Reconciliation method and device of financial institution and storage medium

Abstract

The invention discloses an account checking method and device for a financial institution and a storage medium. Relates to the field of artificial intelligence, and the method comprises the steps: obtaining online payment information and income and expenditure detail information of a target account type under the condition of obtaining the authorization of a target user; detecting whether the number of transaction data in the online payment information is equal to the number of transaction data in the income and expenditure detail information, and obtaining a number detection result; determining a range detection result corresponding to the online payment information and the revenue and expenditure detail information by using a batch reconciliation detection model under the condition that the number detection result is equal; and matching the transaction data of the online payment information and the income and expenditure detail information in the target number interval according to the range detection result to obtain a target account checking result. Through the method and the device, the problem of relatively low account checking efficiency of a financial institution in related technologies is solved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and more specifically, to a reconciliation method, apparatus, and storage medium for a financial institution. Background Technology

[0002] Secure accounts, a type of financial institution account designed specifically for certain projects or purposes, aim to ensure the transparency and security of account funds through strict fund management and usage rules. Online payments, as a convenient transaction method, have been widely used in various scenarios. However, post-payment reconciliation—verifying the consistency between payment information and the account's actual income and expenditure records—is a crucial step in ensuring fund security and compliance.

[0003] In current technology, financial institutions primarily rely on manual reconciliation, requiring staff to manually compare and reconcile transactions. However, with the increasing sophistication of payment technologies and the surge in payment data, existing reconciliation methods are showing their limitations, struggling to ensure efficiency when dealing with large volumes of transaction data.

[0004] There is currently no effective solution to the problem of low reconciliation efficiency for financial institutions in related technologies. Summary of the Invention

[0005] The main objective of this application is to provide a reconciliation method, apparatus, and storage medium for financial institutions to solve the problem of low reconciliation efficiency in related technologies.

[0006] To achieve the above objectives, according to one aspect of this application, a reconciliation method for financial institutions is provided. The method includes: obtaining online payment information and transaction details for a target account type, with authorization from a target user, wherein both the online payment information and transaction details include multiple transaction records; detecting whether the number of transactions in the online payment information and the number of transactions in the transaction details are equal, obtaining a transaction count detection result; if the transaction count detection result indicates that the number of transactions is equal, using a batch reconciliation detection model to determine a range detection result corresponding to the online payment information and transaction details, wherein the range detection result includes a target number range of transaction records where there are inconsistencies between the online payment information and the transaction details; and matching the transaction records in the online payment information and transaction details within the target number range based on the range detection result, obtaining a target reconciliation result.

[0007] Furthermore, the process of determining the range detection results corresponding to online payment information and transaction details using a batch reconciliation detection model includes: splitting the online payment information and transaction details into multiple transaction number intervals to obtain payment information splitting results and transaction detail splitting results, where each transaction number interval's payment information splitting result and transaction detail splitting result includes a preset number of transaction data; extracting features from each transaction number interval's payment information splitting result and transaction detail splitting result to obtain multiple interval features, where each interval feature has a one-to-one correspondence with the number of transactions; inputting each interval feature into the batch reconciliation detection model, and using the batch reconciliation detection model to output the interval detection results corresponding to the interval features, where the interval detection results indicate that the transaction data in the interval is consistent or inconsistent; determining the target transaction number interval based on the transaction number interval to which the interval feature with inconsistent transaction data belongs, and determining the range detection results based on the target transaction number interval.

[0008] Furthermore, the transaction data includes transaction amount and transaction account. Feature extraction is performed on the payment information splitting results and detailed information splitting results for each transaction number interval to obtain multiple interval features, including: for any transaction number interval, the payment information splitting results and detailed information splitting results are used to determine a first mean amount feature and a first variance amount feature based on the transaction amount of multiple transaction data in the payment information splitting results; the transaction amount of multiple transaction data in the detailed information splitting results is used to determine a second mean amount feature and a second variance amount feature; the transaction account numbers of multiple transaction data in the payment information splitting results are used to determine a first account number percentage feature and a first account number total value feature; the transaction account numbers of multiple transaction data in the detailed information splitting results are used to determine a second account number percentage feature and a second account number total value feature; the first mean amount feature, the first variance amount feature, the second mean amount feature, the second variance amount feature, the first account number percentage feature, the first account number total value feature, the second account number percentage feature, and the second account number total value feature are used to determine interval features corresponding to the transaction number interval; and multiple interval features are determined based on the interval features corresponding to multiple transaction number intervals.

[0009] Furthermore, the training steps of the batch reconciliation detection model include: obtaining a batch reconciliation training dataset, wherein the batch reconciliation training dataset includes multiple batch samples and labels for each batch sample, the batch samples include online payment samples and income and expenditure detail samples, and each online payment sample and income and expenditure detail sample includes a preset number of transaction data samples; extracting features from each batch sample to obtain multiple sample features, wherein the sample features correspond one-to-one with the batch samples; and training a preset machine learning model using the multiple sample features and the labels corresponding to the sample features to obtain the batch reconciliation detection model.

[0010] Furthermore, based on the range detection results, the online payment information and transaction details are matched within the target number of transactions to obtain the target reconciliation results, which include: determining multiple target payment transaction data in key-value pair format based on the online payment information within the target number of transactions, wherein the key-value pair includes a key name and a key value; determining multiple target detail transaction data in key-value pair format based on the transaction details within the target number of transactions; and matching the key values ​​of target payment transaction data and target detail transaction data with the same key name to obtain the target reconciliation results.

[0011] Furthermore, after obtaining the transaction count detection result by checking whether the number of transactions in the online payment information and the number of transactions in the income and expenditure details information are equal, the method further includes: if the transaction count detection result is that the number of transactions is not equal, deleting the transaction data in the online payment information that is after a first preset time to obtain the corrected online payment information; deleting the transaction data in the income and expenditure details information that is before a second preset time to obtain the corrected income and expenditure details information; and correcting the transaction count detection result to be equal if the number of transactions in the corrected online payment information and the number of transactions in the corrected income and expenditure details information are equal.

[0012] To achieve the above objectives, according to another aspect of this application, a reconciliation device for a financial institution is provided. The device includes: an information acquisition unit, configured to acquire online payment information and transaction details of a target account type, provided that authorization is obtained from a target user; wherein both the online payment information and the transaction details include multiple transaction data; a transaction count detection unit, configured to detect whether the number of transactions in the online payment information and the number of transactions in the transaction details are equal, and obtain a transaction count detection result; an interval determination unit, configured to determine a range detection result corresponding to the online payment information and transaction details using a batch reconciliation detection model, provided that the transaction count detection result indicates that the number of transactions is equal; wherein the range detection result includes a target transaction count interval where there is inconsistency between the online payment information and the transaction details; and a reconciliation matching unit, configured to match the transaction data in the online payment information and the transaction details within the target transaction count interval based on the range detection result, and obtain a target reconciliation result.

[0013] Furthermore, the interval determination unit includes: an information splitting module, used to split online payment information and transaction details information separately to obtain payment information splitting results and transaction details splitting results for multiple transaction number intervals, wherein each transaction number interval's payment information splitting result and transaction details splitting result includes a preset number of transaction data; a feature extraction module, used to extract features from the payment information splitting results and transaction details splitting results for each transaction number interval to obtain multiple interval features, wherein the interval features and transaction numbers have a one-to-one correspondence; a model prediction module, used to input each interval feature into a batch reconciliation detection model, and use the batch reconciliation detection model to output the interval detection result corresponding to the interval feature, wherein the interval detection result indicates that the interval transaction data is consistent or inconsistent; and a result determination module, used to determine the target transaction number interval based on the transaction number interval to which the interval feature with inconsistent transaction data belongs, and to determine the range detection result based on the target transaction number interval.

[0014] Furthermore, the transaction data includes transaction amount and transaction account. The feature extraction module includes: an amount feature submodule, used to determine the first amount mean feature and the first amount variance feature based on the transaction amounts of multiple transaction data in the payment information splitting results and detailed information splitting results for any number of transactions; a number feature submodule, used to determine the first account number percentage feature and the first account number total value feature based on the transaction accounts of multiple transaction data in the payment information splitting results; a second account number percentage feature and the second account number total value feature based on the transaction accounts of multiple transaction data in the detailed information splitting results; and an interval feature determination submodule, used to determine the interval features corresponding to the number of transactions interval based on the first amount mean feature, the first amount variance feature, the second amount mean feature, the second amount variance feature, the first account number percentage feature, the first account number total value feature, the second account number percentage feature, and the second account number total value feature; and to determine multiple interval features based on the interval features corresponding to multiple number of transactions intervals.

[0015] Furthermore, the device also includes: a dataset acquisition unit for acquiring a batch reconciliation training dataset, wherein the batch reconciliation training dataset includes multiple batch samples and labels for each batch sample, the batch samples include online payment samples and income and expenditure detail samples, and each online payment sample and income and expenditure detail sample includes a preset number of transaction data samples; a sample feature extraction unit for extracting features from each batch sample to obtain multiple sample features, wherein the sample features and batch samples have a one-to-one correspondence; and a model training unit for training a preset machine learning model using the multiple sample features and the labels corresponding to the sample features to obtain a batch reconciliation detection model.

[0016] Furthermore, the reconciliation matching unit includes: a first determining module, used to determine multiple target payment transaction data in key-value pair format based on transaction data in the target number range of online payment information, wherein the key-value pair includes a key name and a key value; a second determining module, used to determine multiple target detail transaction data in key-value pair format based on transaction data in the target number range of income and expenditure details information; and a key-value matching module, used to match the key values ​​of target payment transaction data and target detail transaction data that belong to the same key name to obtain the target reconciliation result.

[0017] Furthermore, the device also includes: a first deletion unit, used to detect whether the number of transactions in the online payment information and the number of transactions in the income and expenditure details information are equal, and after obtaining the transaction count detection result, if the transaction count detection result is not equal, delete the transaction data in the online payment information that is after a first preset time to obtain the corrected online payment information; a second deletion unit, used to delete the transaction data in the income and expenditure details information that is before a second preset time to obtain the corrected income and expenditure details information; and a result correction unit, used to correct the transaction count detection result to be equal if the number of transactions in the corrected online payment information and the number of transactions in the corrected income and expenditure details information are equal.

[0018] According to another aspect of this application, a computer-readable storage medium is provided, which includes a stored program, wherein, when the program is executed, it controls the device on which the computer-readable storage medium is located to perform any reconciliation method of a financial institution.

[0019] According to another aspect of this application, an electronic device is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs include methods for performing reconciliation of any financial institution.

[0020] According to another aspect of this application, a computer program product is provided, including computer instructions that, when executed by a processor, implement the steps of the reconciliation method for a financial institution as described above.

[0021] In this embodiment, with the authorization of the target user, online payment information and transaction details of the target account type are obtained. Both online payment information and transaction details include multiple transaction data. The number of transactions in the online payment information and the number of transactions in the transaction details are checked to obtain a transaction count detection result. If the transaction count is equal, a batch reconciliation detection model is used to determine the range detection result corresponding to the online payment information and transaction details. The range detection result includes a target number range of transaction data where there is inconsistency between the online payment information and the transaction details. Based on the range detection result, the transaction data in the online payment information and transaction details within the target number range are matched to obtain the target reconciliation result. This solves the technical problem of low reconciliation efficiency for financial institutions in the prior art.

[0022] By checking whether the number of transactions in online payment information and the number of transactions in income and expenditure details are equal, the number of transactions is counted, thus determining the result of the first level of detection. If the number of transactions is equal, a batch reconciliation detection model is used to determine the range detection result corresponding to the online payment information and income and expenditure details, narrowing the matching range to the target number interval within the range detection result, thus determining the result of the second level of detection. Based on the range detection result, the transaction data within the target number interval is matched to obtain the target reconciliation result, thus determining the result of the third level of detection. Through these three levels of detection, initial screening is performed based on the number of transactions, and the batch reconciliation detection model quickly locates the target number interval. Matching is then performed based on the target number interval, avoiding the time-consuming process of matching all transaction data sequentially, thereby improving reconciliation efficiency. Attached Figure Description

[0023] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0024] Figure 1 A hardware block diagram of a computer terminal for implementing a reconciliation method for financial institutions is shown.

[0025] Figure 2 This is a flowchart of a reconciliation method for financial institutions provided according to an embodiment of this application;

[0026] Figure 3 This is a schematic diagram of a financial institution's reconciliation device according to an embodiment of this application;

[0027] Figure 4 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0028] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0030] It should be noted that the information collected in this application (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for display, data used for analysis, etc.) are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of this data all comply with relevant laws, regulations, and standards, necessary confidentiality measures have been taken, and they do not violate public order and good morals. Corresponding access points are provided for users to choose to authorize or refuse. For example, interfaces are set up between this system and relevant users or organizations, providing users with corresponding access points to choose to agree to or refuse automated decision-making results; if the user chooses to refuse, the process proceeds to the expert decision-making stage.

[0031] Example 1

[0032] According to an embodiment of this application, a method embodiment for reconciliation of accounts by a financial institution is also provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0033] The method embodiment provided in Embodiment 1 of this application can be executed on a mobile terminal, computer terminal, or similar computing device. Figure 1A hardware block diagram of a computer terminal (or mobile device) for implementing reconciliation methods used by financial institutions is shown. Figure 1 As shown, the computer terminal 10 (or mobile device) may include one or more processors 102 (shown as 102a, 102b, ..., 102n in the figure) 102 (processor 102 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0034] It should be noted that the aforementioned one or more processors 102 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 10 (or mobile device). As involved in the embodiments of this application, the data processing circuits serve as a processor control mechanism (e.g., selection of a variable resistor termination path connected to an interface).

[0035] 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 reconciliation method of financial institutions in this embodiment of the application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned reconciliation method of financial institutions. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0036] The transmission device 106 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 10. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 106 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0037] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 10 (or mobile device).

[0038] Under the aforementioned operating environment, this application provides the following: Figure 2 The reconciliation methods of financial institutions are shown. Figure 2 This is a flowchart of a reconciliation method for financial institutions according to Embodiment 1 of this application.

[0039] Step S201: With the authorization of the target user, obtain the online payment information and transaction details of the target account type.

[0040] It should be noted that both online payment information and transaction details include multiple transaction records. The number of transactions in online payment information can refer to the channel number, while the number of transactions in transaction details can refer to the transaction serial number. For example, online payment information may include 5,000 transactions using channel numbers 1 to 5000, and transaction details may include 5,000 transactions using transaction serial numbers 1 to 5000.

[0041] Optionally, the target user is the user who generates online payment information and transaction details. Authorization can be obtained from the target user through methods such as clicking authorization in a mobile application before generating and recording the online payment information and transaction details, and the target user agrees to have their online payment information and transaction details used for reconciliation. The target account type can be a managed sub-account (i.e., a secure account). Online payment information for the day can be obtained through the online payment system before 24:00 each day, and transaction details for the day can be obtained through the account transaction details program interface.

[0042] Step S202: Check whether the number of transactions in the online payment information and the number of transactions in the income and expenditure details are equal, and obtain the transaction count detection result.

[0043] Optionally, after obtaining online payment information and transaction details, the len function of the Python programming language can be used to calculate the number of transactions A in the online payment information and the number of transactions B in the transaction details. When the number of transactions A and the number of transactions B are equal, the transaction count detection result is that the number of transactions is equal; otherwise, the transaction count detection result is that the number of transactions is not equal, and the target reconciliation result is output as the number of transactions in the online payment information and the transaction details is inconsistent.

[0044] Step S203: If the number of transactions is equal, use the batch reconciliation detection model to determine the range detection result corresponding to the online payment information and income and expenditure details.

[0045] It should be noted that the range detection result includes the target number range of transactions where there are inconsistencies between online payment information and transaction details. When the number of the target number range in the range detection result is zero, the output target reconciliation result is that the online payment information and transaction details are consistent.

[0046] Optionally, the batch reconciliation detection model can be a model trained based on machine learning models such as logistic regression and support vector machines. This model is used to predict whether there are inconsistencies between online payment transaction data within a certain number of transactions (e.g., 1001-2000) and transaction data of income and expenditure details within the same number of transactions (e.g., 1001-2000). The batch reconciliation detection model does not directly locate specific numbers of transactions; instead, it performs consistency checks only on a batch of transaction data within a certain number of transactions.

[0047] For example, online payment information includes 5,000 transaction data from channel number 1 to channel number 5000, and income and expenditure details include 5,000 transaction data from transaction serial number 1 to transaction serial number 5000. By using a batch reconciliation detection model, a range detection result is obtained. The range detection result can include at least one target number range, which can be 1001-2000 (that is, there are inconsistent transaction data between online payment information and income and expenditure details within the target number range of 1001-2000; in other words, there are inconsistent transaction data between transaction data from channel number 1000 to channel number 2000 and transaction data from transaction serial number 1001 to transaction serial number 2000. For example, the batch reconciliation detection model outputs the above target number range because the transaction data from channel number 1005 is inconsistent with the transaction data from transaction serial number 1005).

[0048] Step S204: Based on the range detection results, match the transaction data of online payment information and income and expenditure details within the target number of transactions to obtain the target reconciliation result.

[0049] Optionally, after obtaining the range detection results, at least one target number of transactions range can be determined based on the range detection results, and transaction data within the target number of transactions range can be matched one by one.

[0050] For example, for the target number of transactions in the range of 1001-2000, matching can be performed transaction by transaction. When the transaction data with the number of transactions is matched to 1001, the transaction data J with channel number 1001 in the online payment information is obtained, and the transaction data K with transaction serial number 1001 in the income and expenditure details information is obtained. The data content in transaction data J and transaction data K is matched precisely. If the result of the precise string matching is that the data content in transaction data J and transaction data K are completely consistent, then the matching of the next number of transactions is started, until the matching of the 2000 transactions is completed. The number of transactions with inconsistent data content during the matching process is recorded to obtain at least one target number of transactions. All target numbers of transactions are determined as the target reconciliation result.

[0051] In summary, by checking whether the number of transactions in online payment information and the number of transactions in income and expenditure details are equal, the number of transactions is counted, thus determining the result of the first level of detection. When the number of transactions is equal, a batch reconciliation detection model is used to determine the range detection result corresponding to the online payment information and income and expenditure details, and the matching range is narrowed down to the target number range within the range detection result, thus determining the result of the second level of detection. Based on the range detection result, the transaction data within the target number range is matched to obtain the target reconciliation result, thus determining the result of the third level of detection. Through these three levels of detection, initial screening is performed based on the number of transactions, and the batch reconciliation detection model quickly locates the target number range. Matching is then performed based on the target number range, avoiding the time-consuming process of matching all transaction data sequentially, thereby improving reconciliation efficiency.

[0052] To improve reconciliation efficiency, optionally, the batch reconciliation detection model can be used to determine the range detection results corresponding to online payment information and transaction details. This includes: splitting the online payment information and transaction details into multiple transaction number intervals, each including a preset number of transaction data; extracting features from each interval to obtain multiple interval features, where each interval feature corresponds one-to-one with the number of transactions; inputting each interval feature into the batch reconciliation detection model, which outputs the interval detection results corresponding to the interval features, where the interval detection results indicate either consistent or inconsistent transaction data; determining the target transaction number interval based on the transaction number interval to which the interval feature with inconsistent transaction data belongs, and then determining the range detection results based on the target transaction number interval.

[0053] Optionally, online payment information and transaction details can be split according to a rule that a predetermined number of transaction data points are included within a certain number of transaction ranges. Feature extraction can be performed by extracting statistical features from multiple data points of each transaction data point in the split payment information (or transaction details) results, thus obtaining the interval features corresponding to the number of transaction ranges. The interval features are then input into the batch reconciliation detection model to obtain the interval detection results for that number of transaction ranges.

[0054] For example, taking online payment information as an example, if the online payment information includes 5,000 transaction data entries from channel number 1 to channel number 5000, with a preset quantity of 1,000, then the above online payment information can be split into five payment information splitting results: transaction data from channel number 1 to channel number 1000, transaction data from channel number 1001 to channel number 2000, transaction data from channel number 2001 to channel number 3000, transaction data from channel number 3001 to channel number 4000, and transaction data from channel number 4001 to channel number 5000. Similarly, five detailed information splitting results can be obtained. For the transaction data from channel number 1 to channel number 1000, statistical features such as the mean, maximum, minimum, and variance of these 1,000 transaction data entries can be extracted to obtain interval features.

[0055] In summary, by separately splitting online payment information and transaction details, we obtained payment information splitting results and transaction details splitting results for multiple transaction number ranges. Then, through feature extraction, we obtained the interval features corresponding to the transaction number ranges. By inputting the interval features into the batch reconciliation detection model, we obtained the interval detection results for the transaction number ranges, which enabled us to quickly locate the transaction number ranges with inconsistent transaction data and improved reconciliation efficiency.

[0056] To improve reconciliation efficiency, optionally, the transaction data includes transaction amount and transaction account. Feature extraction is performed on the payment information splitting results and detailed information splitting results for each transaction number interval to obtain multiple interval features, including: for any transaction number interval, determining a first mean amount feature and a first variance amount feature based on the transaction amount of multiple transaction data in the payment information splitting results; determining a second mean amount feature and a second variance amount feature based on the transaction amount of multiple transaction data in the detailed information splitting results; determining a first account number percentage feature and a first account number total value feature based on the transaction account of multiple transaction data in the payment information splitting results; determining a second account number percentage feature and a second account number total value feature based on the transaction account of multiple transaction data in the detailed information splitting results; determining interval features corresponding to the transaction number interval based on the first mean amount feature, the first variance amount feature, the second mean amount feature, the second variance amount feature, the first account number percentage feature, the first account number total value feature, the second account number percentage feature, and the second account number total value feature; and determining multiple interval features based on the interval features corresponding to multiple transaction number intervals.

[0057] Optionally, the data content in the transaction data can be the transaction amount and the transaction account, and the transaction account can include the receiving account and the paying account. Taking a preset number of 3 transactions as an example, suppose there is a list of transaction data [100, a1, b1], [200, aa1, bb1], [300, a1a, b1b], where the data with index 0 in the list is the transaction amount, the data with index 1 is the paying account, and the data with index 2 is the receiving account. Assuming the above transaction data is the result of payment information splitting, then the first amount mean feature 200 is obtained by calculating (100+200+300 / 3), and the first amount variance feature 6666 is obtained by calculating the variance of the number sequence (100, 200, 300). The first (or second) account number proportion feature is used to represent all transaction data. The average proportion of numbers in the trading accounts is calculated as follows: [100, a1, b1] has a 1 / 2 proportion of numbers (meaning only 2 / 4 of a1 and b1 are numbers), [200, aa1, bb1] has a 1 / 3 proportion of numbers, and [300, a1a, b1b] has a 1 / 3 proportion of numbers. Therefore, calculating (1 / 2 + 1 / 3 + 1 / 3) divided by the preset quantity 3 yields the characteristic of the number proportion of the first account as 7 / 18. The characteristic of the total number value of the first (or second) account is the sum of the individual numbers in all trading accounts. The sum of a1, b1, aa1, bb1, a1a, and b1b is 6, therefore the characteristic of the total number value of the first account is 6. The calculation methods for the second average amount feature, the second variance amount feature, the second account number percentage feature, and the second account number total value feature are the same as those for the first average amount feature, the first variance amount feature, the first account number percentage feature, and the first account number total value feature. That is, the difference between "second" and "first" in all the above features is only the calculation object they target. Features with "first" in their names target the payment information breakdown results, while features with "second" in their names target the detailed information breakdown results. Assume the second average amount feature is... The above feature extraction process involves only simple numerical calculations, and its computation time is far less than the time required to match transaction data.

[0058] In summary, by determining the first mean value feature, the first variance value feature, the second mean value feature, the second variance value feature, the first account number percentage feature, the first account number total value feature, the second account number percentage feature, and the second account number total value feature, interval features were obtained, laying the data foundation for the batch reconciliation detection model and improving reconciliation efficiency.

[0059] To improve reconciliation efficiency, the training steps of the batch reconciliation detection model optionally include: obtaining a batch reconciliation training dataset, wherein the batch reconciliation training dataset includes multiple batch samples and labels for each batch sample, the batch samples include online payment samples and income and expenditure detail samples, and each online payment sample and income and expenditure detail sample includes a preset number of transaction data samples; extracting features from each batch sample to obtain multiple sample features, wherein the sample features correspond one-to-one with the batch samples; and training a preset machine learning model using the multiple sample features and the labels corresponding to the sample features to obtain the batch reconciliation detection model.

[0060] For example, multiple batch samples can be collected and labeled to obtain corresponding labels. The feature extraction method for batch samples is the same as the feature extraction method for obtaining interval features. Sample features are used in the training stage of the batch reconciliation detection model, and interval features are used in the inference stage of the batch reconciliation detection model. The preset machine learning model can be a logistic regression model to improve the inference speed.

[0061] In summary, by training a pre-defined machine learning model using a batch reconciliation training dataset, a batch reconciliation detection model was obtained, thus improving reconciliation efficiency.

[0062] To improve reconciliation efficiency, optionally, based on the range detection results, the online payment information and transaction details within the target number of transactions can be matched to obtain the target reconciliation result, including: determining multiple target payment transaction data in key-value pair format based on the online payment information within the target number of transactions, wherein the key-value pair includes a key name and a key value; determining multiple target detail transaction data in key-value pair format based on the transaction details within the target number of transactions; and matching the key values ​​of target payment transaction data and target detail transaction data with the same key name to obtain the target reconciliation result.

[0063] For example, assuming the target number of transactions is between 1001 and 2000, we would retrieve 1000 transaction records from channel number 1001 to channel number 2000 in the online payment information, using the channel number as the key and the data content (transaction amount, payment account, receiving account, etc.) as the key-value pair to construct 1000 target payment transaction records; similarly, we would retrieve 1000 transaction records from transaction serial number 1001 to transaction serial number 2000 in the income and expenditure details information, using the transaction serial number as the key and the data content (transaction amount, payment account, receiving account, etc.) as the key-value pair to construct 1000 target detailed transaction records. Following the order of transactions within the target transaction range, match the key values ​​of target payment transactions and target detail transactions with the same key name, one by one. The matching algorithm can be the SimHash algorithm (a text matching algorithm). If the SimHash algorithm calculates that the key values ​​of target payment transactions and target detail transactions with the same key name are greater than a preset similarity threshold, then the target payment transaction data and target detail transaction data are consistent. Repeat the above process, recording the number of transactions with inconsistent data content during the matching process (i.e., the sequence number of the key name), obtaining at least one target transaction count, and determining all target transaction counts as the target reconciliation result.

[0064] In summary, by determining the target payment transaction data and target detailed transaction data in key-value pair format, and then matching the key values ​​of target payment transaction data and target detailed transaction data with the same key name, the target reconciliation result is obtained, thus improving reconciliation efficiency.

[0065] To improve reconciliation accuracy, optionally, after checking whether the number of transactions in the online payment information and the number of transactions in the income and expenditure details information are equal, the method further includes: if the number of transactions is not equal, deleting transaction data after a first preset time in the online payment information to obtain corrected online payment information; deleting transaction data before a second preset time in the income and expenditure details information to obtain corrected income and expenditure details information; and correcting the number of transactions to be equal if the number of transactions in the corrected online payment information and the number of transactions in the corrected income and expenditure details information are equal.

[0066] For example, the transaction count result might show 3000 transactions in the online payment information and 2998 transactions in the income and expenditure details information. Since there's a delay between the generation of income and expenditure details and online payment information, it's possible that online payment information was generated at 11:58 PM, while income and expenditure details were generated at 12:01 AM the following day. A first preset time could be 11:55 PM, and a second preset time could be 12:05 AM. Transaction data after the first preset time in the online payment information is deleted to obtain corrected online payment information. Similarly, transaction data before the second preset time in the income and expenditure details is deleted to obtain corrected income and expenditure details. If the number of transactions in the corrected online payment information and the corrected income and expenditure details are equal, the transaction count result is corrected to be equal. If the number of transactions in the corrected online payment information and the corrected income and expenditure details are still not equal, the transaction count result remains unchanged.

[0067] In summary, by deleting transaction data after the first preset time in the online payment information to obtain corrected online payment information, and by deleting transaction data before the second preset time in the income and expenditure details information to obtain corrected income and expenditure details information, the error detection results are corrected, misjudgment caused by information generation delays is avoided, and the accuracy of reconciliation is improved.

[0068] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0069] Example 2

[0070] This application also provides a reconciliation device for financial institutions. It should be noted that this reconciliation device can be used to execute the reconciliation method for financial institutions provided in this application. The following describes the reconciliation device for financial institutions provided in this application.

[0071] According to embodiments of this application, an apparatus for implementing the above-mentioned reconciliation method of financial institutions is also provided, such as... Figure 3 As shown, the device includes:

[0072] The information acquisition unit 301 is used to acquire online payment information and income and expenditure details of the target account type when the target user authorizes it. The online payment information and income and expenditure details include multiple transaction data.

[0073] The transaction count detection unit 302 is used to detect whether the number of transaction data in the online payment information is equal to the number of transaction data in the income and expenditure details information, and obtain the transaction count detection result.

[0074] The interval determination unit 303 is used to determine the range detection result corresponding to the online payment information and income and expenditure details information by using the batch reconciliation detection model when the number of transactions is equal. The range detection result includes the target number range of transaction data where there is inconsistency between the online payment information and income and expenditure details information.

[0075] The reconciliation matching unit 304 is used to match online payment information and income and expenditure details within the target number of transactions based on the range detection results, so as to obtain the target reconciliation result.

[0076] The reconciliation device for financial institutions provided in this application embodiment, through an information acquisition unit 301, acquires online payment information and transaction details of the target account type with the authorization of the target user. Both the online payment information and the transaction details include multiple transaction data. A transaction count detection unit 302 detects whether the number of transactions in the online payment information and the number of transactions in the transaction details are equal, obtaining a transaction count detection result. If the transaction count detection result indicates that the number of transactions is equal, an interval determination unit 303 uses a batch reconciliation detection model to determine a range detection result corresponding to the online payment information and transaction details. The range detection result includes a target transaction count range where there are inconsistencies between the online payment information and the transaction details. A reconciliation matching unit 304 matches the transaction data in the online payment information and transaction details within the target transaction count range based on the range detection result, obtaining the target reconciliation result. This solves the problem of low reconciliation efficiency for financial institutions in related technologies, thereby improving the reconciliation efficiency of financial institutions.

[0077] Optionally, in the reconciliation device for financial institutions provided in this application embodiment, the interval determination unit 303 includes: an information splitting module, used to split online payment information and income and expenditure details information respectively to obtain payment information splitting results and details information splitting results for multiple transaction number intervals, wherein each transaction number interval's payment information splitting result and details information splitting result includes a preset number of transaction data; a feature extraction module, used to extract features from each transaction number interval's payment information splitting result and details information splitting result to obtain multiple interval features, wherein the interval features and transaction number have a one-to-one correspondence; a model prediction module, used to input each interval feature into a batch reconciliation detection model, and use the batch reconciliation detection model to output the interval detection result corresponding to the interval feature, wherein the interval detection result is that the interval transaction data is consistent or inconsistent; and a result determination module, used to determine the target transaction number interval based on the transaction number interval to which the interval feature with the interval detection result is inconsistent, and to determine the range detection result based on the target transaction number interval.

[0078] Optionally, in the reconciliation device for financial institutions provided in this application embodiment, the transaction data includes transaction amount and transaction account. The feature extraction module includes: an amount feature submodule, used to determine a first amount mean feature and a first amount variance feature based on the transaction amount of multiple transaction data in the payment information splitting result and the detailed information splitting result for any number of transactions; a second amount mean feature and a second amount variance feature based on the transaction amount of multiple transaction data in the detailed information splitting result; an account feature submodule, used to determine a first account number percentage feature and a first account number total value feature based on the transaction account of multiple transaction data in the payment information splitting result; a second account number percentage feature and a second account number total value feature based on the transaction account of multiple transaction data in the detailed information splitting result; and an interval feature determination submodule, used to determine the interval features corresponding to the number of transactions interval based on the first amount mean feature, the first amount variance feature, the second amount mean feature, the second amount variance feature, the first account number percentage feature, the first account number total value feature, the second account number percentage feature, and the second account number total value feature; and to determine multiple interval features based on the interval features corresponding to multiple number of transactions intervals.

[0079] Optionally, in the reconciliation device for financial institutions provided in this application embodiment, the device further includes: a dataset acquisition unit, used to acquire a batch reconciliation training dataset, wherein the batch reconciliation training dataset includes multiple batch samples and labels for each batch sample, the batch samples include online payment samples and income and expenditure detail samples, and each online payment sample and income and expenditure detail sample includes a preset number of transaction data samples; a sample feature extraction unit, used to extract features from each batch sample to obtain multiple sample features, wherein the sample features and batch samples have a one-to-one correspondence; and a model training unit, used to train a preset machine learning model using the multiple sample features and labels corresponding to the sample features to obtain a batch reconciliation detection model.

[0080] Optionally, in the reconciliation device for financial institutions provided in this application embodiment, the reconciliation matching unit 304 includes: a first determining module, used to determine multiple target payment transaction data in key-value pair format based on transaction data in the target number range of online payment information, wherein the key-value pair includes a key name and a key value; a second determining module, used to determine multiple target detail transaction data in key-value pair format based on transaction data in the target number range of income and expenditure detail information; and a key-value matching module, used to match the key values ​​of target payment transaction data and target detail transaction data that belong to the same key name to obtain the target reconciliation result.

[0081] Optionally, in the reconciliation device for financial institutions provided in this application embodiment, the device further includes: a first deletion unit, used to detect whether the number of transaction data in the online payment information and the number of transaction data in the income and expenditure details information are equal, and after obtaining the transaction count detection result, if the transaction count detection result is not equal, delete the transaction data in the online payment information that is after a first preset time to obtain the corrected online payment information; a second deletion unit, used to delete the transaction data in the income and expenditure details information that is before a second preset time to obtain the corrected income and expenditure details information; and a result correction unit, used to correct the transaction count detection result to be equal if the number of transaction data in the corrected online payment information and the number of transaction data in the corrected income and expenditure details information are equal.

[0082] It should be noted that the information acquisition unit 301, the transaction count detection unit 302, the interval determination unit 303, and the reconciliation matching unit 304 mentioned above correspond to steps S201 to S204 in Embodiment 1. The instances and application scenarios implemented by the units and corresponding steps are the same, but are not limited to the content disclosed in Embodiment 1. It should be noted that the above modules or units can be hardware or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above modules can also be part of the device and run in the computer terminal 10 provided in Embodiment 1.

[0083] Example 3

[0084] Embodiments of this application may provide an electronic device. Figure 4 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 4 As shown, the electronic device may include: one or more ( Figure 4 (Only one is shown) Processor 1002, memory 1004, memory controller, and peripheral interface, wherein the peripheral interface is connected to the radio frequency module, audio module and display.

[0085] The memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the methods and apparatus in the embodiments of this application. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby implementing the above-described methods. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include memory remotely located relative to the processor, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0086] The processor can access information and applications stored in memory via a transmission device to perform the following steps: With authorization from the target user, obtain online payment information and transaction details for the target account type, where both online payment information and transaction details include multiple transaction records; check if the number of transactions in the online payment information and the number of transactions in the transaction details are equal, obtaining a transaction count detection result; if the transaction count detection result indicates equality, use a batch reconciliation detection model to determine a range detection result corresponding to the online payment information and transaction details, where the range detection result includes a target transaction count range where there are inconsistencies between the online payment information and the transaction details; match the transaction data in the online payment information and transaction details within the target transaction count range based on the range detection result, obtaining the target reconciliation result.

[0087] The processor can also access information and applications stored in the memory via a transmission device to perform the following steps: First, split the online payment information and transaction details to obtain multiple transaction number intervals for each payment information interval and transaction details interval. Each transaction number interval includes a preset number of transaction data. Second, extract features from each transaction number interval to obtain multiple interval features, where each interval feature corresponds one-to-one with the number of transactions. Third, input each interval feature into a batch reconciliation detection model and output the interval detection result corresponding to the interval feature. The interval detection result indicates whether the transaction data is consistent or inconsistent. Fourth, determine the target transaction number interval based on the transaction number interval to which the interval feature with inconsistent transaction data belongs, and determine the range detection result based on the target transaction number interval.

[0088] The processor can also invoke information and applications stored in the memory via a transmission device to perform the following steps: For any interval of transaction counts, based on the transaction amounts of multiple transaction data in the payment information splitting results and detailed information splitting results, determine a first mean value feature and a first variance value feature; based on the transaction amounts of multiple transaction data in the detailed information splitting results, determine a second mean value feature and a second variance value feature; based on the transaction accounts of multiple transaction data in the payment information splitting results, determine a first account number percentage feature and a first account number total value feature; based on the transaction accounts of multiple transaction data in the detailed information splitting results, determine a second account number percentage feature and a second account number total value feature; based on the first mean value feature, the first variance value feature, the second mean value feature, the second variance value feature, the first account number percentage feature, the first account number total value feature, the second account number percentage feature, and the second account number total value feature, determine the interval features corresponding to the interval of transaction counts; based on the interval features corresponding to multiple interval of transaction counts, determine multiple interval features.

[0089] The processor can also access information and applications stored in the memory via a transmission device to perform the following steps: acquiring a batch reconciliation training dataset, wherein the batch reconciliation training dataset includes multiple batch samples and labels for each batch sample, the batch samples include online payment samples and income and expenditure detail samples, and each online payment sample and income and expenditure detail sample includes a preset number of transaction data samples; extracting features from each batch sample to obtain multiple sample features, wherein the sample features correspond one-to-one with the batch samples; training a preset machine learning model using the multiple sample features and the labels corresponding to the sample features to obtain a batch reconciliation detection model.

[0090] The processor can also invoke information and applications stored in the memory via the transmission device to perform the following steps: based on the transaction data of online payment information within the target number of transactions, determine multiple target payment transaction data in key-value pair format, wherein the key-value pair includes a key name and a key value; based on the transaction data of income and expenditure details within the target number of transactions, determine multiple target detail transaction data in key-value pair format; match the key values ​​of target payment transaction data and target detail transaction data with the same key name to obtain the target reconciliation result.

[0091] The processor can also call the information and application stored in the memory through the transmission device to perform the following steps: if the number of transactions is not equal, delete the transaction data after the first preset time in the online payment information to obtain the corrected online payment information; delete the transaction data before the second preset time in the income and expenditure details information to obtain the corrected income and expenditure details information; if the number of transactions in the corrected online payment information and the number of transactions in the corrected income and expenditure details information are equal, correct the number of transactions to be equal.

[0092] This application provides a reconciliation solution for financial institutions. With the authorization of the target user, online payment information and transaction details for the target account type are obtained. Both the online payment information and transaction details include multiple transaction records. The number of transactions in the online payment information and the transaction details is checked to determine if they are equal, resulting in a transaction count detection result. If the transaction counts are equal, a batch reconciliation detection model is used to determine a range detection result corresponding to the online payment information and transaction details. This range detection result includes a target number range of transactions where inconsistencies exist between the online payment information and the transaction details. Based on the range detection result, the transaction data in the online payment information and transaction details within the target number range are matched to obtain the target reconciliation result. This solution addresses the technical problem of low reconciliation efficiency for financial institutions in existing technologies.

[0093] Those skilled in the art will understand that Figure 4 The structure shown is for illustrative purposes only. Electronic devices can also be smartphones, tablets, handheld computers, mobile internet devices (MIDs), PADs, and other terminal devices. Figure 4 This does not limit the structure of the aforementioned electronic device. For example, electronic devices may also include components that are more... Figure 4 The more or fewer components shown (such as network interfaces, display devices, etc.), or having the same Figure 4 The different configurations shown.

[0094] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing the hardware related to the terminal device. The program can be stored in a computer-readable storage medium, which may include: flash drive, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0095] Example 4

[0096] Embodiments of this application also provide a storage medium. Optionally, in this embodiment, the storage medium can be used to store the program code executed by the reconciliation method for financial institutions provided in Embodiment 1.

[0097] Optionally, in this embodiment, the storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any mobile terminal in a group of mobile terminals.

[0098] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: Upon obtaining authorization from the target user, acquiring online payment information and transaction details of the target account type, wherein both the online payment information and transaction details include multiple transaction data; detecting whether the number of transaction data in the online payment information and the number of transaction data in the transaction details are equal, obtaining a transaction count detection result; if the transaction count detection result indicates that the number of transactions is equal, using a batch reconciliation detection model to determine a range detection result corresponding to the online payment information and transaction details, wherein the range detection result includes a target number range of transaction data where there is inconsistency between the online payment information and the transaction details; matching the transaction data in the online payment information and transaction details within the target number range based on the range detection result, obtaining a target reconciliation result.

[0099] Optionally, in this embodiment, the computer-readable storage medium is further configured to store program code for performing the following steps: splitting online payment information and transaction details information to obtain payment information splitting results and transaction details splitting results for multiple transaction number intervals, wherein each transaction number interval's payment information splitting result and transaction details splitting result includes a preset number of transaction data; extracting features from each transaction number interval's payment information splitting result and transaction details splitting result to obtain multiple interval features, wherein the interval features and transaction numbers have a one-to-one correspondence; inputting each interval feature into a batch reconciliation detection model, and using the batch reconciliation detection model to output interval detection results corresponding to the interval features, wherein the interval detection results indicate that the interval transaction data is consistent or inconsistent; determining a target transaction number interval based on the transaction number interval to which the interval feature with inconsistent transaction data belongs, and determining a range detection result based on the target transaction number interval.

[0100] Optionally, in this embodiment, the computer-readable storage medium is further configured to store program code for performing the following steps: for any given interval of payment information splitting results and detailed information splitting results, determining a first mean amount feature and a first variance amount feature based on the transaction amounts of multiple transaction data in the payment information splitting results; determining a second mean amount feature and a second variance amount feature based on the transaction amounts of multiple transaction data in the detailed information splitting results; determining a first account number percentage feature and a first account number total value feature based on the transaction accounts of multiple transaction data in the payment information splitting results; determining a second account number percentage feature and a second account number total value feature based on the transaction accounts of multiple transaction data in the detailed information splitting results; determining interval features corresponding to the interval of payment number intervals based on the first mean amount feature, the first variance amount feature, the second mean amount feature, the second variance amount feature, the first account number percentage feature, the first account number total value feature, the second account number percentage feature, and the second account number total value feature; and determining multiple interval features based on the interval features corresponding to multiple intervals of payment number intervals.

[0101] Optionally, in this embodiment, the computer-readable storage medium is further configured to store program code for performing the following steps: obtaining a batch reconciliation training dataset, wherein the batch reconciliation training dataset includes multiple batch samples and labels for each batch sample, the batch samples include online payment samples and income and expenditure detail samples, and each online payment sample and income and expenditure detail sample includes a preset number of transaction data samples; extracting features from each batch sample to obtain multiple sample features, wherein the sample features and batch samples have a one-to-one correspondence; training a preset machine learning model using the multiple sample features and the labels corresponding to the sample features to obtain a batch reconciliation detection model.

[0102] Optionally, in this embodiment, the computer-readable storage medium is further configured to store program code for performing the following steps: determining multiple target payment transaction data in key-value pair format based on transaction data of online payment information within a target number of transactions, wherein the key-value pair includes a key name and a key value; determining multiple target detail transaction data in key-value pair format based on transaction data of income and expenditure details within a target number of transactions; and matching the key values ​​of target payment transaction data and target detail transaction data that belong to the same key name to obtain a target reconciliation result.

[0103] Optionally, in this embodiment, the computer-readable storage medium is further configured to store program code for performing the following steps: if the number of transactions detected is unequal, delete transaction data in the online payment information that occurs after a first preset time to obtain corrected online payment information; delete transaction data in the income and expenditure details information that occurs before a second preset time to obtain corrected income and expenditure details information; if the number of transactions in the corrected online payment information and the number of transactions in the corrected income and expenditure details information are equal, correct the number of transactions detected to be equal.

[0104] This application also provides a computer program product that, when executed on a data processing device, is suitable for performing the reconciliation method steps of a financial institution.

[0105] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0106] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0107] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0108] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0109] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0110] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0111] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A reconciliation method for financial institutions, characterized in that, include: With the authorization of the target user, obtain online payment information and transaction details of the target account type, wherein the online payment information and transaction details include multiple transaction data. The number of transactions in the online payment information is checked against the number of transactions in the income and expenditure details information to obtain the transaction count result. If the number of transactions is equal, a batch reconciliation detection model is used to determine the range detection result corresponding to the online payment information and the income and expenditure details information. The range detection result includes the target number range of transaction data where there is inconsistency between the online payment information and the income and expenditure details information. Based on the range detection results, the online payment information and the income and expenditure details are matched with the transaction data within the target number of transactions range to obtain the target reconciliation result.

2. The method according to claim 1, characterized in that, The batch reconciliation detection model is used to determine the range detection results corresponding to the online payment information and the income and expenditure details information, including: The online payment information and the income and expenditure details are split separately to obtain payment information splitting results and details splitting results for multiple transaction number intervals. Each payment information splitting result and details splitting result for each transaction number interval includes a preset number of transaction data. Feature extraction is performed on the payment information splitting result and the detailed information splitting result for each of the number of transactions intervals to obtain multiple interval features, wherein the interval features and the number of transactions have a one-to-one correspondence; Each of the interval features is input into the batch reconciliation detection model, and the batch reconciliation detection model is used to output the interval detection result corresponding to the interval feature, wherein the interval detection result is either the interval transaction data is consistent or the interval transaction data is inconsistent; The target number of transactions interval is determined based on the number of transactions interval to which the interval features of inconsistent transaction data in the interval detection results belong, and the range detection results are determined based on the target number of transactions interval.

3. The method according to claim 2, characterized in that, The transaction data includes transaction amount and transaction account. Feature extraction is performed on the payment information breakdown results and the detailed information breakdown results for each transaction range to obtain multiple range features, including: For any interval of transaction count, the payment information splitting result and the detailed information splitting result are used to determine a first amount mean feature and a first amount variance feature based on the transaction amount of multiple transaction data in the payment information splitting result, and a second amount mean feature and a second amount variance feature are used to determine the transaction amount of multiple transaction data in the detailed information splitting result. Based on the transaction accounts of multiple transaction data in the payment information splitting result, the first account's numerical proportion feature and the first account's numerical total value feature are determined; based on the transaction accounts of multiple transaction data in the detailed information splitting result, the second account's numerical proportion feature and the second account's numerical total value feature are determined. The interval features corresponding to the number of transactions are determined based on the first mean value feature, the first variance feature, the second mean value feature, the second variance feature, the first account number percentage feature, the first account number total value feature, the second account number percentage feature, and the second account number total value feature. The multiple interval features are determined based on the interval features corresponding to the multiple intervals of the number of transactions.

4. The method according to claim 2, characterized in that, The training steps of the batch reconciliation detection model include: Obtain a batch reconciliation training dataset, wherein the batch reconciliation training dataset includes multiple batch samples and a label for each batch sample, the batch samples include online payment samples and income and expenditure detail samples, and each online payment sample and income and expenditure detail sample includes a preset number of transaction data samples; Feature extraction is performed on each batch of samples to obtain multiple sample features, wherein the sample features and the batch of samples have a one-to-one correspondence. The preset machine learning model is trained using the multiple sample features and the labels corresponding to the sample features to obtain the batch reconciliation detection model.

5. The method according to claim 1, characterized in that, Based on the range detection results, the online payment information and the transaction details are matched with the transaction data within the target number of transactions range to obtain the target reconciliation results, including: Based on the online payment information and the transaction data within the target number of transactions range, determine multiple target payment transaction data in key-value pair format, wherein the key-value pair includes a key name and a key value; Based on the transaction data in the target number range according to the income and expenditure details, determine multiple target detailed transaction data in key-value pair format; The target payment transaction data and the target detailed transaction data with the same key name are matched to obtain the target reconciliation result.

6. The method according to claim 1, characterized in that, After obtaining the transaction count result by detecting whether the number of transactions in the online payment information and the number of transactions in the income and expenditure details information are equal, the method further includes: If the number of transactions is not equal, the transaction data that is after a first preset time in the online payment information will be deleted to obtain the corrected online payment information. The transaction data that occurred before the second preset time in the income and expenditure details are deleted to obtain the corrected income and expenditure details. If the number of transactions in the corrected online payment information is equal to the number of transactions in the corrected income and expenditure details information, the transaction count detection result will be corrected to show that the number of transactions is equal.

7. A reconciliation device for a financial institution, characterized in that, include: The information acquisition unit is used to acquire online payment information and income and expenditure details of the target account type when authorized by the target user, wherein the online payment information and the income and expenditure details include multiple transaction data. The transaction count detection unit is used to detect whether the number of transaction data in the online payment information is equal to the number of transaction data in the income and expenditure details information, and obtain the transaction count detection result; An interval determination unit is used to determine, when the number of transactions is equal, a range detection result corresponding to the online payment information and the income and expenditure details information using a batch reconciliation detection model. The range detection result includes the target number range of transaction data where there is inconsistency between the online payment information and the income and expenditure details information. The reconciliation matching unit is used to match the online payment information and the income and expenditure details within the target number of transactions based on the range detection results, so as to obtain the target reconciliation result.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the computer-readable storage medium is located to perform the reconciliation method of a financial institution as described in any one of claims 1 to 6.

9. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, executes the reconciliation method of a financial institution as described in any one of claims 1 to 6.

10. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the steps of the reconciliation method for financial institutions as described in any one of claims 1 to 6.

Images