An AI-based automatic reconciliation optimization and processing method and system

Abstract

The present invention discloses an automatic reconciliation optimization and processing method and system based on AI, which relates to the field of data processing technology. First, the enterprise financial data is automatically collected and analyzed through the ERP system to obtain the matching evaluation value of the enterprise financial data, and the data is automatically adjusted. At the same time, the reconciliation rule data of the enterprise financial data is monitored to obtain the reconciliation rule compliance value of the enterprise financial data, and the reconciliation rules of the enterprise financial data are optimized. Reconciliation is performed according to the optimized reconciliation rules to obtain the reconciliation data of the enterprise financial data, and the reconciliation abnormal values ​​of the enterprise financial data are analyzed to identify the abnormalities, generate reports through AI, and give reminders, which is helpful to improve the efficiency and accuracy of financial data processing and reduce manual errors and omissions, and help improve the adaptability and effectiveness of reconciliation rules.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an AI-based automatic reconciliation optimization and processing method and system. Background Art

[0002] With the increasing complexity and expansion of corporate business, traditional manual reconciliation methods are time-consuming, labor-intensive and error-prone, and can no longer meet the financial management needs of modern enterprises. Artificial intelligence technology, especially advances in automated processing, intelligent analysis and decision support, has made automatic reconciliation possible. Through AI technology, the system can automatically identify and process financial data, perform intelligent comparison and analysis, thereby realizing the automation and intelligence of the reconciliation process and significantly improving reconciliation efficiency and accuracy.

[0003] For example, the invention patent with announcement number CN117421254B announces an automated testing method and system for reconciliation business, which belongs to the field of data processing technology, and specifically includes: building a visualization page, determining the reconciliation test business requirements of a specified account in response to user operation data, automatically generating an automated testing script based on the reconciliation test business requirements and abnormal server nodes, and acquiring the reconciliation business data of different server nodes according to a preset order and abnormal server nodes, thereby determining the reconciliation files of different server nodes and the reconciliation results between server nodes, acquiring abnormal reconciliation results of different server nodes under different historical test times, and optimizing the preset order in combination with the account change records under the historical test times corresponding to the abnormal reconciliation results and the reconciliation processing priority, thereby improving the processing efficiency of the reconciliation business.

[0004] For example, the invention patent with announcement number CN117725451B announces a method and system for automatic reconciliation of multi-dimensional transaction data, which can be applied to emerging software and new information technology services, as well as big data analysis and mining services. It obtains two multi-dimensional target transaction data to be reconciled, performs initial clustering according to the initial k value to obtain multiple initial clustering clusters, and then determines the data distribution characteristic index according to the shape regularity of each initial clustering cluster, the overall density of the multi-dimensional target transaction data, and the difference in data point distances. The initial k value is corrected, and then the corresponding multi-dimensional target transaction data is clustered again according to the corrected k value. Then, the multi-source difference characteristic index is obtained based on the difference in the clustering again. Finally, the target k value is obtained based on the multi-source difference characteristic index.

[0005] Based on the above findings, existing technical solutions may have problems such as inaccurate corporate financial data, unreasonable reconciliation rules, and high error rates in manual reconciliation, which may lead to unreasonable corporate financial data, causing the company to make wrong financial decisions and fail to discover and correct financial errors in a timely manner. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides an AI-based automatic reconciliation optimization and processing method and system, which solves the problems designed in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: an AI-based automatic reconciliation optimization and processing method and system, including automatically collecting enterprise financial data from the enterprise's ERP system, analyzing the enterprise financial data, obtaining a matching evaluation value of the enterprise financial data, and automatically adjusting the enterprise financial data according to the matching evaluation value of the enterprise financial data.

[0008] Monitor the reconciliation process of the enterprise's financial data, obtain the reconciliation rule data of the enterprise's financial data, analyze the reconciliation rule data of the enterprise's financial data, obtain the reconciliation rule compliance value of the enterprise's financial data, and optimize the reconciliation rules of the enterprise's financial data based on the reconciliation rule compliance value of the enterprise's financial data.

[0009] According to the optimized reconciliation rules of enterprise financial data, the enterprise financial data is reconciled through AI technology to obtain the reconciliation data of the enterprise financial data, and the reconciliation data of the enterprise financial data is analyzed to obtain the reconciliation anomalies of the enterprise financial data. According to the reconciliation anomalies of the enterprise financial data, a report is generated through AI to prompt relevant personnel to supplement.

[0010] Furthermore, the matching evaluation value of the enterprise financial data is specifically analyzed as follows:

[0011] The enterprise financial data is collected in the ERP system, and the enterprise financial data includes: the duration of each billing cycle, the time difference between the start of each bill and the financial settlement, the transaction date of each bill, the payment date of each bill, and the delay time of each bill.

[0012] The duration of each billing cycle, the time difference between the start of each bill and the financial settlement, the transaction date of each bill, the payment date of each bill, and the delay time of each bill are analyzed and processed to obtain a matching evaluation value of the enterprise's financial data, which is used to evaluate the compliance of the enterprise's financial data.

[0013] Furthermore, the enterprise financial data is automatically adjusted according to the matching evaluation value of the enterprise financial data, and the specific analysis is as follows:

[0014] Extracting the matching evaluation value of the enterprise financial data, matching it with the data conformity corresponding to each interval of the matching evaluation value of the enterprise financial data stored in the database, and obtaining the data conformity category of the enterprise financial data, wherein the data conformity of the enterprise financial data includes: a first conformity degree, a second conformity degree, and a third conformity degree, wherein the first conformity degree, the second conformity degree, and the third conformity degree are all used to describe the level of data conformity, and the levels of data conformity of the first conformity degree, the second conformity degree, and the third conformity degree decrease step by step;

[0015] If the data compliance of the enterprise financial data is the first compliance level or the second compliance level, there is no need to adjust the data update frequency of the enterprise financial data;

[0016] If the data compliance of the enterprise's financial data is the third compliance level, the data update frequency of the enterprise's financial data is adjusted.

[0017] Furthermore, the reconciliation rule data of the enterprise financial data is obtained, and the reconciliation rule data of the enterprise financial data is analyzed to obtain the reconciliation rule compliance value of the enterprise financial data. The specific analysis is:

[0018] The reconciliation rule data of the enterprise financial data includes: the length of the reconciliation cycle and the number of reconciliation operation interruptions;

[0019] The matching evaluation value of the enterprise financial data is extracted, and combined with the reconciliation rule data of the enterprise financial data to obtain the reconciliation rule compliance value of the enterprise financial data, and the reconciliation rule compliance value of the enterprise financial data is used to evaluate the compliance of the reconciliation rule of the enterprise financial data.

[0020] Furthermore, the reconciliation rules of the enterprise financial data are optimized according to the reconciliation rule compliance value of the enterprise financial data, and the specific analysis is as follows:

[0021] The reconciliation rules for the enterprise financial data include: the review frequency of the enterprise financial data and the amount of data processed by the reconciliation operation.

[0022] The reconciliation rule compliance value of the enterprise financial data is extracted, and matched with the review frequency of the enterprise accounting data corresponding to each interval of the reconciliation rule compliance value of the enterprise financial data stored in the database to obtain the review frequency of the enterprise accounting data.

[0023] The reconciliation rule compliance value of the enterprise financial data is extracted and compared with the reconciliation rule compliance value threshold of the enterprise financial data stored in the database. If the reconciliation rule compliance value of the enterprise financial data is higher than or equal to the reconciliation rule compliance value threshold of the enterprise financial data stored in the database, the reconciliation process of the enterprise financial data is continuously monitored.

[0024] If the reconciliation rule compliance value of the enterprise financial data is lower than the reconciliation rule compliance value threshold of the enterprise financial data stored in the database, the amount of data processed by the reconciliation operation is adjusted.

[0025] Furthermore, the reconciliation data of the enterprise financial data is obtained by reconciling the enterprise financial data through AI technology according to the optimized reconciliation rules of the enterprise financial data. The specific process is as follows:

[0026] The review frequency of the enterprise's accounting data and the data volume of the reconciliation operation processing are extracted to perform reconciliation processing on the enterprise's financial data in batches until all the enterprise's financial data have been reconciled, and the reconciliation data of each time of the enterprise's financial data are obtained, wherein the reconciliation data of each time of the enterprise's financial data include: the number of mismatches in each reconciliation data, the number of differences in the status of each reconciliation transaction, the number of inconsistencies in the time of each reconciliation transaction, and the number of abnormal transactions in each reconciliation.

[0027] Furthermore, the reconciliation data of the enterprise financial data is analyzed to obtain the reconciliation abnormal values ​​of the enterprise financial data, and the specific analysis is as follows:

[0028] The number of mismatches in reconciliation data, the number of differences in reconciliation transaction status, the number of inconsistencies in reconciliation transaction time, and the number of abnormal reconciliation transactions are extracted, and combined with the reconciliation rule compliance value of the enterprise's financial data, the reconciliation abnormal value of the enterprise's financial data is processed to obtain the reconciliation abnormal value of the enterprise's financial data. The reconciliation abnormal value of the enterprise's financial data is used to evaluate the financial security of the enterprise.

[0029] Furthermore, the financial security of the enterprise is evaluated and analyzed specifically as follows:

[0030] Extracting the reconciliation abnormal values ​​of the enterprise's financial data and matching them with the financial security conditions of the enterprise corresponding to each interval of the reconciliation abnormal values ​​of the enterprise's financial data stored in the database, to obtain the financial security conditions of the enterprise, the financial security conditions of the enterprise including: a first security level, a second security level and a third security level, the first security level, the second security level and the third security level are all used to describe the level of financial data security, and the first security level, the second security level and the third security level of financial data security increase step by step;

[0031] When the financial security situation of an enterprise is at the first security level or the second security level, the system will issue an early warning and adjust and correct erroneous record entries. When the financial security situation of an enterprise is at the third security level, the enterprise's financial data will be continuously monitored.

[0032] Furthermore, the reconciliation abnormal values ​​of the enterprise's financial data are used to generate a report through AI to prompt relevant personnel to supplement. The specific analysis is as follows:

[0033] Extract reconciliation anomalies from corporate financial data, analyze anomalies through AI and generate reports, send the reports to relevant persons in charge, and prompt them to make corrections, and use AI to track subsequent corrections.

[0034] The second aspect of this article provides an AI-based automatic reconciliation optimization and processing system, including:

[0035] The financial data adjustment module is used to automatically collect enterprise financial data from the enterprise's ERP system, analyze the enterprise financial data, obtain the matching evaluation value of the enterprise financial data, and automatically adjust the enterprise financial data based on the matching evaluation value of the enterprise financial data.

[0036] The reconciliation rule optimization module is used to monitor the reconciliation process of the enterprise's financial data, obtain the reconciliation rule data of the enterprise's financial data, analyze the reconciliation rule data of the enterprise's financial data, obtain the reconciliation rule compliance value of the enterprise's financial data, and optimize the reconciliation rules of the enterprise's financial data according to the reconciliation rule compliance value of the enterprise's financial data.

[0037] The abnormal data reporting module is used to reconcile the enterprise financial data through AI technology according to the optimized reconciliation rules of the enterprise financial data, obtain the reconciliation data of the enterprise financial data, and analyze the reconciliation data of the enterprise financial data to obtain the reconciliation abnormal values ​​of the enterprise financial data. According to the reconciliation abnormal values ​​of the enterprise financial data, a report is generated through AI to prompt relevant personnel to supplement.

[0038] The present invention has the following beneficial effects:

[0039] (1) The present invention automatically collects and analyzes enterprise financial data through the ERP system to obtain matching evaluation values ​​of the enterprise financial data, and automatically adjusts the data. At the same time, it monitors the reconciliation rule data of the enterprise financial data to obtain the reconciliation rule compliance value of the enterprise financial data, optimizes the reconciliation rules of the enterprise financial data, performs reconciliation according to the optimized reconciliation rules, obtains the reconciliation data of the enterprise financial data, and analyzes the reconciliation abnormal values ​​of the enterprise financial data, identifies the abnormalities, generates reports through AI, and issues reminders, which helps to improve the efficiency and accuracy of financial data processing and reduce manual errors and omissions, and helps to improve the adaptability and effectiveness of reconciliation rules.

[0040] (2) The present invention collects the duration of each billing cycle, the time difference between the start of each bill and the financial settlement, the transaction date of each bill, the payment date of each bill, and the delay time of each bill in the enterprise ERP system, obtains the matching evaluation value of the enterprise's financial data through analysis and processing, evaluates the compliance of the enterprise's financial data, and reviews and adjusts the enterprise's financial data based on the evaluation results, which helps to improve the accuracy and reliability of the financial data and provide strong support for the financial management of the enterprise.

[0041] (3) The present invention obtains and analyzes the reconciliation rules of the enterprise's financial data, combines the matching evaluation value of the enterprise's financial data, processes to obtain the reconciliation rule compliance value of the enterprise's financial data, and then obtains the review frequency according to the matching value of the reconciliation rules of the enterprise's financial data, and adjusts the amount of data processed by the reconciliation operation, which helps to improve the automation level of financial management.

[0042] (4) The present invention obtains the reconciliation data of each enterprise's financial data, and combines it with the reconciliation rule compliance value of the enterprise's financial data to obtain the reconciliation abnormal value of the enterprise's financial data, and evaluates the financial security of the enterprise. When the security level is low, the system will issue an early warning and adjust the error record; when it is highly secure, it will continue to monitor and generate a report through AI analysis, which will be sent to the person in charge to prompt corrections and track subsequent corrections, which will help improve the timeliness and accuracy of financial management.

[0043] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is a schematic diagram of the process of the present invention.

[0045] Figure 2 It is the curve of the number of interruptions of the reconciliation operation and the conformity value of the reconciliation rule of the enterprise financial data of the present invention.

[0046] Figure 3 It is a schematic diagram of the system module of the present invention. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 are within the scope of protection of the present invention.

[0048] In the description of the present invention, it should be understood that the terms "opening", "upper", "lower", "thickness", "top", "middle", "length", "inside", "all around" and the like indicating orientation or positional relationship are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the components or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0049] See also Figure 1 The embodiment of the present invention provides a technical solution: an AI-based automatic reconciliation optimization and processing method, including automatically collecting enterprise financial data from the enterprise's ERP system, analyzing the enterprise financial data, obtaining a matching evaluation value of the enterprise financial data, and automatically adjusting the enterprise financial data according to the matching evaluation value of the enterprise financial data.

[0050] Monitor the reconciliation process of the enterprise's financial data, obtain the reconciliation rule data of the enterprise's financial data, analyze the reconciliation rule data of the enterprise's financial data, obtain the reconciliation rule compliance value of the enterprise's financial data, and optimize the reconciliation rules of the enterprise's financial data based on the reconciliation rule compliance value of the enterprise's financial data.

[0051] According to the optimized reconciliation rules of enterprise financial data, the enterprise financial data is reconciled through AI technology to obtain the reconciliation data of the enterprise financial data, and the reconciliation data of the enterprise financial data is analyzed to obtain the reconciliation anomalies of the enterprise financial data. According to the reconciliation anomalies of the enterprise financial data, a report is generated through AI to prompt relevant personnel to supplement.

[0052] It should be noted that AI first automatically extracts and synchronizes corporate financial data by connecting to the company's ERP system. In this process, AI uses the API interface to automatically obtain the latest financial records from the ERP system.

[0053] It should be noted that the enterprise financial data is collected and analyzed through the enterprise ERP system to obtain the matching evaluation value of the enterprise financial data, and the enterprise financial data is adjusted. At the same time, the reconciliation rule data of the enterprise financial data is monitored, and the reconciliation rule compliance value of the enterprise financial data is obtained through processing. The reconciliation rules of the enterprise financial data are optimized, and reconciliation is performed according to the optimized reconciliation rules to obtain the reconciliation data of the enterprise financial data. The reconciliation anomaly values ​​of the enterprise financial data are analyzed, anomalies are identified, and reports are generated and reminders are issued through AI, which helps to improve the efficiency and accuracy of financial data processing.

[0054] It should be noted that reconciliation of corporate financial data through AI technology includes: inputting corporate financial data into the AI ​​reconciliation system, the system retrieves the corporate financial statements provided by the bank and, based on the reconciliation rules for the corporate financial data, reconciles the corporate financial data with the corporate financial statements provided by the bank item by item, identifies transaction records and marks differences, and generates a reconciliation result report after the reconciliation is completed. The reconciliation result report includes successfully matched transaction record entries, unmatched transaction records, and differences.

[0055] The matching evaluation value of the enterprise financial data is obtained above. The specific analysis process is as follows:

[0056] The enterprise financial data is collected in the ERP system, and the enterprise financial data includes: the duration of each billing cycle, the time difference between the start of each bill and the financial settlement, the transaction date of each bill, the payment date of each bill, and the delay time of each bill.

[0057] The duration of each billing cycle, the time difference between the start of each bill and the financial settlement, the transaction date of each bill, the payment date of each bill, and the delay time of each bill are analyzed and processed to obtain a matching evaluation value of the enterprise's financial data, which is used to evaluate the compliance of the enterprise's financial data.

[0058] The above automatically adjusts the enterprise financial data based on the matching evaluation value of the enterprise financial data. The specific analysis is as follows:

[0059] Extracting the matching evaluation value of the enterprise financial data, matching it with the data conformity corresponding to each interval of the matching evaluation value of the enterprise financial data stored in the database, and obtaining the data conformity category of the enterprise financial data, wherein the data conformity of the enterprise financial data includes: a first conformity degree (also called high conformity), a second conformity degree (also called basic conformity) and a third conformity degree (also called general conformity), wherein the first conformity degree, the second conformity degree and the third conformity degree are all used to describe the level of data conformity, and the levels of data conformity of the first conformity degree, the second conformity degree and the third conformity degree decrease step by step;

[0060] If the data compliance of the enterprise financial data is the first compliance level or the second compliance level, there is no need to adjust the data update frequency of the enterprise financial data;

[0061] If the data compliance of the enterprise's financial data is the third compliance level, the data update frequency of the enterprise's financial data is adjusted.

[0062] It should be noted that the duration of a billing cycle refers to the length of time of the billing cycle, the time period from the billing start date to the billing end date; the time difference between the billing start date and the financial settlement date is the difference between the billing cycle start date and the financial settlement date; the time difference between the transaction in the bill and the actual payment date refers to the time difference between the transaction date and the payment completion date; the billing delay time is the time difference between the transaction date and the bill generation date, where the duration of a billing cycle is generally a fixed time range, and the billing delay time is generally not a fixed number of days, but depends on the relative time between the transaction date and the billing date.

[0063] It should be noted that the account settings can be found in the ERP system, and the relevant information of account period management can be viewed to obtain the duration of the billing cycle, the billing start date and the financial settlement date. Based on the difference between the billing start date and the financial settlement date, the time difference between the billing start date and the financial settlement date can be obtained. Through the transaction record query module of the ERP system, the transaction date and the payment completion date can be found, and the difference can be calculated to obtain the time difference between the transaction in the bill and the actual payment. Through the ERP system, the transaction date and the bill generation date can be found, and the difference can be calculated to obtain the bill delay time.

[0064] It should be noted that the matching evaluation value of corporate financial data is analyzed in the following specific process:

[0065] ;

[0066] In the formula, Represents the matching evaluation value of the enterprise financial data, represents the duration of the i-th billing cycle, It represents the time difference between the start of the i-th bill and the financial settlement. It represents the time difference between the jth transaction in the i-th bill and the actual payment. represents the delay time of the i-th bill, Indicates the reference value corresponding to the duration of the set billing cycle. Indicates the reference value corresponding to the time difference between the set bill start and financial settlement. Indicates the maximum reference value corresponding to the time difference between each transaction in the set bill and the actual payment time. Indicates the reference value corresponding to the set bill delay time. Indicates the correction factor corresponding to the duration of the set billing cycle. Indicates the correction factor corresponding to the difference between the set billing start time and the financial settlement time. Indicates the correction factor corresponding to the time difference between each transaction in the set bill and the actual payment time. It represents the correction factor corresponding to the set bill delay time, i represents the number of each bill in the enterprise financial data, , m represents the total number of bills in the enterprise financial data, j represents the number of each transaction in the bill, , n represents the total number of transactions in the bill.

[0067] The correction factor corresponding to the duration of the billing cycle set in the database is recorded as , the role of this correction factor is to quantify the impact of the duration of the billing cycle on the matching evaluation value of the enterprise's financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the duration of the billing cycle and the correction factor preset in the database. When the duration of the real-time billing cycle is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0068] The correction factor corresponding to the difference between the billing start time and the financial settlement time set in the database is recorded as , the role of this correction factor is to quantify the impact of the time difference between the start of billing and financial settlement on the matching evaluation value of the enterprise's financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the time difference between the start of billing and financial settlement and the correction factor preset in the database. When the real-time time difference between the start of billing and financial settlement is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0069] The correction factor corresponding to the time difference between each transaction in the bill set in the database and the actual payment time is recorded as , the role of this correction factor is to quantify the impact of the time difference between each transaction in the bill and the actual payment on the matching evaluation value of the enterprise's financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the time difference between each transaction in the bill and the actual payment and the correction factor preset in the database. When the time difference between each transaction in the real-time bill and the actual payment is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0070] The correction factor corresponding to the bill delay time set in the database is recorded as , the role of this correction factor is to quantify the impact of the billing delay time on the matching evaluation value of the enterprise's financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the billing delay time and the correction factor preset in the database. When the real-time billing delay time is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0071] It should be noted that the duration of each billing cycle, the time difference between the start of each bill and financial settlement, the time difference between each transaction in each bill and the actual payment, and the delay time of each bill are related. The duration of the billing cycle includes the transaction time range, which directly affects the time difference between each transaction in each bill and the actual payment. For example, the longer the billing cycle, the greater the number of transactions and payment time difference that may occur; an increase in the time difference between the start of a bill and financial settlement may affect the payment and processing cycle in the bill, thereby extending the time difference between each transaction in the bill and the actual payment; the billing delay time is usually closely related to the time difference between transactions and payments. For example, as the time difference between transactions and payments increases, the billing delay time will be longer. From the above, it can be seen that the matching evaluation value of the enterprise's financial data is jointly determined by the duration of each billing cycle, the time difference between the start of each bill and financial settlement, the time difference between each transaction in each bill and the actual payment, and the delay time of each bill.

[0072] It should be noted that the matching evaluation value of the enterprise financial data is matched with the data conformity of the enterprise financial data in the matching evaluation value interval stored in the database. The data conformity of the enterprise financial data is preset in the database, and the data conformity of the enterprise financial data can be directly obtained from the database when used. The corresponding relationship can be a pre-set mapping relationship, for example, the matching evaluation value of the enterprise financial data and the data conformity of the enterprise financial data in the database form a mapping set, and the matching evaluation value of the real-time enterprise financial data is input into the mapping set to obtain the data conformity of the enterprise financial data, wherein the mapping relationship can be a one-to-one corresponding relationship. When the data compliance of enterprise financial data reaches high compliance or basic compliance, there is no need to adjust the update frequency of enterprise financial data, which helps to ensure the timeliness and accuracy of the data; when the data compliance of enterprise financial data is evaluated as generally compliant, the update frequency of enterprise financial data is extracted and compared with the update frequency range of enterprise financial data stored in the database. If the update frequency of enterprise financial data is not within the update frequency range of historical enterprise financial data stored in the database, the update frequency of enterprise financial data is adjusted, and the average of the update frequencies of historical enterprise financial data stored in the database is extracted. The update frequency of enterprise financial data is divided into the average of the update frequencies of historical enterprise financial data to obtain a ratio. If the update frequency of enterprise financial data is greater than the maximum value of the update frequency range of historical financial data, the update frequency of enterprise financial data is shortened according to the ratio. If the update frequency of enterprise financial data is less than the minimum value of the update frequency range of historical financial data, the update frequency of enterprise financial data is increased according to the ratio.

[0073] The above-mentioned reconciliation rule data of the enterprise financial data is obtained, and the reconciliation rule data of the enterprise financial data is analyzed to obtain the reconciliation rule compliance value of the enterprise financial data. The specific analysis process is:

[0074] The reconciliation rule data of the enterprise financial data includes: the length of the reconciliation cycle and the number of reconciliation operation interruptions;

[0075] The matching evaluation value of the enterprise financial data is extracted, and combined with the reconciliation rule data of the enterprise financial data to obtain the reconciliation rule compliance value of the enterprise financial data, and the reconciliation rule compliance value of the enterprise financial data is used to evaluate the compliance of the reconciliation rule of the enterprise financial data.

[0076] The above optimizes the reconciliation rules of the enterprise financial data according to the reconciliation rule compliance value of the enterprise financial data. The specific analysis process is as follows:

[0077] The reconciliation rules for the enterprise financial data include: the review frequency of the enterprise financial data and the amount of data processed by the reconciliation operation.

[0078] The reconciliation rule compliance value of the enterprise financial data is extracted, and matched with the review frequency of the enterprise accounting data corresponding to each interval of the reconciliation rule compliance value of the enterprise financial data stored in the database to obtain the review frequency of the enterprise accounting data.

[0079] The reconciliation rule compliance value of the enterprise financial data is extracted and compared with the reconciliation rule compliance value threshold of the enterprise financial data stored in the database. If the reconciliation rule compliance value of the enterprise financial data is higher than or equal to the reconciliation rule compliance value threshold of the enterprise financial data stored in the database, the reconciliation process of the enterprise financial data is continuously monitored.

[0080] If the reconciliation rule compliance value of the enterprise financial data is lower than the reconciliation rule compliance value threshold of the enterprise financial data stored in the database, the amount of data processed by the reconciliation operation is adjusted.

[0081] It should be noted that the length of a reconciliation cycle refers to the time spent from the start to the end of a reconciliation cycle; the number of reconciliation operation interruptions refers to the number of interruption events caused by faults, errors or human factors during the reconciliation process; the length of the reconciliation cycle is obtained by querying the start and end timestamps of the reconciliation in the reconciliation system, processing the difference, and then obtaining the length of the reconciliation cycle; using log analysis tools (such as grep, awk, etc.) to analyze the logs of the reconciliation system, extract log entries related to reconciliation interruptions, and count the number of entries to obtain the number of reconciliation operation interruptions.

[0082] It should be noted that the reconciliation rules for corporate financial data comply with the value, and the specific calculation formula is:

[0083] ;

[0084] In the formula, Indicates that the reconciliation rules of the enterprise's financial data meet the value. Indicates the length of the reconciliation cycle. Indicates the number of times the reconciliation operation was interrupted. Represents the matching evaluation value of the enterprise financial data, Indicates the reference value corresponding to the set reconciliation cycle duration. It indicates the reference value corresponding to the set number of reconciliation operation interruptions. e is a natural constant. Indicates the correction factor corresponding to the set reconciliation cycle duration. Indicates the correction factor corresponding to the set number of reconciliation operation interruptions. Indicates the correction factor corresponding to the matching evaluation value of the set enterprise financial data.

[0085] The correction factor corresponding to the reconciliation cycle length set in the database is recorded as , the role of this correction factor is to quantify the impact of the reconciliation cycle duration on the reconciliation rule compliance value of the company's financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the reconciliation cycle duration and the correction factors preset in the database. When the real-time reconciliation cycle duration is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0086] The correction factor corresponding to the number of reconciliation operation interruptions set in the database is recorded as , the role of this correction factor is to quantify the impact of the number of reconciliation operation interruptions on the reconciliation rule compliance value of the enterprise's financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the number of reconciliation operation interruptions and the correction factors preset in the database. When the real-time number of reconciliation operation interruptions is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0087] The correction factor corresponding to the matching evaluation value of the enterprise financial data set in the database is recorded as , the role of this correction factor is to quantify the impact of the matching evaluation value of the enterprise's financial data on the compliance value of the reconciliation rules of the enterprise's financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the matching evaluation value of the enterprise's financial data and the correction factor preset in the database. When the matching evaluation value of the real-time enterprise's financial data is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0088] like Figure 2 As shown, Figure 2 Represents the curve of the number of reconciliation operation interruptions - the reconciliation rule compliance value of the enterprise financial data, where the x-axis represents the number of reconciliation operation interruptions, and the y-axis represents the reconciliation rule compliance value of the enterprise financial data. Three different sets of example parameters are defined in the figure, corresponding to different situations of the three curves, represented by solid lines, dashed lines and dot-dash lines respectively, and the corresponding curve labels are a, b, and c respectively. Assume that represents the matching evaluation value of the enterprise financial data =1.6, indicating the reference value corresponding to the set reconciliation cycle duration =3 (days), indicating the reference value corresponding to the set number of reconciliation operation interruptions =3 (times), indicating the correction factor corresponding to the set reconciliation cycle duration =0.3, indicating the correction factor corresponding to the number of reconciliation operation interruptions =0.2, indicating the correction factor corresponding to the matching evaluation value of the enterprise's financial data =0.5, when the reconciliation cycle is long =1 (day), the diagram of the reconciliation rule compliance value of the enterprise financial data is shown in curve a. =2 (days), the diagram of the reconciliation rule compliance value of the enterprise financial data is shown in curve b. =3 (days), the schematic diagram of the reconciliation rules of the enterprise's financial data conforming to the value is shown in curve c.

[0089] Table 1 Example data of the reconciliation rules for corporate financial data Number of reconciliation operation interruptions (times) Reconciliation cycle duration (days) The reconciliation rules of corporate financial data comply with the value 0 2 2.442 1 2 1.784 2 2 1.595

[0090] As shown in Table 1, Table 1 is example data of reconciliation rule compliance values ​​of enterprise financial data, which lists the reconciliation rule compliance values ​​of enterprise financial data and the number of reconciliation operation interruptions.

[0091] As shown in Table 1, in a specific embodiment, assuming that the matching evaluation value of the enterprise financial data is =1.6, the reference value corresponding to the set reconciliation cycle duration =3 (days), the reference value corresponding to the number of interruptions of the reconciliation operation =3 (times), the correction factor corresponding to the set reconciliation cycle duration =0.3, correction factor corresponding to the number of reconciliation operation interruptions =0.2, the correction factor corresponding to the matching evaluation value of the enterprise financial data =0.5.

[0092] It should be noted that there is a correlation between the length of the reconciliation cycle and the number of interruptions in the reconciliation operation. The more times the reconciliation operation is interrupted, the more time it takes to troubleshoot and solve the problem each time, which will cause the reconciliation cycle to become longer. The higher the matching evaluation value of the enterprise's financial data, the higher the compliance of the enterprise's financial data, and the smaller the possibility of differences and errors in the reconciliation process, thereby improving the reconciliation rule compliance value of the enterprise's financial data. From the above, it can be seen that the reconciliation rule compliance value of the enterprise's financial data is jointly determined by the length of the reconciliation cycle, the number of interruptions in the reconciliation operation, and the matching evaluation value of the enterprise's financial data.

[0093] It should be noted that the reconciliation rule compliance value of the enterprise financial data is extracted and matched with the review frequency of the enterprise account data corresponding to each interval of the reconciliation rule compliance value of the enterprise financial data stored in the database. The review frequency of the enterprise account data is preset in the database, and the review frequency of the enterprise account data can be directly obtained from the database when used. The corresponding relationship can be a pre-set mapping relationship. For example, the reconciliation rule compliance value of the enterprise financial data and the review frequency of the enterprise account data in the database form a mapping set, and the real-time reconciliation rule compliance value of the enterprise financial data is input into the mapping set to obtain the review frequency of the enterprise account data. The mapping relationship can be a one-to-one relationship. The review frequency refers to the number of times or cycles that the enterprise reconciles within a certain period of time.

[0094] It should be noted that the reconciliation rule compliance value of the enterprise financial data is compared with the reconciliation rule compliance value threshold of the enterprise financial data stored in the database. If the reconciliation rule compliance value of the enterprise financial data is higher than or equal to the reconciliation rule compliance value threshold of the enterprise financial data stored in the database, the reconciliation process is continuously monitored. If the reconciliation rule compliance value of the enterprise financial data is lower than the reconciliation rule compliance value threshold of the enterprise financial data stored in the database, the amount of data processed by the reconciliation operation is adjusted. The adjustment of the amount of data processed by the reconciliation operation includes: extracting the amount of data processed by the reconciliation operation and comparing it with the average of the historical amounts of data processed by the reconciliation operation stored in the database. If the amount of data processed by the reconciliation operation is compared with the average of the historical amounts of data processed by the reconciliation operation stored in the database, if the amount of data processed by the reconciliation operation is greater than or equal to the average of the historical amounts of data processed by the reconciliation operation stored in the database, the amount of data processed by the reconciliation operation is halved to reduce excessive system pressure caused by processing too much data at a single time; if the amount of data processed by the reconciliation operation is less than the average of the historical amounts of data processed by the reconciliation operation stored in the database, adjacent processing batches are merged to increase the amount of data processed by the reconciliation operation, thereby improving system resource utilization.

[0095] According to the optimized reconciliation rules of enterprise financial data, the enterprise financial data is reconciled through AI technology to obtain the reconciliation data of the enterprise financial data. The specific process is as follows:

[0096] The review frequency of the enterprise's accounting data and the data volume of the reconciliation operation processing are extracted to perform reconciliation processing on the enterprise's financial data in batches until all the enterprise's financial data have been reconciled, and the reconciliation data of each time of the enterprise's financial data are obtained, wherein the reconciliation data of each time of the enterprise's financial data include: the number of mismatches in each reconciliation data, the number of differences in the status of each reconciliation transaction, the number of inconsistencies in the time of each reconciliation transaction, and the number of abnormal transactions in each reconciliation.

[0097] The above analysis of the reconciliation data of the enterprise's financial data obtains the reconciliation abnormal values ​​of the enterprise's financial data. The specific analysis process is as follows:

[0098] The number of mismatches in reconciliation data, the number of differences in reconciliation transaction status, the number of inconsistencies in reconciliation transaction time, and the number of abnormal reconciliation transactions are extracted, and combined with the reconciliation rule compliance value of the enterprise's financial data, the reconciliation abnormal value of the enterprise's financial data is processed to obtain the reconciliation abnormal value of the enterprise's financial data. The reconciliation abnormal value of the enterprise's financial data is used to evaluate the financial security of the enterprise.

[0099] The above assessment of the financial security of the enterprise is carried out in the following specific analysis process:

[0100] The reconciliation anomalies of the enterprise's financial data are extracted and matched with the financial security situations of the enterprises corresponding to each interval of the reconciliation anomalies of the enterprise's financial data stored in the database to obtain the financial security situation of the enterprise. The financial security situation of the enterprise includes: basic security (also called the first security level), intermediate security (also called the second security level), and advanced security (also called the third security level). The first security level, the second security level, and the third security level are all used to describe the level of financial data security. The levels of financial data security of the first security level, the second security level, and the third security level increase step by step.

[0101] When the financial security situation of an enterprise is basic security or intermediate security, the system will issue an early warning prompt and adjust and correct erroneous record entries. When the financial security situation of an enterprise is advanced security, the enterprise's financial data will be continuously monitored.

[0102] It should be noted that batch reconciliation processing refers to splitting the enterprise financial data into small batches for reconciliation processing based on the amount of data processed by the reconciliation operation, which can reduce the time and resources required for a single reconciliation and help improve the overall processing efficiency. When the enterprise financial data is reconciled in batches, the reconciliation data is obtained after each processing, and the reconciliation data of each time the enterprise financial data is processed is obtained after all the reconciliation processing is completed.

[0103] It should be noted that the number of data mismatches refers to the number of times no corresponding records are found during the reconciliation of the company's financial data; the number of transaction status differences refers to the cumulative number of different transaction statuses during the reconciliation of the company's financial data, for example, the bank confirms payment, but the company's financial data shows it as pending; the number of inconsistent transaction times refers to the cumulative number of times the transaction time recorded in the company's reconciliation data is inconsistent with the actual transaction time during the reconciliation of the company's financial data; abnormal transactions include refunds and returns. The number of abnormal transactions refers to the cumulative number of special transactions during the reconciliation of the company's financial data, special transactions.

[0104] It should be noted that when reconciling corporate financial data through AI technology, for example, by importing corporate financial data into the AI ​​reconciliation system through professional reconciliation software (such as Xero and QuickBooks), the AI ​​reconciliation system will first retrieve the corporate financial bills and, based on the preset reconciliation rules, check the corporate financial data against the bank bills one by one, identify transaction records and mark the differences. After the reconciliation is completed, a reconciliation result report is generated, which includes successfully matched transaction record entries, unmatched transaction records, and differences, and the error entries are classified, the types of error entries are obtained and recorded. After the reconciliation is completed, AI technology further parses the text content in the bank bill through natural language processing technology, and counts the classified error entries to generate detailed reports such as the number of reconciliation data mismatches, the number of transaction status differences, the number of inconsistent transaction times, and the number of abnormal transactions.

[0105] It should be noted that the reconciliation abnormal values ​​of the enterprise financial data are obtained by extracting the reconciliation data of each time of the enterprise financial data and processing them according to the reconciliation rules of the enterprise financial data. The reconciliation abnormal values ​​of the enterprise financial data are matched with the financial security situation of the enterprise stored in the database to evaluate the financial security situation of the enterprise. When the evaluation result is basic security and intermediate security, the system triggers the early warning mechanism, notifies the financial personnel by sending system messages, and reviews the error entries generated in the reconciliation process, makes preliminary modifications, and adjusts and corrects the error record entries, including: adjusting the transaction status, synchronizing the transaction time, and re-reconciling after the adjustment to obtain the reconciliation abnormal values ​​of the enterprise financial data, and evaluates the financial security situation of the enterprise. If the evaluation result is high-level security, the reconciliation is completed. If the evaluation result is not high-level security, the adjustment continues. When the evaluation results are not high-level security for multiple times, an alarm is issued to remind the relevant financial personnel to handle it. When the evaluation result is high-level security, the enterprise financial data is continuously monitored. By evaluating the financial security situation of the enterprise, it helps to improve the accuracy of the enterprise financial data and ensure financial security.

[0106] It should be noted that the specific calculation formula for the reconciliation abnormal value of the enterprise's financial data is:

[0107] ;

[0108] In the formula, Indicates the reconciliation abnormal value of the enterprise's financial data, Indicates the number of mismatches in the j-th reconciliation data. Indicates the number of differences in the status of the j-th reconciliation transaction, Indicates the number of inconsistent times for the jth reconciliation transaction. represents the number of abnormal transactions in the jth reconciliation. Indicates that the reconciliation rules of the enterprise's financial data meet the value. Indicates the maximum allowed value for the number of mismatches in reconciliation data. Indicates the maximum allowed value of the number of reconciliation transaction status differences. Indicates the maximum allowed value of the number of inconsistent transaction times. Indicates the maximum allowed value for the number of abnormal reconciliation transactions. Indicates the correction factor corresponding to the number of mismatches in reconciliation data. Indicates the correction factor corresponding to the number of differences in the reconciliation transaction status. Indicates the correction factor corresponding to the number of inconsistent reconciliation transaction times. Indicates the correction factor corresponding to the set number of abnormal reconciliation transactions. It indicates the correction factor corresponding to the compliance value of the reconciliation rule of the set enterprise financial data, and j indicates the number of each reconciliation time. , n represents the total number of reconciliations.

[0109] The correction factor corresponding to the number of mismatches of reconciliation data set in the database is recorded as , the role of this correction factor is to quantify the impact of the number of reconciliation data mismatches on the reconciliation anomalies of the company's financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the number of reconciliation data mismatches and the correction factors preset in the database. When the real-time number of reconciliation data mismatches is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0110] The correction factor corresponding to the number of reconciliation transaction status differences set in the database is recorded as , the role of this correction factor is to quantify the impact of the number of differences in the status of reconciliation transactions on the reconciliation anomalies of the company's financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the number of differences in the status of reconciliation transactions and the correction factors preset in the database. When the real-time number of differences in the status of reconciliation transactions is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0111] The correction factor corresponding to the number of inconsistent reconciliation transaction times set in the database is recorded as , the role of this correction factor is to quantify the impact of the number of inconsistent reconciliation transaction times on the reconciliation anomalies of the company's financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the number of inconsistent reconciliation transaction times and the correction factors preset in the database. When the real-time number of inconsistent reconciliation transaction times is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0112] The correction factor corresponding to the number of abnormal reconciliation transactions set in the database is recorded as , the role of this correction factor is to quantify the impact of the number of abnormal reconciliation transactions on the abnormal reconciliation values ​​of corporate financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the number of abnormal reconciliation transactions and the correction factors preset in the database. When the real-time number of abnormal reconciliation transactions is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0113] The correction factor corresponding to the reconciliation rule compliance value of the enterprise financial data set in the database is recorded as , the role of this correction factor is to quantify the impact of the reconciliation rule compliance value of the enterprise financial data on the reconciliation anomaly of the enterprise financial data. In practical applications, the correction factor can be directly obtained from the database, and its corresponding relationship is a pre-set mapping relationship. For example, a one-to-one or many-to-one mapping set is formed between the reconciliation rule compliance value of the enterprise financial data and the correction factor preset in the database. When the real-time reconciliation rule compliance value of the enterprise financial data is input into this mapping set, the corresponding correction factor can be obtained. In this example, the value range of the correction factor is limited to [0,1].

[0114] It should be noted that there is a correlation between the number of mismatches in reconciliation data, the number of differences in transaction status, the number of inconsistencies in transaction time, and the number of abnormal transactions in each reconciliation. Data mismatch may be caused by differences in transaction status, so an increase in the number of differences in transaction status may lead to an increase in the number of mismatches in reconciliation data; the number of inconsistencies in transaction time is proportional to the number of data mismatches, and an increase in the number of inconsistencies in transaction time will lead to an increase in the number of mismatches in reconciliation data; an increase in the number of abnormal transactions may lead to inaccurate data and thus an increase in the number of data mismatches, so an increase in the number of abnormal transactions in each reconciliation will cause an increase. From the above, it can be seen that the reconciliation abnormal values ​​of corporate financial data are jointly determined by the number of mismatches in reconciliation data, the number of differences in reconciliation transaction status, the number of inconsistencies in reconciliation transaction time, and the number of abnormal reconciliation transactions.

[0115] Specifically, based on the abnormal values ​​of the reconciliation of the company's financial data, AI generates a report to prompt relevant personnel to supplement. The specific analysis process is as follows:

[0116] Extract reconciliation anomalies from corporate financial data, analyze anomalies through AI and generate reports, send the reports to relevant persons in charge, and prompt them to make corrections, and use AI to track subsequent corrections.

[0117] It should be noted that in the process of reconciling corporate financial data through AI technology, the corporate financial data is first input into the AI ​​reconciliation system. The system will retrieve the corporate financial bills provided by the bank and, based on the preset reconciliation rules, reconcile the corporate financial data with the bank bills one by one, identify the transaction records and mark the differences. After the reconciliation is completed, the system generates a reconciliation result report, which includes successfully matched transaction records, unmatched transaction records and differences. On this basis, AI technology further identifies and marks abnormal transactions in corporate financial data through preset rules and historical data experience, and automatically matches abnormal situations and generates reports based on historical data and preset solutions. The report content includes information such as an overview of the error items, type analysis, and solutions. The report is automatically sent to relevant financial personnel, who review the error items based on the report content, supplement the causes of the anomalies, and correct the erroneous data based on the AI ​​solution. The AI ​​system will also continue to track these corrections, update the data and optimize its recognition and correction algorithms to improve the efficiency and accuracy of reconciliation.

[0118] Specifically, see Figure 3The second aspect of this article provides an AI-based automatic reconciliation optimization and processing system, including: a financial data adjustment module, which is used to automatically collect enterprise financial data from the enterprise's ERP system, analyze the enterprise financial data, obtain the matching evaluation value of the enterprise financial data, and automatically adjust the enterprise financial data according to the matching evaluation value of the enterprise financial data.

[0119] The reconciliation rule optimization module is used to monitor the reconciliation process of the enterprise's financial data, obtain the reconciliation rule data of the enterprise's financial data, analyze the reconciliation rule data of the enterprise's financial data, obtain the reconciliation rule compliance value of the enterprise's financial data, and optimize the reconciliation rules of the enterprise's financial data according to the reconciliation rule compliance value of the enterprise's financial data.

[0120] The abnormal data reporting module is used to reconcile the enterprise financial data through AI technology according to the optimized reconciliation rules of the enterprise financial data, obtain the reconciliation data of the enterprise financial data, and analyze the reconciliation data of the enterprise financial data to obtain the reconciliation abnormal values ​​of the enterprise financial data. According to the reconciliation abnormal values ​​of the enterprise financial data, a report is generated through AI to prompt relevant personnel to supplement.

[0121] It should be noted that the financial data adjustment module collects and analyzes the company's financial data through the company's ERP system, obtains the matching evaluation value of the company's financial data, and adjusts the company's financial data. At the same time, the reconciliation rule optimization module monitors the reconciliation rule data of the company's financial data, and combines it with the reconciliation rule data of the company's financial data to obtain the reconciliation rule compliance value of the company's financial data, and optimizes the reconciliation rules of the company's financial data. The abnormal data reporting module reconciles according to the optimized reconciliation rules, obtains the reconciliation data of the company's financial data, and analyzes the reconciliation abnormal values ​​of the company's financial data, identifies anomalies, generates reports through AI and issues reminders, which helps to improve the efficiency and accuracy of financial data processing.

[0122] It should be noted that the above-mentioned AI-based automatic reconciliation optimization and processing method or system also includes a database for storing and managing corporate financial data. In this embodiment, the database is used to store the matching evaluation threshold of corporate financial data, the reference value corresponding to the duration of the set billing cycle, the reference value corresponding to the set time difference between the start of the bill and the financial settlement, the maximum reference value corresponding to the time difference between each transaction in the bill and the actual payment, the reference value corresponding to the set billing delay time, the correction factor corresponding to the duration of the set billing cycle, the correction factor corresponding to the set time difference between the start of the bill and the financial settlement, the correction factor corresponding to the time difference between each transaction in the bill and the actual payment, the correction factor corresponding to the set billing delay time, the data compliance of corporate financial data, the update frequency range of corporate financial data, the average update frequency of historical corporate financial data, the reconciliation rule compliance value threshold of corporate financial data, and the set reconciliation cycle time. The reference value corresponding to the set length, the reference value corresponding to the set number of reconciliation operation interruptions, the correction factor corresponding to the set reconciliation cycle length, the correction factor corresponding to the set number of reconciliation operation interruptions, the correction factor corresponding to the set matching assessment value of the enterprise's financial data, the review frequency of the enterprise's accounting data, the average amount of data processed by historical reconciliation operations, the financial security of the enterprise, the maximum allowable value corresponding to the set number of reconciliation data mismatches, the maximum allowable value corresponding to the set number of reconciliation transaction status differences, the maximum allowable value corresponding to the set number of reconciliation transaction time inconsistencies, the maximum allowable value corresponding to the number of abnormal reconciliation transactions, the correction factor corresponding to the set number of reconciliation data mismatches, the correction factor corresponding to the set number of reconciliation transaction status differences, the correction factor corresponding to the set number of reconciliation transaction time inconsistencies, the correction factor corresponding to the set number of abnormal reconciliation transactions, and the correction factor corresponding to the set reconciliation rule compliance value of the enterprise's financial data.

[0123] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0124] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An AI-based automatic reconciliation optimization and processing method, characterized in that: include: Automatically collect enterprise financial data from the enterprise's ERP system, analyze the enterprise financial data, obtain the matching evaluation value of the enterprise financial data, and automatically adjust the enterprise financial data according to the matching evaluation value of the enterprise financial data; Monitor the reconciliation process of the enterprise's financial data, obtain the reconciliation rule data of the enterprise's financial data, analyze the reconciliation rule data of the enterprise's financial data, obtain the reconciliation rule compliance value of the enterprise's financial data, and optimize the reconciliation rule of the enterprise's financial data according to the reconciliation rule compliance value of the enterprise's financial data; According to the optimized reconciliation rules of enterprise financial data, the enterprise financial data is reconciled through AI technology to obtain the reconciliation data of the enterprise financial data, and the reconciliation data of the enterprise financial data is analyzed to obtain the reconciliation anomalies of the enterprise financial data. According to the reconciliation anomalies of the enterprise financial data, a report is generated through AI to prompt relevant personnel to supplement.

2. According to claim 1, an AI-based automatic reconciliation optimization and processing method is characterized by: The matching evaluation value of the enterprise financial data is specifically analyzed as follows: Collecting enterprise financial data in the ERP system, the enterprise financial data including: the duration of each billing cycle, the time difference between the start of each billing and the financial settlement, the transaction date of each billing, the payment date of each billing, and the delay time of each billing; The duration of each billing cycle, the time difference between the start of each bill and the financial settlement, the transaction date of each bill, the payment date of each bill, and the delay time of each bill are analyzed and processed to obtain a matching evaluation value of the enterprise's financial data, which is used to evaluate the compliance of the enterprise's financial data.

3. According to claim 1, an AI-based automatic reconciliation optimization and processing method is characterized by: According to the matching evaluation value of the enterprise financial data, the enterprise financial data is automatically adjusted, and the specific analysis is as follows: Extracting the matching evaluation value of the enterprise financial data, matching it with the data conformity corresponding to each interval of the matching evaluation value of the enterprise financial data stored in the database, and obtaining the data conformity category of the enterprise financial data, wherein the data conformity of the enterprise financial data includes: a first conformity degree, a second conformity degree, and a third conformity degree, wherein the first conformity degree, the second conformity degree, and the third conformity degree are all used to describe the level of data conformity, and the levels of data conformity of the first conformity degree, the second conformity degree, and the third conformity degree decrease step by step; If the data compliance of the enterprise financial data is the first compliance level or the second compliance level, there is no need to adjust the data update frequency of the enterprise financial data; If the data compliance of the enterprise's financial data is the third compliance level, the data update frequency of the enterprise's financial data is adjusted.

4. According to claim 1, an AI-based automatic reconciliation optimization and processing method is characterized by: The reconciliation rule data of the enterprise financial data is obtained, and the reconciliation rule data of the enterprise financial data is analyzed to obtain the reconciliation rule compliance value of the enterprise financial data. The specific analysis is as follows: The reconciliation rule data of the enterprise financial data includes: the length of the reconciliation cycle and the number of reconciliation operation interruptions; The matching evaluation value of the enterprise financial data is extracted, and combined with the reconciliation rule data of the enterprise financial data to obtain the reconciliation rule compliance value of the enterprise financial data, and the reconciliation rule compliance value of the enterprise financial data is used to evaluate the compliance of the reconciliation rule of the enterprise financial data.

5. According to claim 1, an AI-based automatic reconciliation optimization and processing method is characterized by: The reconciliation rules of the enterprise financial data are optimized according to the reconciliation rule compliance value of the enterprise financial data. The specific analysis is as follows: The reconciliation rules for the enterprise financial data include: the review frequency of the enterprise financial data and the amount of data processed by the reconciliation operation; Extract the reconciliation rule compliance value of the enterprise financial data, match it with the review frequency of the enterprise accounting data corresponding to each interval of the reconciliation rule compliance value of the enterprise financial data stored in the database, and obtain the review frequency of the enterprise accounting data; Extract the reconciliation rule compliance value of the enterprise financial data, and compare it with the reconciliation rule compliance value threshold of the enterprise financial data stored in the database. If the reconciliation rule compliance value of the enterprise financial data is higher than or equal to the reconciliation rule compliance value threshold of the enterprise financial data stored in the database, continuously monitor the reconciliation process of the enterprise financial data; If the reconciliation rule compliance value of the enterprise financial data is lower than the reconciliation rule compliance value threshold of the enterprise financial data stored in the database, the amount of data processed by the reconciliation operation is adjusted.

6. According to claim 5, an AI-based automatic reconciliation optimization and processing method is characterized by: According to the optimized reconciliation rules of enterprise financial data, the enterprise financial data is reconciled by using AI technology to obtain the reconciliation data of the enterprise financial data. The specific process is as follows: The review frequency of the enterprise's accounting data and the data volume of the reconciliation operation processing are extracted to perform reconciliation processing on the enterprise's financial data in batches until all the enterprise's financial data have been reconciled, and the reconciliation data of each time of the enterprise's financial data are obtained, wherein the reconciliation data of each time of the enterprise's financial data include: the number of mismatches in each reconciliation data, the number of differences in the status of each reconciliation transaction, the number of inconsistencies in the time of each reconciliation transaction, and the number of abnormal transactions in each reconciliation.

7. The AI-based automatic reconciliation optimization and processing method according to claim 6 is characterized by: The reconciliation data of the enterprise financial data is analyzed to obtain the reconciliation abnormal values ​​of the enterprise financial data. The specific analysis is as follows: The number of mismatches in reconciliation data, the number of differences in reconciliation transaction status, the number of inconsistencies in reconciliation transaction time, and the number of abnormal reconciliation transactions are extracted, and combined with the reconciliation rule compliance value of the enterprise's financial data, the reconciliation abnormal value of the enterprise's financial data is processed to obtain the reconciliation abnormal value of the enterprise's financial data. The reconciliation abnormal value of the enterprise's financial data is used to evaluate the financial security of the enterprise.

8. According to claim 7, an AI-based automatic reconciliation optimization and processing method is characterized by: The financial security of the enterprise is evaluated as follows: Extracting the reconciliation abnormal values ​​of the enterprise's financial data and matching them with the financial security conditions of the enterprise corresponding to each interval of the reconciliation abnormal values ​​of the enterprise's financial data stored in the database, to obtain the financial security conditions of the enterprise, the financial security conditions of the enterprise including: a first security level, a second security level and a third security level, the first security level, the second security level and the third security level are all used to describe the level of financial data security, and the first security level, the second security level and the third security level of financial data security increase step by step; When the financial security situation of an enterprise is at the first security level or the second security level, the system will issue an early warning and adjust and correct erroneous record entries. When the financial security situation of an enterprise is at the third security level, the enterprise's financial data will be continuously monitored.

9. The AI-based automatic reconciliation optimization and processing method according to claim 1, characterized in that: According to the abnormal reconciliation values ​​of the enterprise's financial data, a report is generated through AI to prompt relevant personnel to supplement. The specific analysis is as follows: Extract reconciliation anomalies from corporate financial data, analyze anomalies through AI and generate reports, send the reports to relevant persons in charge, and prompt them to make corrections, and use AI to track subsequent corrections.

10. An AI-based automatic reconciliation optimization and processing system, characterized by: include: The financial data adjustment module is used to automatically collect the enterprise's financial data from the enterprise's ERP system, analyze the enterprise's financial data, obtain the matching evaluation value of the enterprise's financial data, and automatically adjust the enterprise's financial data according to the matching evaluation value of the enterprise's financial data; The reconciliation rule optimization module is used to monitor the reconciliation process of the enterprise's financial data, obtain the reconciliation rule data of the enterprise's financial data, analyze the reconciliation rule data of the enterprise's financial data, obtain the reconciliation rule compliance value of the enterprise's financial data, and optimize the reconciliation rule of the enterprise's financial data according to the reconciliation rule compliance value of the enterprise's financial data; The abnormal data reporting module is used to reconcile the enterprise financial data through AI technology according to the optimized reconciliation rules of the enterprise financial data, obtain the reconciliation data of the enterprise financial data, and analyze the reconciliation data of the enterprise financial data to obtain the reconciliation abnormal values ​​of the enterprise financial data. According to the reconciliation abnormal values ​​of the enterprise financial data, a report is generated through AI to prompt relevant personnel to supplement.

Images