used to retrieve the front-end transaction information table, the associated system parameter table and the transaction details corresponding to each financial product in the financial transaction result before detecting the new financial transaction result and the transaction summary, when the new financial transaction result is detected; a first information extraction unit, which is used to extract the transaction information in the front-end transaction information table based on the first type of transaction element set by using a key element sorter to obtain a first transaction information set, wherein the first type of transaction element set includes at least one of the following: transaction counterparty characters, transaction type characters, transaction amount characters, and transaction date characters; a second information extraction unit, which is used to extract the transaction information in the front-end transaction information table based on the second type of transaction element set by using a key element sorter to obtain a first transaction information set. A transaction element set is used to extract the transaction information in the associated system parameter table using a key element sorter to obtain a second transaction information set, wherein the second transaction element set includes at least one of the following: account line characters, middle line characters, and account name characters; a third information extraction unit is used to extract the transaction information in the transaction details based on the third transaction element set using a key element sorter to obtain a third transaction information set, wherein the third transaction element set includes at least one of the following: counterparty abbreviation characters, currency characters, and transaction settlement category characters; a transaction summary generation unit is used to generate the transaction summary based on the first transaction information set, the second transaction information set, and the third transaction information set.

[0015] Optionally, the transaction element extraction unit includes: a counterparty identification module, which is used to use the natural language processing model to identify the transaction subject element entity in the transaction summary, obtain the transaction account and account name involving the counterparty element, and obtain the counterparty information; an institutional entity identification module, which is used to use the natural language processing model to identify the institutional element entity in the transaction summary, obtain the account bank and intermediary bank of the associated financial institution element, and obtain the transaction financial institution information.

[0016] Optionally, the transaction element extraction unit also includes: a first detection module, used to detect whether the counterparty information in the transaction element set contains the customer code SWI FTCODE of the account bank; a beneficiary information generation module, used to generate beneficiary information based on the counterparty's abbreviation information when it is detected that the counterparty information does not contain the customer code SWI FTCODE of the account bank, and store the beneficiary information into a cash remittance route mapping table; or, when it is detected that the counterparty information contains the customer code SWI FTCODE of the account bank, use the customer code of the account bank as the beneficiary information.

[0017] Optionally, the remittance route matching unit includes: a matching saturation calculation module, which is used to calculate the matching saturation of each candidate remittance route according to the matched element fields in the cash remittance route mapping table when multiple candidate remittance routes are generated by matching each transaction element in the transaction element set with the cash remittance route mapping table; a sorting module, which is used to sort all the matching saturations to obtain a sorting result; and a remittance route selection module, which is used to select the candidate remittance route with the highest matching saturation in the sorting result and that meets preset business rules as the transaction remittance route.

[0018] Optionally, the route matching unit includes: a route sending module, which is used to send the route information of the transaction route to the business end; an adjustment indication receiving module, which is used to receive the route adjustment indication transmitted by the business end when the business end has not confirmed the transaction route; a route re-matching module, which is used to re-match the transaction route according to the transaction-related business type and business restriction information in the route adjustment indication, obtain a new transaction route, and send the new transaction route to the business end; a clearing module, which is used to use the transaction route to complete the transaction clearing process when receiving the route confirmation instruction from the business end.

[0019] Optionally, the transaction route matching device based on natural language processing also includes: a transaction order information extraction unit, which is used to extract the transaction account information, transaction financial institution information and transaction party information in the transaction order after detecting whether the financial transaction result contains the transaction order. When it is detected that the financial transaction result contains the transaction order, a route generation unit is used to extract the transaction account information, transaction financial institution information and transaction party information in the transaction order; generate corresponding transaction routes and route mappings according to the transaction account information, the transaction financial institution information and the transaction party information, and use the transaction route for transaction authorization; a clearing unit is used to start the payment and collection flow based on the transaction confirmation information when the transaction authorization is obtained, generate a clearing message based on the payment and collection flow, send the clearing message to the transaction counterparty, and complete the transaction clearing process.

[0020] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is further provided, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned transaction route matching methods based on natural language processing.

[0021] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, comprising one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement any one of the above-mentioned transaction route matching methods based on natural language processing.

[0022] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program, which, when executed by a processor, implements the steps of any one of the above-mentioned transaction route matching methods based on natural language processing.

[0023] In the present disclosure, when a new financial transaction result and a transaction summary are detected, it is possible to detect whether the financial transaction result contains a transaction completion note, and when no transaction completion note is detected, the remittance self-matching mechanism is started. When the remittance self-matching mechanism is started, a natural language processing model is used to extract counterparty information, transaction financial institution information and beneficiary information in the transaction summary to obtain a set of transaction elements, wherein the natural language processing model is a pre-trained model for parsing summary text, and the transaction financial institution information includes at least: transaction account bank, transaction intermediary bank, and each transaction element in the transaction element set is matched with the cash remittance mapping table, and the remittance route with the highest matching saturation and in compliance with preset business rules is selected as the transaction remittance route, and the transaction remittance route is used to complete the transaction settlement process.

[0024] Based on the above-mentioned public content, in the absence of a transaction order or incomplete transaction order information, key information in the transaction summary can be automatically extracted, and transaction remittances can be automatically matched to improve transaction processing efficiency. At the same time, it can ensure that the transaction clearing process is completed before the clearing deadline, avoid missing the clearing window due to untimely manual processing, and reduce the occurrence of transaction clearing risks, thereby solving the technical problem in related technologies that for financial transactions without transaction orders, the remittance routes cannot be automatically matched, which affects the automatic processing efficiency of transactions and leads to transaction clearing risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0026] Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing a transaction matching method based on natural language processing is shown;

[0027] Figure 2 is a flow chart of an optional transaction route matching method based on natural language processing according to an embodiment of the present invention;

[0028] Figure 3 is a flowchart of an optional automatic matching of transaction channels based on natural language processing according to an embodiment of the present invention;

[0029] Figure 4 is a schematic diagram of an optional transaction channel matching device based on natural language processing according to an embodiment of the present invention;

[0030] Figure 5 It is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0033] To facilitate those skilled in the art to understand the present invention, some terms or nouns involved in the embodiments of the present invention are explained below:

[0034] Transaction: A contract that specifies a set of terms and key elements agreed upon with a counterparty.

[0035] Payment and collection flow: In the financial market backend system, it is an information flow of payment or collection generated according to specific rules after a transaction is concluded, usually including information such as counterparty, delivery date, transaction currency, delivery amount, etc., which serves as the basis for subsequent settlement.

[0036] Remittance path: In the financial market backend system, it is the account number, opening bank, account number and other information of the parties involved in the clearing that is used after the transaction is completed. It is used to define the clearing path required for the clearing to be transferred from one party to another.

[0037] It should be noted that the transaction remittance matching method and device based on natural language processing in the present disclosure can be used in the field of financial technology (in the case of automatic matching of transaction remittances based on natural language processing), and can also be used in any field other than the field of financial technology (in the case of automatic matching of transaction remittances based on natural language processing). The present disclosure does not limit the application field of the transaction remittance matching method and device based on natural language processing.

[0038] It should be noted that the information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) collected by this disclosure are information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data are in compliance with the relevant laws, regulations and standards of the relevant regions, necessary confidentiality measures are taken, and public order and good customs are not violated, and corresponding operation portals are provided for users to choose to authorize or refuse. For example, an interface is set up between this system and relevant users or organizations. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0039] It should be noted that in the present disclosure, when collecting and analyzing customer information, corresponding operation entrances are provided for users to choose to agree or reject the automated decision results; if the user chooses to reject, the expert decision-making process is entered.

[0040] The following embodiments of the present invention can be applied to various systems / applications / devices for transaction exchange route matching based on natural language processing. The present invention automatically extracts and parses transaction information through natural language processing technology, without the need for manual maintenance of exchange route mapping, which greatly reduces the transaction preprocessing time, especially when processing transactions without transaction orders, and significantly improves the transaction processing efficiency. Automatic exchange route matching ensures that the transaction enters the clearing process in a timely manner, especially for situations where the transaction is completed close to the clearing deadline, which can effectively avoid the clearing risk caused by exchange route matching delays.

[0041] The present invention also provides a mechanism for adjusting the saturation of sink mapping according to business feedback, which can gradually optimize the matching logic and improve the matching accuracy.

[0042] It should be noted that financial transactions are concluded at the terminal front desk. After the transaction is concluded, it flows into the back-end system of the financial institution. After the back-end system accesses the transaction, it will pre-process the transaction. During the pre-processing, it will match the remittance route, match the confirmation routing rules, match the clearing routing rules, match the accounting entity and other information. After the matching is completed, the transaction is authorized, the corresponding accounts are calculated, and the corresponding confirmation is generated. After the confirmation is sent to the counterparty and the confirmation sent by the counterparty is matched, the payment and receipt flow will be started, and then a clearing message will be generated and sent to the counterparty. The calculated accounts and the cleared accounts will be checked in both directions to complete the entire transaction process.

[0043] The present invention is described in detail below in conjunction with various embodiments.

[0044] Embodiment 1

[0045] According to an embodiment of the present invention, an embodiment of a transaction route matching method based on natural language processing is provided. It should be noted that the steps shown in the flowchart of 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 can be executed in an order different from that shown here.

[0046] The transaction route matching method embodiment based on natural language processing provided in the first embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 The hardware structure block diagram of a computer terminal (or mobile device) for implementing a transaction matching method based on natural language processing is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more ( Figure 1 , 102a, 102b, ..., 102n are used to illustrate) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU (Microcontroller Unit) or a programmable logic device FPGA (Field Programmable Gate Array)), 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 interface), a Universal Serial Bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. Those skilled in the art can understand that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0047] It should be noted that the one or more processors 102 and / or other data processing circuits described above may generally be referred to herein as "data processing circuits". The data processing circuits may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuit may be a single independent processing module, or may be incorporated in whole or in part into any of the other components in the computer terminal 10 (or mobile device). As described in the embodiments of the present application, the data processing circuit acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0048] 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 transaction exchange matching method based on natural language processing in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, realizing the above-mentioned transaction exchange matching method based on natural language processing. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the computer terminal 10 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0049] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0050] The display may be, for example, a touch screen liquid crystal display (LCD), which may enable a user to interact with a user interface of the computer terminal 10 (or mobile device).

[0051] It should be noted that, compared with the prior art, in some cases, financial transactions may not be accompanied by detailed transaction notes, or the information on the transaction transaction notes is not sufficient to automatically match the remittance route. The present invention extracts key information from the transaction summary and combines the existing counterparty and currency information in the system to automatically match or generate cash remittance routes, so that transactions can enter the clearing process in a timely and accurate manner. At the same time, in a clearing environment with huge transaction volumes and time sensitivity, the traditional manual matching remittance route method may not be able to meet the needs due to low efficiency. The present invention uses an automated remittance route matching mechanism to not only improve the speed of transaction processing, but also reduce errors caused by human factors, thereby improving the efficiency and accuracy of the entire clearing process.

[0052] Furthermore, this embodiment can also quickly identify key information in the transaction summary when the transaction is close to the clearing deadline, automatically generate and match the remittance path, so that the transaction enters the clearing process in time to avoid missing the clearing window.

[0053] It should be noted that this embodiment can be applied to the backend system / equipment of a financial institution.

[0054] Under the above operating environment, this application provides Figure 2 The transaction route matching method based on natural language processing is shown. Figure 2 is a flowchart of an optional transaction path matching method based on natural language processing according to an embodiment of the present invention. Figure 2 As shown, the method comprises the following steps:

[0055] Optionally, before detecting a new financial transaction result and a transaction summary, when a new financial transaction result is detected, the front-end transaction information table, the associated system parameter table and the transaction details corresponding to each financial product in the financial transaction result are retrieved; based on the first type of transaction element set, a key element sorter is used to extract the transaction information in the front-end transaction information table to obtain a first transaction information set, wherein the first type of transaction element set includes at least one of the following: counterparty characters, transaction type characters, transaction amount characters, and transaction date characters; based on the second type of transaction element set, a key element sorter is used to extract the transaction information in the associated system parameter table to obtain a second transaction information set, wherein the second type of transaction element set includes at least one of the following: account line characters, middle line characters, and account name characters; based on the third type of transaction element set, a key element sorter is used to extract the transaction information in the transaction details to obtain a third transaction information set, wherein the third type of transaction element set includes at least one of the following: counterparty abbreviation characters, currency characters, and transaction settlement category characters; based on the first transaction information set, the second transaction information set, and the third transaction information set, a transaction summary is generated.

[0056] The steps of retrieving the front-end transaction information table, the associated system parameter table and the transaction details corresponding to each financial product in the financial transaction result refer to that after the new financial transaction result is received or generated, the system will automatically retrieve the three key data sources related to the transaction. Among them, the front-end transaction information table may include but is not limited to the detailed terms and conditions of the transaction, such as the type of transaction, the counterparty, the transaction amount, the currency, the transaction date, etc., which are crucial for the subsequent transaction processing and remittance matching. The associated system parameter table records the parameters related to transaction settlement, such as the SWI FTCODE (the customer code of the account bank) and the full English name of the account bank and the intermediary bank, which are the basis for generating cash remittance mapping. The transaction details record more detailed transaction information records, including the abbreviation of the counterparty, the transaction currency, the specific clearing category and other information, which may not be fully included in the transaction order, but are crucial for generating transaction summaries and remittance mapping.

[0057] In this embodiment, the key element sorter can be used to extract the transaction type, counterparty, transaction amount, currency, transaction date and other information from the front-end transaction information table to form a first transaction information set, which can be used to directly or indirectly generate the remittance information required for the transaction. The key element sorter can also be used to extract the SWI FTCODE (customer code of the account line) and English full name of the account bank and the intermediary bank from the associated system parameter table to form a second transaction information set, which is used to generate complete remittance information, especially when the counterparty has maintained the SWI FTCODE, accurate remittance information can be quickly generated. In addition, in this embodiment, the key element sorter can also be used to extract the counterparty abbreviation characters, currency characters and transaction settlement category characters from the transaction details to form a third transaction information set.

[0058] Afterwards, a transaction summary can be generated based on the first transaction information set, the second transaction information set and the third transaction information set. Specifically, all key transaction information extracted from the three data sources are integrated to generate a comprehensive transaction summary. The transaction summary contains the core elements of the transaction, such as the counterparty, transaction currency, clearing category, and the SWIFTCODE and name of the financial institution. The generation of the transaction summary provides a clear and structured information basis for the subsequent automatic remittance matching. Even in the case of incomplete transaction information or no transaction order, accurate remittance matching can be performed based on natural language processing technology. In an automated manner, the key information of the transaction is extracted and integrated from multiple data sources to provide more comprehensive and accurate data support for remittance matching. The role of the key element sorter is to automatically identify and extract specific information related to the transaction remittance, such as the account bank, intermediary bank, SWIFTCODE, account name, counterparty abbreviation, transaction currency, etc., so that the system can generate a structured transaction summary based on this information. This structured information integration method not only simplifies the transaction processing process and reduces the need for manual intervention, but also improves the accuracy and efficiency of information processing, especially for transactions with incomplete information. Through the key information in the transaction summary, combined with natural language processing technology, it can automatically complete the exchange matching, avoiding delays and errors in manual processing, thereby reducing the risk of missing the clearing window and ensuring the smooth progress of transactions.

[0059] It should be noted that in the process of generating transaction summaries, it is necessary to take into account the characteristics and specific details of various transactions, as well as possible missing or uncertain information. In this embodiment, the key element sorter can accurately identify and extract different transaction elements, and at the same time have certain intelligent processing capabilities, such as processing fuzzy information in the text, identifying synonyms or abbreviations, etc., to ensure that the generated transaction summary can fully and accurately reflect the key information of the transaction, and provide a solid data foundation for subsequent exchange matching and transaction processing.

[0060] Step S201, when a new financial transaction result and a transaction summary are detected, whether the financial transaction result contains a transaction completion order is detected, and when no transaction completion order is detected, the exchange self-matching mechanism is started.

[0061] In the financial market backend system, whenever a new financial transaction result and a transaction summary are identified or received, the processing flow of the present invention will be started first. The system will automatically detect whether the financial transaction result contains a detailed transaction order, which usually contains all the detailed information of the transaction, such as the two parties to the transaction, the transaction amount, the currency, the clearing path, etc. If the transaction result contains a transaction order, the system will automatically match the exchange route according to the information in the transaction order; however, if the transaction result does not contain a transaction order or the transaction order information is incomplete, the system will automatically start the exchange route self-matching mechanism based on natural language processing, automatically parse the key elements of the transaction through the transaction summary information, and then generate or match the exchange route, which is crucial to improve the processing efficiency of transactions without transaction orders and reduce manual intervention.

[0062] Step S202, when the Huilu self-matching mechanism is activated, a natural language processing model is used to extract the counterparty information, transaction financial institution information and beneficiary information in the transaction summary to obtain a set of transaction elements, wherein the natural language processing model is a pre-trained model for parsing summary text, and the transaction financial institution information includes at least: transaction account bank, transaction intermediary bank.

[0063] After starting the self-matching mechanism of the exchange route, the pre-trained natural language processing model can be used to deeply analyze the transaction summary. The model extracts the transaction element set by identifying key entities in the summary text, such as the counterparty, the financial institutions involved in the transaction (including the correspondent bank and the intermediary bank), and the beneficiary. It should be noted that the counterparty information may include the name of the counterparty; the transaction financial institution information at least includes the names of the account bank and the intermediary bank, which play a key role in the transaction settlement; the beneficiary information refers to the entity to which the transaction funds ultimately flow. Through this step, the system can extract specific key information that can be used for exchange route matching from the abstract transaction summary.

[0064] Optionally, step S202 includes: using a natural language processing model to identify the transaction subject element entity in the transaction summary, obtain the transaction account and account name involving the counterparty element, and obtain the counterparty information; using a natural language processing model to identify the institution element entity in the transaction summary, obtain the account bank and intermediary bank of the associated financial institution element, and obtain the transaction financial institution information.

[0065] It should be noted that the natural language processing model selected in this embodiment can be a BERT model, through which counterparty information and transaction financial institution information are extracted from the transaction summary. Here, the transaction subject element entity includes key information such as the name and account of the counterparty. By analyzing the text, the model can identify and extract the counterparty's transaction account and account name, thereby obtaining detailed information of the counterparty. This process ensures that the counterparty can be identified as accurately as possible even when the information is incomplete by parsing the semantic relationship in the summary. Similarly, the institutional element entities in the transaction summary are identified by the natural language processing model. These entities involve financial institutions used in the transaction, such as the correspondent bank and the intermediary bank. The model will identify and extract the name information of these financial institutions, and then obtain the account bank and intermediary bank information involved in the transaction.

[0066] After extracting the transaction element set, further process the counterparty information to generate beneficiary information. Optionally, step S202 also includes: detecting whether the counterparty information in the transaction element set contains the customer code SWIFTCODE of the account bank; if it is detected that the counterparty information does not contain the customer code SWIFTCODE of the account bank, generate the beneficiary information according to the counterparty's abbreviation information, and store the beneficiary information in the cash remittance route mapping table; or, if it is detected that the counterparty information contains the customer code SWIFTCODE of the account bank, use the customer code of the account bank as the beneficiary information.

[0067] Among them, when detecting the SWIFTCODE in the counterparty information, the system will automatically detect whether the counterparty information contains the SWIFTCODE, that is, the customer code of the account bank. SWIFTCODE is a code widely used in inter-bank communication and clearing internationally. Its existence or not is directly related to the efficiency and accuracy of remittance matching. If the counterparty information contains the SWIFTCODE, then this code is directly used as the beneficiary information and stored in the cash remittance mapping table for subsequent remittance matching. The existence of SWIFTCODE ensures the certainty and efficiency of remittance information. On the contrary, if the SWIFTCODE is not detected, the system will generate the beneficiary information based on the counterparty's abbreviation information and store it in the cash remittance mapping table. This process may require intelligent matching in combination with historical data, transaction habits, etc. to generate the beneficiary information as accurately as possible to ensure that the transaction can proceed smoothly even in the case of incomplete information.

[0068] By extracting transaction elements, especially information about counterparties and financial institutions, through natural language processing (NLP) technology, the entities and relationships involved in the transaction can be accurately identified. The above natural language processing model is pre-trained. During the model training phase, a large amount of annotated financial transaction data, including different types of transaction summaries, is used to ensure the generalization and accuracy of the model.

[0069] Step S203, matching each transaction element in the transaction element set with the cash remittance route mapping table, selecting the remittance route with the highest matching saturation and meeting the preset business rules as the transaction remittance route, and using the transaction remittance route to complete the transaction settlement process.

[0070] After extracting the transaction element set, the system will compare and match these transaction elements with the cash remittance route mapping table in the system through an intelligent matching algorithm. The cash remittance route mapping table is a database that stores various remittance route information and matching rules, including relevant information of account banks, intermediary banks, counterparties and clearing paths. The system will calculate the matching saturation, that is, the degree of consistency between the transaction element and the preset remittance route information, based on the matching degree between each element in the transaction element set and the mapping table. Generally speaking, the higher the matching saturation, the more accurate the matching between the transaction element and a certain remittance route. The system will select the remittance route with the highest matching saturation and that meets the preset business rules (such as transaction type, currency restrictions, etc.) as the transaction remittance route. Once the transaction remittance route is determined, the system will automatically complete the transaction clearing process according to the selected remittance route, including generating a clearing message, sending it to the counterparty, and performing account verification until the transaction clearing is successfully completed. This ensures that even when the transaction information is incomplete, the system can complete the remittance route matching and transaction clearing with high efficiency and accuracy, thereby improving the processing efficiency and security of financial market transactions.

[0071] Optionally, step S203 also includes: when each transaction element in the transaction element set is matched with the cash remittance route mapping table to generate multiple candidate remittance routes, calculating the matching saturation of each candidate remittance route according to the matched element field in the cash remittance route mapping table; sorting all matching saturations to obtain a sorting result; and selecting the candidate remittance route with the highest matching saturation in the sorting result and meeting the preset business rules as the transaction remittance route.

[0072] Among them, when matching each transaction element in the transaction element set in the cash remittance route mapping table, it is necessary to select the remittance route with the highest matching saturation and in compliance with the preset business rules as the transaction remittance route. The matching process and remittance route selection involved specifically include: the generation of multiple candidate remittance routes and the calculation of matching saturation, that is, when the information in the transaction element set matches the records in the cash remittance route mapping table, multiple candidate remittance routes may be generated. For each candidate remittance route, the system will calculate its matching saturation according to the number and quality of the matched element fields (such as counterparty, currency, account line SWIFTCODE, etc.). The higher the matching saturation, the higher the matching degree of the remittance route with the transaction element, and the more it can meet the needs of transaction settlement. Then, the matching saturation sorting and remittance route selection are implemented, and the system sorts the matching saturation of all candidate remittance routes to obtain a sorting result. The sorting is usually performed in the order of matching saturation from high to low. In the sorting result, the system selects the candidate remittance route with the highest matching saturation and in compliance with the preset business rules as the transaction remittance route. Preset business rules may include but are not limited to transaction type, currency, business scope of the account bank, etc., to ensure that the selected remittance route not only technically matches the transaction elements but is also reasonable in terms of business logic.

[0073] Through the above implementation steps, the system can automatically identify the most appropriate exchange route, reduce the need for manual intervention, and improve the efficiency and accuracy of transaction processing. The calculation and sorting logic of matching saturation ensures that even when the information is incomplete or there are multiple matching options, the system can still make accurate decisions and select the most appropriate exchange route for transaction settlement.

[0074] Through the above steps, when a new financial transaction result and a transaction summary are detected, it can be detected whether the financial transaction result contains a transaction order, and when the transaction order is not detected, the remittance self-matching mechanism is started. When the remittance self-matching mechanism is started, a natural language processing model is used to extract the counterparty information, transaction financial institution information and beneficiary information in the transaction summary to obtain a transaction element set, wherein the natural language processing model is a pre-trained model for parsing summary texts, and the transaction financial institution information includes at least: transaction account bank, transaction intermediary bank, and each transaction element in the transaction element set is matched with the cash remittance mapping table, and the remittance route with the highest matching saturation and in accordance with the preset business rules is selected as the transaction remittance route, and the transaction remittance route is used to complete the transaction settlement process. In this embodiment, when there is no transaction order or the transaction order information is incomplete, key information in the transaction summary can be automatically extracted, and the transaction route can be automatically matched to improve the transaction processing efficiency. At the same time, it can ensure that the transaction clearing process is completed before the clearing deadline, avoid missing the clearing window due to untimely manual processing, and reduce the occurrence of transaction clearing risks, thereby solving the technical problem in the related technology that for financial transactions without transaction orders, the route cannot be automatically matched, which affects the automatic processing efficiency of the transaction and leads to transaction clearing risks.

[0075] Optionally, when the transaction remittance route is used to complete the transaction settlement process, this embodiment includes: sending the remittance route information of the transaction remittance route to the business end; when the business end has not confirmed the transaction remittance route, receiving the remittance route adjustment instruction transmitted by the business end; re-matching the transaction remittance route according to the transaction-related business type and business restriction information in the remittance route adjustment instruction, obtaining a new transaction remittance route, and sending the new transaction remittance route to the business end; when receiving the remittance route confirmation instruction from the business end, using the transaction remittance route to complete the transaction settlement process.

[0076] Among them, the steps of using transaction remittance route to complete the transaction settlement process specifically include the interaction process with the business end. First, the system sends the automatically matched transaction remittance route information to the business end, and the business end personnel make the final confirmation. The confirmation process is an important part of the transaction settlement process, ensuring the accuracy of the transaction remittance route and the compliance of the transaction. If the business end fails to confirm the automatically matched transaction remittance route, the system will receive the remittance route adjustment instructions transmitted by the business end, which may include additional conditions such as the transaction-related business type and business restriction information. According to these adjustment instructions, the system will re-match the transaction remittance route to obtain a new transaction remittance route that better meets business needs. This process may involve re-querying and matching the cash remittance route mapping table, as well as recalculating and sorting the matching saturation. When the business end confirms the transaction remittance route, the system will automatically start the transaction settlement process based on the confirmed transaction remittance route information. This includes steps such as generating a settlement message, executing collection and payment operations, and sending a settlement message to the counterparty until the transaction settlement is successfully completed.

[0077] It should be noted that in actual operation, the confirmation and adjustment of transaction exchange routes is a dynamic process, which may require multiple iterations to be finalized. It can respond quickly according to the feedback and instructions from the business side, adjust the matching strategy, and ensure that the selection of transaction exchange routes not only meets the technical matching requirements, but also complies with the business side's operating habits and compliance requirements. At the same time, the system can also record and analyze the reasons for each exchange route adjustment, continuously optimize the matching algorithm, and improve the accuracy and efficiency of subsequent automatic matching of transaction exchange routes. At the same time, it improves the automation level of transaction processing, enhances the flexibility and business adaptability of the system, and improves the speed and accuracy of financial market transaction processing.

[0078] Optionally, after detecting whether the financial transaction result includes the transaction completion note, it also includes: when it is detected that the financial transaction result includes the transaction completion note, extracting the transaction account information, transaction financial institution information and transaction party information in the transaction completion note; generating corresponding transaction remittances and remittance mappings based on the transaction account information, transaction financial institution information and transaction party information, and using the transaction remittance to authorize the transaction; when the transaction is authorized, initiating the payment and receipt flow based on the transaction confirmation information, generating a clearing message based on the payment and receipt flow, and sending the clearing message to the counterparty to complete the transaction clearing process.

[0079] When it is detected that a new financial transaction result contains a transaction order, key transaction account information, transaction financial institution information and transaction party information can be directly extracted from the transaction order. This information forms the basis for generating the transaction route. Then, the corresponding transaction route and route mapping are generated based on this information. This is an important step in the transaction settlement process, which defines the specific path of how transaction funds are transferred from one party to another. Once the transaction route is generated, the system will authorize the transaction, which is a necessary step before the transaction is settled to confirm the legality and compliance of the transaction. After the transaction is authorized successfully, the payment and collection flow is initiated based on the transaction confirmation information, that is, the specific operation of the transaction settlement is initiated, including generating a settlement message and sending it to the counterparty, and finally completing the settlement process of the transaction.

[0080] Specifically, when it is detected that a financial transaction is accompanied by a transaction note, the backend system can directly extract the complete transaction account information, transaction financial institution information and transaction party information from the transaction note, which not only simplifies the process of generating the remittance route, but also improves the accuracy and completeness of the transaction information. The transaction transaction note usually contains all the details of the transaction, including the transaction amount, currency, transaction date, detailed information of the counterparty, etc., and is an important voucher for transaction settlement. The system automatically generates the transaction remittance route based on the transaction account information, transaction financial institution information and transaction party information. This process may involve querying historical remittance routes, matching clearing rules and verifying the information of both parties. The generated transaction remittance route will be stored in the cash remittance route mapping table as the basis for subsequent automatic matching of remittance routes. At the same time, the generated remittance route mapping will cover all necessary information of the transaction to ensure the smooth progress of the settlement process.

[0081] Transaction authorization is a key step before transaction settlement. It ensures the legality and compliance of the transaction and is the prerequisite for transaction settlement. After successful authorization, the system will start the payment flow based on the transaction confirmation information. The confirmation information is the confirmation of the transaction details by both parties, including the transaction amount, currency, transaction date, etc., which is the basis for starting the transaction settlement. The system will generate specific payment instructions based on the confirmation information and start the settlement process.

[0082] Furthermore, after the payment and collection flow is initiated, the system will generate a clearing message, which is an electronic file containing detailed information on the transaction clearing and is an important communication medium in the transaction clearing process. The clearing message will be sent to the counterparty to confirm the clearing details of the transaction. At the same time, the generation and sending of the clearing message also marks the beginning of the transaction clearing process. The entire clearing process will be carried out according to the content of the clearing message, including steps such as fund transfer and account verification, until the transaction is cleared.

[0083] In actual applications, the clearing process may need to be customized according to different transaction types, counterparties, clearing rules and other factors. Therefore, the system needs to have certain learning and optimization capabilities, and be able to continuously adjust the strategies for generating transaction routes and clearing messages based on historical transaction data and clearing results to improve clearing efficiency and accuracy. In addition, the system should also provide monitoring and analysis tools for the clearing process to help business personnel understand the clearing status, handle abnormal situations in a timely manner, and ensure the smooth progress of the clearing process.

[0084] The embodiment of the present invention provides an efficient and accurate natural language processing-based automatic matching and clearing process initiation solution for transaction routes. Whether it is a transaction with a deal order or a transaction without a deal order, the system can intelligently analyze the transaction summary information, generate or match the accurate transaction route, automatically authorize the transaction, generate a clearing message and send it to the counterparty, and complete the entire transaction clearing process. It not only improves the automation level of transaction processing, but also reduces the clearing risk caused by incomplete information or manual operation errors, which is of great value for the optimization of financial market transaction processing systems.

[0085] Another optional specific implementation is described in detail below.

[0086] To solve the problem of not being able to automatically match the remittance route, this embodiment provides a solution for automatically extracting summary information in transactions based on a natural language processing model, and automatically matching the most accurate remittance route based on the summary information. Specifically, the information extraction component involved in this embodiment is first described.

[0087] 1. Key element sorter.

[0088] By sorting out the front-end transaction information tables corresponding to different products, the summary information corresponding to the corresponding fields is obtained. The Correspondent Bank keyword in the summary information extracts the account bank information, and the Intermediary Bank keyword extracts the intermediary bank information. The account bank and intermediary bank information in the clearing counterparty in the system parameter table are associated to obtain the corresponding SWI FTCODE (account bank customer code) or English full name (account bank account name) and deposit it into the cash remittance route table. The counterparty is obtained from the transaction details, and the clearing counterparty in the associated system table is obtained to obtain the corresponding SWI FTCODE (account bank customer code) or English full name (account bank account name) and deposit it into the cash remittance route table. The remittance route mapping is generated according to the counterparty's English abbreviation and currency, the clearing subcategory of the transaction and other information.

[0089] 2. Key element parser.

[0090] Natural language processing (NLP) algorithms are often required to identify key information and sentiment in transactions.

[0091] For the implementation mode of the present invention, a named entity recognition method can be used (of course, a naive Bayes algorithm, a support vector machine SVM algorithm, etc. can also be used) to obtain key information such as the account line, intermediary bank, account number, etc. in the summary (the account line information needs to be entered, and the intermediary bank and account number information are optional), combined with the existing counterparty, currency and other related information in the transaction (already existing in the transaction information table), generate cash remittance information and remittance matching information mapping, and call the transaction preprocessing process to complete the remittance matching.

[0092] 3. Huilu matching engine.

[0093] According to the results of summary extraction in the key element parser, such as account line, account number, counterparty, currency and other information, the remittance route information is matched. On the one hand, the account line, intermediary bank and beneficiary generated according to the counterparty in the extraction result are used to generate a cash remittance route according to the existing system logic. If the counterparty maintains the SWIFTCODE, the beneficiary information is the SWIFT type. If not maintained, the English abbreviation of the counterparty is taken as the beneficiary's account name and stored in the cash remittance route. On the other hand, the counterparty, currency and clearing subcategory in the transaction in the extraction result are used as matching conditions to generate a matching mapping and insert it into the existing cash remittance route mapping table.

[0094] When a transaction matches a remittance route, it is determined based on the matching degree of the matching fields in the mapping table. If multiple matches are possible, they are sorted according to the matching saturation, and the ones with higher saturation are given priority to the business. After the match, manual authorization is performed, and the business confirms whether the remittance route is acceptable. If it is not what the business wants, the transaction can be restarted to match the remittance route that the business wants to match by adjusting the remittance route mapping saturation (increasing restrictions on business categories and business types, etc.), gradually optimizing the matching logic and matching accuracy. If no matching remittance route is found based on the matching information, a to-do list is generated and handled by the business.

[0095] Figure 3 is a flowchart of an optional automatic matching of transaction channels based on natural language processing according to an embodiment of the present invention. Figure 3 As shown, including:

[0096] 31. When a new transaction inflow is detected, check whether the financial transaction is accompanied by a transaction note. If there is a transaction note, the transaction note information already contains detailed information such as the account line and account number, which can be automatically matched or a matching remittance path can be generated. If it does not contain a transaction note and the business has not maintained the mapping in advance, it is necessary to use the self-matching mechanism to complete the automatic matching of the remittance path.

[0097] 32. When pre-processing a transaction, it is necessary to match the remittance route and the clearing route rules. If the remittance route does not match, manual transaction authorization will be implemented, and the subsequent matching of the clearing route rules will also fail. In the clearing route rules, the message type to be generated is matched according to the transaction and remittance route information. Here, it is necessary to use the key element sorter to classify the transaction elements, implement the transaction element parsing through the key element parser, and implement the transaction remittance route matching through the remittance route matching engine.

[0098] 33. Verify whether the account line information exists through the content obtained from the summary information. If it does not exist, the remittance route cannot be generated. If it exists, the counterparty account line and beneficiary information are generated based on the account line information and counterparty information, and the corresponding matching mapping is generated. The transaction is manually authorized to the business. The remittance route information can be displayed to the business on the remittance route sheet page, and the accuracy of the extraction results can be gradually learned and optimized.

[0099] 34. If no useful element information can be extracted based on the summary information, a to-do task will be generated in the system to remind the business that the transaction has not been matched to the remittance route and requires manual maintenance. The transaction will also be manually authorized.

[0100] Based on the above technical solution, the present invention provides an automatic matching device for transaction remittances based on natural language processing, which can solve the problem of untimely manual processing when the transaction cannot be automatically matched to the remittance, and can automatically generate a matching mapping between remittances and remittances according to summary information, complete automatic matching of transaction remittances, and automatically complete subsequent clearing processes, effectively reducing the clearing risk caused by missing the clearing deadline.

[0101] The following is a detailed description in conjunction with another embodiment.

[0102] Embodiment 2

[0103] A transaction route matching device based on natural language processing provided in this embodiment includes multiple implementation units, each of which corresponds to each implementation step in the above-mentioned embodiment 1. Its specific implementation methods and beneficial effects can refer to the above-mentioned method embodiments and will not be repeated here.

[0104] Figure 4 is a schematic diagram of an optional transaction path matching device based on natural language processing according to an embodiment of the present invention, such as Figure 4 As shown, the transaction route matching device based on natural language processing may include: a transaction order detection unit 41, a transaction element extraction unit 42, and a route matching unit 43.

[0105] The transaction order detection unit 41 is used to detect whether the financial transaction result contains a transaction order when a new financial transaction result and a transaction summary are detected, and to start the exchange self-matching mechanism when no transaction order is detected.

[0106] The transaction element extraction unit 42 is used to extract the counterparty information, transaction financial institution information and beneficiary information in the transaction summary by using a natural language processing model when the exchange self-matching mechanism is started, so as to obtain a set of transaction elements, wherein the natural language processing model is a pre-trained model for parsing summary text, and the transaction financial institution information includes at least: transaction account bank and transaction intermediary bank.

[0107] The remittance route matching unit 43 is used to match each transaction element in the transaction element set with the cash remittance route mapping table, select the remittance route with the highest matching saturation and in compliance with the preset business rules as the transaction remittance route, and use the transaction remittance route to complete the transaction settlement process.

[0108] The above-mentioned transaction remittance route matching device based on natural language processing can detect whether the financial transaction result contains a transaction transaction order through the transaction order detection unit 41 when a new financial transaction result and transaction summary are detected, and start the remittance route self-matching mechanism when the transaction transaction order is not detected. When the transaction element extraction unit 42 starts the remittance route self-matching mechanism, the natural language processing model is used to extract the counterparty information, transaction financial institution information and beneficiary information in the transaction summary to obtain a transaction element set, wherein the natural language processing model is a pre-trained model for parsing summary text, and the transaction financial institution information includes at least: transaction account bank, transaction intermediary bank, and each transaction element in the transaction element set is matched with the cash remittance route mapping table through the remittance route matching unit 43, and the remittance route with the highest matching saturation and in accordance with the preset business rules is selected as the transaction remittance route, and the transaction remittance route is used to complete the transaction settlement process. In this embodiment, when there is no transaction order or the transaction order information is incomplete, key information in the transaction summary can be automatically extracted, and the transaction route can be automatically matched to improve the transaction processing efficiency. At the same time, it can ensure that the transaction clearing process is completed before the clearing deadline, avoid missing the clearing window due to untimely manual processing, and reduce the occurrence of transaction clearing risks, thereby solving the technical problem in the related technology that for financial transactions without transaction orders, the route cannot be automatically matched, which affects the automatic processing efficiency of the transaction and leads to transaction clearing risks.

[0109] Optionally, the transaction route matching device based on natural language processing also includes: an information table retrieval unit, which is used to retrieve the front-end transaction information table, the associated system parameter table and the transaction details corresponding to each financial product in the financial transaction result before detecting the new financial transaction result and the transaction summary, when the new financial transaction result is detected; a first information extraction unit, which is used to extract the transaction information in the front-end transaction information table based on the first type of transaction element set by using a key element sorter to obtain a first transaction information set, wherein the first type of transaction element set includes at least one of the following: counterparty characters, transaction type characters, transaction amount characters, and transaction date characters; a second information extraction unit, which is used to extract the transaction information in the front-end transaction information table based on the first type of transaction element set by using a key element sorter to obtain a first transaction information set. The second type of transaction element set uses a key element sorter to extract transaction information in the associated system parameter table to obtain a second transaction information set, wherein the second type of transaction element set includes at least one of the following: account line characters, middle line characters, account name characters; a third information extraction unit is used to extract transaction information in transaction details based on the third type of transaction element set using a key element sorter to obtain a third transaction information set, wherein the third type of transaction element set includes at least one of the following: counterparty abbreviation characters, currency characters and transaction settlement category characters; a transaction summary generation unit is used to generate a transaction summary based on the first transaction information set, the second transaction information set and the third transaction information set.

[0110] Optionally, the transaction element extraction unit includes: a counterparty identification module, which is used to use a natural language processing model to identify the transaction subject element entity in the transaction summary, obtain the transaction account and account name involving the counterparty element, and obtain the counterparty information; an institutional entity identification module, which is used to use a natural language processing model to identify the institutional element entity in the transaction summary, obtain the account bank and intermediary bank of the associated financial institution element, and obtain the transaction financial institution information.

[0111] Optionally, the transaction element extraction unit also includes: a first detection module, used to detect whether the counterparty information in the transaction element set contains the customer code SWI FTCODE of the account bank; a beneficiary information generation module, used to generate beneficiary information based on the counterparty's abbreviation information when it is detected that the counterparty information does not contain the customer code SWI FTCODE of the account bank, and store the beneficiary information into the cash remittance route mapping table; or, when it is detected that the counterparty information contains the customer code SWI FTCODE of the account bank, use the customer code of the account bank as the beneficiary information.

[0112] Optionally, the remittance route matching unit includes: a matching saturation calculation module, which is used to calculate the matching saturation of each candidate remittance route according to the matched element fields in the cash remittance route mapping table when multiple candidate remittance routes are generated by matching each transaction element in the transaction element set with the cash remittance route mapping table; a sorting module, which is used to sort all matching saturations to obtain a sorting result; and a remittance route selection module, which is used to select the candidate remittance route with the highest matching saturation in the sorting result and that meets the preset business rules as the transaction remittance route.

[0113] Optionally, the route matching unit includes: a route sending module, which is used to send the route information of the transaction route to the business end; an adjustment indication receiving module, which is used to receive the route adjustment indication transmitted by the business end when the business end has not confirmed the transaction route; a route re-matching module, which is used to re-match the transaction route according to the transaction-related business type and business restriction information in the route adjustment indication, obtain a new transaction route, and send the new transaction route to the business end; a clearing module, which is used to use the transaction route to complete the transaction clearing process when receiving the route confirmation instruction from the business end.

[0114] Optionally, the transaction route matching device based on natural language processing also includes: a transaction order information extraction unit, which is used to extract the transaction account information, transaction financial institution information and transaction party information in the transaction order after detecting whether the financial transaction result contains the transaction order. When it is detected that the financial transaction result contains the transaction order, a route generation unit is used to extract the transaction account information, transaction financial institution information and transaction party information in the transaction order; generate corresponding transaction routes and route mappings according to the transaction account information, transaction financial institution information and transaction party information, and use the transaction route for transaction authorization; a clearing unit is used to start the payment and collection flow based on the transaction confirmation information when the transaction authorization is obtained, generate a clearing message based on the payment and collection flow, send the clearing message to the counterparty, and complete the transaction clearing process.

[0115] The above-mentioned transaction route matching device based on natural language processing may also include a processor and a memory. The above-mentioned transaction order detection unit 41, transaction element extraction unit 42, route matching unit 43, etc. are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize the corresponding functions.

[0116] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the automatic matching of transaction channels based on natural language processing can be achieved by adjusting kernel parameters.

[0117] The above-mentioned memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (fl ash RAM), and the memory includes at least one storage chip.

[0118] Embodiment 3

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

[0120] Among them, the memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the transaction exchange matching method and device based on natural language processing in the embodiment of the present application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, realizing the above-mentioned transaction exchange matching method based on natural language processing. The memory may include a high-speed random access memory, and may also include a 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 a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0121] The processor can call the information and application stored in the memory through the transmission device to perform the following steps: when a new financial transaction result and transaction summary are detected, detect whether the financial transaction result contains a transaction completion order, and when the transaction completion order is not detected, start the remittance self-matching mechanism; when the remittance self-matching mechanism is started, use a natural language processing model to extract the counterparty information, transaction financial institution information and beneficiary information in the transaction summary to obtain a transaction element set, wherein the natural language processing model is a pre-trained model for parsing summary text, and the transaction financial institution information includes at least: transaction account bank, transaction intermediary bank; match each transaction element in the transaction element set with the cash remittance mapping table, select the remittance route with the highest matching saturation and in accordance with the preset business rules as the transaction remittance route, and use the transaction remittance route to complete the transaction settlement process.

[0122] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: before detecting a new financial transaction result and a transaction summary, when a new financial transaction result is detected, call the front-end transaction information table, the associated system parameter table and the transaction details corresponding to each financial product in the financial transaction result; based on the first type of transaction element set, use a key element sorter to extract the transaction information in the front-end transaction information table to obtain a first transaction information set, wherein the first type of transaction element set includes at least one of the following: counterparty characters, transaction type characters, transaction amount characters, and transaction date characters; based on the second type of transaction element set, use a key element sorter to extract the transaction information in the associated system parameter table to obtain a second transaction information set, wherein the second type of transaction element set includes at least one of the following: account line characters, middle line characters, and account name characters; based on the third type of transaction element set, use a key element sorter to extract the transaction information in the transaction details to obtain a third transaction information set, wherein the third type of transaction element set includes at least one of the following: counterparty abbreviation characters, currency characters, and transaction settlement category characters; based on the first transaction information set, the second transaction information set, and the third transaction information set, generate a transaction summary.

[0123] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: using a natural language processing model to identify the transaction subject element entity in the transaction summary, obtain the transaction account and account name involving the counterparty element, and obtain the counterparty information; using a natural language processing model to identify the institution element entity in the transaction summary, obtain the account bank and intermediary bank of the associated financial institution element, and obtain the transaction financial institution information.

[0124] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: detect whether the counterparty information in the transaction element set contains the customer code SWI FTCODE of the account bank; when it is detected that the counterparty information does not contain the customer code SWI FTCODE of the account bank, generate beneficiary information based on the counterparty's abbreviation information, and store the beneficiary information in the cash remittance route mapping table; or, when it is detected that the counterparty information contains the customer code SWI FTCODE of the account bank, use the customer code of the account bank as the beneficiary information.

[0125] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: when each transaction element in the transaction element set is matched with the cash remittance route mapping table to generate multiple candidate remittance routes, the matching saturation of each candidate remittance route is calculated according to the matched element fields in the cash remittance route mapping table; all matching saturations are sorted to obtain a sorting result; and the candidate remittance route with the highest matching saturation in the sorting result and that meets the preset business rules is selected as the transaction remittance route.

[0126] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: sending the transaction remittance information of the transaction remittance to the business end; when the business end has not confirmed the transaction remittance, receiving the remittance adjustment instruction transmitted by the business end; re-matching the transaction remittance according to the transaction-related business type and business restriction information in the remittance adjustment instruction, obtaining a new transaction remittance, and sending the new transaction remittance to the business end; when receiving the remittance confirmation instruction from the business end, using the transaction remittance to complete the transaction settlement process.

[0127] The processor can also call the information and application programs stored in the memory through the transmission device to perform the following steps: after detecting whether the financial transaction result contains a transaction completion order, if it is detected that the financial transaction result contains a transaction completion order, extract the transaction account information, transaction financial institution information and transaction party information in the transaction completion order; generate corresponding transaction remittances and remittance mappings based on the transaction account information, transaction financial institution information and transaction party information, and use the transaction remittance to authorize the transaction; when the transaction is authorized, start the payment and collection flow based on the transaction confirmation information, generate a clearing message based on the payment and collection flow, and send the clearing message to the counterparty to complete the transaction clearing process.

[0128] It can be understood by those skilled in the art that Figure 5 The structure shown is for illustration only, and the electronic device may also be a terminal device such as a smart phone, a tablet computer, a PDA, a mobile Internet device (Mobile Internet Devices, MID), a PAD, etc. Figure 5 The structure of the electronic device is not limited. Figure 5 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 5 Different configurations are shown.

[0129] A person of ordinary skill in the art can understand that all or part of the steps in the various natural language processing-based transaction route matching methods in the above-mentioned embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, and the storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0130] Embodiment 4

[0131] The embodiment of the present application also provides a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the transaction exchange matching method based on natural language processing provided in the first embodiment.

[0132] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is further provided, the computer-readable storage medium including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the transaction route matching methods based on natural language processing in the above-mentioned embodiment 1.

[0133] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0134] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the transaction route matching method based on natural language processing described in each embodiment of the present application.

[0135] The present application also provides a computer program product, including a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the transaction route matching method based on natural language processing described in each embodiment of the present application are implemented.

[0136] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0137] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0138] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0139] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0140] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0141] If the integrated unit is implemented in the form of 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 the present invention, 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, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0142] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A transaction path matching method based on natural language processing, characterized in that: include: When a new financial transaction result and a transaction summary are detected, detecting whether the financial transaction result contains a transaction order, and when the transaction order is not detected, starting the exchange self-matching mechanism; In the case of starting the exchange self-matching mechanism, a natural language processing model is used to extract the counterparty information, transaction financial institution information and beneficiary information in the transaction summary to obtain a transaction element set, wherein the natural language processing model is a pre-trained model for parsing summary text, and the transaction financial institution information at least includes: transaction account bank, transaction intermediary bank; Each transaction element in the transaction element set is matched with the cash remittance route mapping table, and the remittance route with the highest matching saturation and conforming to the preset business rules is selected as the transaction remittance route, and the transaction remittance route is used to complete the transaction settlement process.

2. The transaction exchange matching method according to claim 1, characterized in that: Before a new financial transaction result is detected, along with a transaction summary, it also includes: When the new financial transaction result is detected, the foreground transaction information table, the associated system parameter table and the transaction details corresponding to each financial product in the financial transaction result are retrieved; Based on the first type of transaction element set, a key element sorter is used to extract the transaction information in the foreground transaction information table to obtain a first transaction information set, wherein the first type of transaction element set includes at least one of the following: a transaction counterparty character, a transaction type character, a transaction amount character, and a transaction date character; Based on the second type of transaction element set, a key element sorter is used to extract the transaction information in the associated system parameter table to obtain a second transaction information set, wherein the second type of transaction element set includes at least one of the following: account line characters, middle line characters, and account name characters; Based on the third type of transaction element set, a key element sorter is used to extract transaction information in the transaction details to obtain a third transaction information set, wherein the third type of transaction element set includes at least one of the following: a counterparty abbreviation character, a currency character, and a transaction settlement category character; The transaction summary is generated based on the first transaction information set, the second transaction information set, and the third transaction information set.

3. The transaction exchange matching method according to claim 1, characterized in that: The step of extracting the counterparty information, the transaction financial institution information and the beneficiary information in the transaction summary by using a natural language processing model to obtain a transaction element set includes: Using the natural language processing model to identify the transaction subject element entity in the transaction summary, obtain the transaction account and account name involved in the counterparty element, and obtain the counterparty information; The natural language processing model is used to identify the institutional element entities in the transaction summary, obtain the account bank and the intermediate bank of the associated financial institution element, and obtain the transaction financial institution information.

4. The transaction exchange matching method according to claim 3, characterized in that: The step of extracting the counterparty information, the transaction financial institution information and the beneficiary information in the transaction summary by using a natural language processing model to obtain a transaction element set also includes: Detect whether the counterparty information in the transaction element set includes the customer code SWIFTCODE of the account line; If it is detected that the counterparty information does not contain the customer code SWIFTCODE of the account bank, generate beneficiary information according to the counterparty's short name information, and store the beneficiary information in the cash remittance route mapping table; or, When it is detected that the counterparty information contains the customer code SWIFTCODE of the account bank, the customer code of the account bank is used as the beneficiary information.

5. The transaction exchange matching method according to claim 1, characterized in that: The step of matching each transaction element in the transaction element set in the cash remittance route mapping table and selecting a remittance route with the highest matching saturation and conforming to preset business rules as the transaction remittance route also includes: When a plurality of candidate remittance routes are generated by matching each transaction element in the transaction element set with the cash remittance route mapping table, the matching saturation of each candidate remittance route is calculated according to the matched element fields in the cash remittance route mapping table; Sorting all the matching saturations to obtain a sorting result; The candidate remittance route with the highest matching saturation and conforming to the preset business rules in the sorting results is selected as the transaction remittance route.

6. The transaction exchange matching method according to claim 1, characterized in that: The steps of completing the transaction settlement process using the transaction exchange route include: Sending the transaction information to the service end; In the case that the business end has not confirmed the transaction remittance route, receiving a remittance route adjustment instruction transmitted by the business end; Re-matching the transaction remittance route according to the transaction-related business type and business restriction information in the remittance route adjustment instruction to obtain a new transaction remittance route, and sending the new transaction remittance route to the business end; When receiving the remittance confirmation instruction from the business end, the transaction remittance is used to complete the transaction settlement process.

7. The transaction exchange matching method according to claim 1, characterized in that: After detecting whether the financial transaction result includes a transaction order, the method further includes: When it is detected that the financial transaction result includes the transaction transaction sheet, extracting the transaction account information, transaction financial institution information and transaction party information in the transaction transaction sheet; Generate a corresponding transaction remittance route and remittance route mapping according to the transaction account information, the transaction financial institution information and the transaction party information, and use the transaction remittance route to perform transaction authorization; When the transaction is authorized, the payment and collection flow is started based on the transaction confirmation information, a settlement message is generated based on the payment and collection flow, and the settlement message is sent to the counterparty to complete the transaction settlement process.

8. A transaction channel matching device based on natural language processing, characterized in that: include: A transaction order detection unit is used to detect whether a transaction order is included in the financial transaction result when a new financial transaction result and a transaction summary are detected, and to start the exchange-route self-matching mechanism when the transaction order is not detected; a transaction element extraction unit, configured to extract the counterparty information, transaction financial institution information and beneficiary information in the transaction summary by using a natural language processing model when the remittance self-matching mechanism is activated, so as to obtain a transaction element set, wherein the natural language processing model is a pre-trained model for parsing summary text, and the transaction financial institution information at least includes: a transaction account bank and a transaction intermediary bank; The remittance route matching unit is used to match each transaction element in the transaction element set with the cash remittance route mapping table, select the remittance route with the highest matching saturation and in compliance with the preset business rules as the transaction remittance route, and use the transaction remittance route to complete the transaction settlement process.

9. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the transaction route matching method based on natural language processing as described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the transaction route matching method based on natural language processing described in any one of claims 1 to 7 are implemented.

Images