Bill settlement method and system

Abstract

The embodiment of the invention discloses a bill settlement method and system, and the method comprises the steps: collecting bill data information to a preset image processing module, and carrying out the preprocessing of the bill data information, and obtaining the refined data information of the bill data; carrying out key information interception on the refinement data information of the bill data to obtain pre-settlement information of the bill data; analyzing and comparing the pre-settlement information of the bill data with the original bill settlement information to obtain a bill settlement difference value; whether the bill settlement difference value is equal to a preset difference value threshold value or not is judged, if yes, pre-settlement information of the bill data is sent to a preset bill data settlement module for settlement, and the pre-settlement information is uploaded to a preset data storage module; if not, the bill checking information is triggered, and the pre-settlement information of the bill data is checked again. According to the invention, analysis and deep learning are carried out on the intelligently identified bill data information and the original bill data information, so that the efficiency and accuracy of bill settlement are improved.

Description

Technical Field

[0001] This invention relates to the field of data processing, and more specifically, to a billing settlement method and system. Background Technology

[0002] With the application of various smart devices and big data technologies, people have placed higher demands on the efficiency and accuracy of bill management. However, traditional bill settlement methods usually rely on manual review and data entry, which is both time-consuming and error-prone, especially when processing large volumes of bills, resulting in low efficiency and difficulty in guaranteeing accuracy. Therefore, existing technologies have shortcomings and urgently need improvement. Summary of the Invention

[0003] In view of the above problems, the purpose of this invention is to provide a bill settlement method and system that automates the processing of bill data through intelligent recognition technology, greatly improving settlement efficiency and accuracy. At the same time, it performs deep learning and analysis on the bill data, providing strong support for the efficiency and accuracy of bill settlement.

[0004] The first aspect of this invention provides a bill settlement method, comprising:

[0005] Based on the preset image acquisition module, the bill data information is collected into the preset image processing module;

[0006] The preset image processing module preprocesses the bill data to obtain refined bill data.

[0007] By extracting key information from the refined data of the billing data, the pre-settlement information of the billing data can be obtained;

[0008] Obtain original billing and settlement information;

[0009] The pre-settlement information of the billing data is analyzed and compared with the original billing settlement information to obtain the billing settlement difference;

[0010] Determine whether the bill settlement difference is equal to the preset difference threshold. If so, send the pre-settlement information of the bill data to the preset bill data settlement module for settlement and upload it to the preset data storage module.

[0011] If not, trigger the bill reconciliation information to recalculate the pre-settlement information of the bill data.

[0012] This plan also includes:

[0013] Time information for obtaining billing data;

[0014] Based on the time information of the bill data, the bill data is sorted according to a preset time rule to obtain bill data information arranged in chronological order;

[0015] Based on preset time period nodes, the bill data information arranged in chronological order is classified and stored to obtain the bill data information within each time period node.

[0016] The bill data is then sent to a preset image acquisition module for data processing.

[0017] In this solution, the step of the preset image processing module preprocessing the bill data to obtain refined bill data specifically includes:

[0018] Retrieve image information from billing data;

[0019] Based on the image information of the bill data, obtain the pixel information of the bill image data;

[0020] Determine whether the pixel information of the bill image data is greater than the preset pixel threshold. If so, set the pixel information of the bill image data that is greater than the preset pixel threshold to the preset first fixed pixel value.

[0021] If not, set the pixel information of bill image data that is less than the preset pixel threshold to the preset second fixed pixel value;

[0022] The determined bill image data is processed and saved to obtain refined bill data.

[0023] In this solution, the step of extracting key information from the refined information of the bill data to obtain the pre-settlement information of the bill data specifically includes:

[0024] Send the refined information of the bill data to the preset bill data intelligent analysis module;

[0025] Based on preset key information rules, set key information for billing data;

[0026] The pre-set bill data intelligent analysis module compares and analyzes key information in the refined information of the bill data to obtain the number of associations of key information.

[0027] Determine if the number of associations of key information exceeds the preset association threshold. If so, mark the refined information of the corresponding bill data to obtain the pre-settlement information of the bill data.

[0028] If not, trigger the manual review information and send the corresponding bill data refinement information to the preset manual review module for manual review and judgment.

[0029] This plan also includes:

[0030] Obtain the original billing documents;

[0031] By statistically analyzing each original invoice voucher, the original invoice data information is obtained;

[0032] The original billing data is reviewed to obtain the review results.

[0033] Determine whether the audit result information meets the preset audit threshold. If so, send the original bill data information to the preset bill data settlement module and wait for further bill settlement.

[0034] If not, the corresponding original bill will be marked as invalid and will not be settled.

[0035] In this solution, the step of triggering bill reconciliation information and recalculating the pre-settlement information of the bill data specifically includes:

[0036] Send the pre-settlement information of the billing data to the preset manual review module;

[0037] The preset manual review module reviews the pre-settlement information and obtains the review results.

[0038] If the review results are abnormal, the corresponding abnormal settlement information will be corrected.

[0039] If the review result is normal, the original billing data will be reviewed and entered to obtain the corrected original billing data.

[0040] This plan also includes:

[0041] Retrieve information on the number of times bill reconciliation information has been triggered;

[0042] Determine if the number of times the bill reconciliation information is triggered exceeds the preset threshold. If not, proceed with the bill settlement according to the normal process.

[0043] If so, a bill settlement warning will be triggered, and intelligent recognition of the current bill data will be paused. The preset manual review module will be activated to settle the bill data.

[0044] A second aspect of the present invention provides a billing system, characterized in that it includes a memory and a processor, wherein the memory stores a billing method program, and the billing method program, when executed by the processor, performs the following steps:

[0045] Based on the preset image acquisition module, the bill data information is collected into the preset image processing module;

[0046] The preset image processing module preprocesses the bill data to obtain refined bill data.

[0047] By extracting key information from the refined data of the billing data, the pre-settlement information of the billing data can be obtained;

[0048] Obtain original billing and settlement information;

[0049] The pre-settlement information of the billing data is analyzed and compared with the original billing settlement information to obtain the billing settlement difference;

[0050] Determine whether the bill settlement difference is equal to the preset difference threshold. If so, send the pre-settlement information of the bill data to the preset bill data settlement module for settlement and upload it to the preset data storage module.

[0051] If not, trigger the bill reconciliation information to recalculate the pre-settlement information of the bill data.

[0052] This plan also includes:

[0053] Time information for obtaining billing data;

[0054] Based on the time information of the bill data, the bill data is sorted according to a preset time rule to obtain bill data information arranged in chronological order;

[0055] Based on preset time period nodes, the bill data information arranged in chronological order is classified and stored to obtain the bill data information within each time period node.

[0056] The bill data is then sent to a preset image acquisition module for data processing.

[0057] In this solution, the step of the preset image processing module preprocessing the bill data to obtain refined bill data specifically includes:

[0058] Retrieve image information from billing data;

[0059] Based on the image information of the bill data, obtain the pixel information of the bill image data;

[0060] Determine whether the pixel information of the bill image data is greater than the preset pixel threshold. If so, set the pixel information of the bill image data that is greater than the preset pixel threshold to the preset first fixed pixel value.

[0061] If not, set the pixel information of bill image data that is less than the preset pixel threshold to the preset second fixed pixel value;

[0062] The determined bill image data is processed and saved to obtain refined bill data.

[0063] In this solution, the step of extracting key information from the refined information of the bill data to obtain the pre-settlement information of the bill data specifically includes:

[0064] Send the refined information of the bill data to the preset bill data intelligent analysis module;

[0065] Based on preset key information rules, set key information for billing data;

[0066] The pre-set bill data intelligent analysis module compares and analyzes key information in the refined information of the bill data to obtain the number of associations of key information.

[0067] Determine if the number of associations of key information exceeds the preset association threshold. If so, mark the refined information of the corresponding bill data to obtain the pre-settlement information of the bill data.

[0068] If not, trigger the manual review information and send the corresponding bill data refinement information to the preset manual review module for manual review and judgment.

[0069] This plan also includes:

[0070] Obtain the original billing documents;

[0071] By statistically analyzing each original invoice voucher, the original invoice data information is obtained;

[0072] The original billing data is reviewed to obtain the review results.

[0073] Determine whether the audit result information meets the preset audit threshold. If so, send the original bill data information to the preset bill data settlement module and wait for further bill settlement.

[0074] If not, the corresponding original bill will be marked as invalid and will not be settled.

[0075] In this solution, the step of triggering bill reconciliation information and recalculating the pre-settlement information of the bill data specifically includes:

[0076] Send the pre-settlement information of the billing data to the preset manual review module;

[0077] The preset manual review module reviews the pre-settlement information and obtains the review results.

[0078] If the review results are abnormal, the corresponding abnormal settlement information will be corrected.

[0079] If the review result is normal, the original billing data will be reviewed and entered to obtain the corrected original billing data.

[0080] This plan also includes:

[0081] Retrieve information on the number of times bill reconciliation information has been triggered;

[0082] Determine if the number of times the bill reconciliation information is triggered exceeds the preset threshold. If not, proceed with the bill settlement according to the normal process.

[0083] If so, a bill settlement warning will be triggered, and intelligent recognition of the current bill data will be paused. The preset manual review module will be activated to settle the bill data.

[0084] The present invention discloses a bill settlement method and system that improves the accuracy of intelligently identified bill data by analyzing and performing deep learning on the intelligently identified bill data information and the original bill data information. Attached Figure Description

[0085] The accompanying drawings are provided to further illustrate embodiments of the present invention and form part of the specification. They are used together with the following detailed description to explain the embodiments of the present invention, but do not constitute a limitation thereof. In the drawings:

[0086] Figure 1 A flowchart of a bill settlement method according to an embodiment of the present invention is shown;

[0087] Figure 2 A flowchart illustrating the steps of intelligently recognizing bill data according to an embodiment of the present invention is shown;

[0088] Figure 3 A block diagram of a billing system according to an embodiment of the present invention is shown. Detailed Implementation

[0089] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0090] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0091] Figure 1 A flowchart of a bill settlement method according to the present invention is shown.

[0092] like Figure 1 As shown, this invention discloses a bill settlement method, including:

[0093] S101, based on the preset image acquisition module, collects bill data information into the preset image processing module;

[0094] S102, The preset image processing module preprocesses the bill data information to obtain the refined bill data information;

[0095] S103, by extracting key information from the refined data of the billing data, the pre-settlement information of the billing data is obtained;

[0096] S104, Obtain original billing settlement information;

[0097] S105, Analyze and compare the pre-settlement information of the bill data with the original bill settlement information to obtain the bill settlement difference;

[0098] S106, determine whether the bill settlement difference is equal to the preset difference threshold. If so, send the pre-settlement information of the bill data to the preset bill data settlement module for settlement, and upload it to the preset data storage module.

[0099] S107, if not, trigger bill reconciliation information to recalculate the pre-settlement information of the bill data.

[0100] It should be noted that the billing data is collected by a preset image acquisition module, such as a scanner, camera, mobile phone, or PC. The billing data is then transmitted in image format to a preset image processing module for processing to obtain refined billing data. Key information is then extracted using intelligent recognition technology to obtain the pre-settlement information x. The original billing data y is obtained, and the pre-settlement information x obtained through intelligent recognition is compared with the original billing data y to obtain the bill settlement difference information z, z = |xy|. If the preset difference threshold is set to 0, then when z = 0, it indicates that the pre-settlement information obtained through intelligent recognition is consistent with the original billing data, and the pre-settlement information obtained through intelligent recognition is sent to the preset billing settlement module for bill settlement. When z ≠ 0, bill recalculation is triggered, indicating that there is a discrepancy between the pre-settlement information obtained through intelligent recognition, and the pre-settlement information needs to be recalculated.

[0101] According to an embodiment of the present invention, it further includes:

[0102] Time information for obtaining billing data;

[0103] Based on the time information of the bill data, the bill data is sorted according to a preset time rule to obtain bill data information arranged in chronological order;

[0104] Based on preset time period nodes, the bill data information arranged in chronological order is classified and stored to obtain the bill data information within each time period node.

[0105] The bill data is then sent to a preset image acquisition module for data processing.

[0106] It should be noted that the time information for obtaining bill data generally refers to the time when the bill occurred. Based on this time information, the bill data is sorted and then categorized according to a preset time rule, namely, by the start and end times of the natural month. For example, the start time is 00:00 on the 1st of each month, and the end time is 24:00 on the 30th of the month (if the month has a 31st, then the end time is 24:00 on the 31st). This yields bill data categorized by month, and the monthly bill data is processed on a monthly cycle. The preset time cycle is set by professionals in the field based on actual conditions such as the volume of bill settlement tasks.

[0107] According to an embodiment of the present invention, the step of the preset image processing module preprocessing bill data information to obtain refined bill data information specifically includes:

[0108] Retrieve image information from billing data;

[0109] Based on the image information of the bill data, obtain the pixel information of the bill image data;

[0110] Determine whether the pixel information of the bill image data is greater than the preset pixel threshold. If so, set the pixel information of the bill image data that is greater than the preset pixel threshold to the preset first fixed pixel value.

[0111] If not, set the pixel information of bill image data that is less than the preset pixel threshold to the preset second fixed pixel value;

[0112] The determined bill image data is processed and saved to obtain refined bill data.

[0113] It should be noted that after the preset image processing module acquires the image information of the bill data, it obtains the pixel information α of the image based on the image information of the bill data. The preset pixel threshold is set to 120. Therefore, when the pixel information α of the bill image data is ≥ 120, the pixels of the bill image data with a pixel value greater than 120 are set to the preset first fixed pixel value, i.e., 255; when α < 120, the pixels of the bill image data with a pixel value less than 120 are set to the preset second fixed pixel value, i.e., 0. Through the above operation, the image information of the bill data is converted into a black and white binary image, thereby simplifying the bill image data to facilitate subsequent image processing and analysis. The preset pixel threshold is set by those skilled in the art according to the actual situation.

[0114] Figure 2 A flowchart illustrating the steps of the present invention for intelligently identifying billing data is shown.

[0115] like Figure 2 As shown in the embodiment of the present invention, the step of obtaining the pre-settlement information of the bill data by extracting key information from the refined information of the bill data specifically includes:

[0116] S201, send the refined information of the bill data to the preset bill data intelligent analysis module;

[0117] S202, Based on preset key information rules, set key information for billing data;

[0118] S203, The pre-set bill data intelligent analysis module compares and analyzes key information of the refined information of the bill data to obtain the number of associations of key information;

[0119] S204, determine whether the number of associations of key information is greater than the preset association number threshold. If so, mark the refined information of the corresponding bill data to obtain the pre-settlement information of the bill data.

[0120] S205, if not, trigger manual review information, send the corresponding bill data refinement information to the preset manual review module, and wait for manual review and judgment.

[0121] It should be noted that the pre-set bill data intelligent analysis module is an intelligent module with functions such as data analysis, data processing, and deep learning. After receiving the refined information of the bill data, it sets keyword information related to bill settlement based on preset key information rules. The preset key information rules mainly include setting the time format to extract the bill settlement time, setting the amount value format to extract the bill settlement amount, and setting the company or unit name to extract the bill's counterparty information, etc. The pre-set bill data intelligent analysis module extracts key information based on the refined information of the bill data and records the number of times the key information is triggered, obtaining the association number N of the key information. The preset association number threshold is set to 2. When the association number N of the key information is ≥ 2, it means that there is a lot of keyword information triggered that is related to bill settlement, and it can be used as the pre-settlement information of the bill data. When the association number N of the key information is < 2, it triggers the manual review information, and the corresponding refined information of the bill data needs to be sent to the preset manual review module for manual review and judgment.

[0122] According to an embodiment of the present invention, it further includes:

[0123] Obtain the original billing documents;

[0124] By statistically analyzing each original invoice voucher, the original invoice data information is obtained;

[0125] The original billing data is reviewed to obtain the review results.

[0126] Determine whether the audit result information meets the preset audit threshold. If so, send the original bill data information to the preset bill data settlement module and wait for further bill settlement.

[0127] If not, the corresponding original bill will be marked as invalid and will not be settled.

[0128] It should be noted that in order to filter out original bill data that meets the bill settlement requirements, the original bill data needs to be reviewed; original bill vouchers are obtained, and each bill voucher is statistically calculated and reviewed, and the original bill data information is reviewed to obtain the review result; only if it meets the preset review threshold, that is, the review is passed, can the original bill data information be sent to the preset bill settlement module for settlement; if the review result is not passed, it means that the original bill does not meet the bill settlement requirements and is marked as an invalid bill.

[0129] According to an embodiment of the present invention, the step of triggering bill reconciliation information and recalculating the pre-settlement information of the bill data specifically includes:

[0130] Send the pre-settlement information of the billing data to the preset manual review module;

[0131] The preset manual review module reviews the pre-settlement information and obtains the review results.

[0132] If the review results are abnormal, the corresponding abnormal settlement information will be corrected.

[0133] If the review result is normal, the original billing data will be reviewed and entered to obtain the corrected original billing data.

[0134] It should be noted that the pre-settlement information of the bill data processed by the bill data intelligent analysis module is sent to the preset manual review module. The preset manual review module reviews the pre-settlement information of the bill data. The review mainly confirms the compliance and accuracy of the pre-settlement information of the bill data and obtains the review result. If the review result is abnormal, the corresponding pre-settlement information needs to be corrected. If the review result is normal, it means that the pre-settlement information of the bill data meets the requirements of bill settlement. Therefore, the original bill data is then reviewed and entered into the system to obtain the reviewed original bill data.

[0135] According to an embodiment of the present invention, it further includes:

[0136] Retrieve information on the number of times bill reconciliation information has been triggered;

[0137] Determine if the number of times the bill reconciliation information is triggered exceeds the preset threshold. If not, proceed with the bill settlement according to the normal process.

[0138] If so, a bill settlement warning will be triggered, and intelligent recognition of the current bill data will be paused. The preset manual review module will be activated to settle the bill data.

[0139] It should be noted that when the bill settlement difference z≠0, bill reconciliation information will be triggered, and the number of times the bill reconciliation information is triggered, T, will be obtained. The preset number threshold is set to 2. When T<2, the bill settlement steps will continue to be executed according to the original process. If T≥2, a bill settlement warning will be triggered, indicating that the bill data currently identified by the intelligent system has errors and the number of errors is too high. The intelligent system needs to be paused and the preset manual review module will be started to settle the bill data.

[0140] According to an embodiment of the present invention, it further includes:

[0141] Retrieve historical billing data;

[0142] Historical billing data is divided and calculated according to preset time period nodes to obtain billing data for each time period;

[0143] The average billing data within the preset time period is calculated based on the billing data information of each time period.

[0144] Retrieve bill settlement data information for the current time period;

[0145] By calculating and analyzing the average bill data within a preset time period and the bill settlement data at the current time period, the bill settlement data deviation value is obtained.

[0146] Determine if the deviation value of the bill settlement data is greater than the preset deviation threshold. If so, trigger the bill settlement anomaly information and send the bill data for the current month to the preset bill data intelligent analysis module for analysis.

[0147] It should be noted that historical billing data is obtained and divided according to preset time period nodes, such as dividing historical billing data into monthly billing data according to the natural month time period; the average monthly billing settlement data P is calculated; the difference between the average monthly billing settlement data P and the current month's billing settlement data Q is calculated to obtain the billing settlement data deviation value L, L=|QP|. The preset deviation threshold is set to 5000. When L≥5000, a billing settlement anomaly information is triggered, indicating that the current month's billing settlement data exceeds or falls significantly below the historical monthly billing average settlement data, indicating a significant anomaly. The current month's billing data needs to be sent to the preset billing data intelligent analysis module for analysis to find the reason for the excess or below-average level in the current month. The preset time period nodes, preset difference thresholds, etc., are all set by those skilled in the art according to the actual situation.

[0148] According to an embodiment of the present invention, the step of triggering abnormal bill settlement information and sending the monthly bill data to a preset intelligent bill data analysis module for analysis specifically includes:

[0149] Retrieve billing and settlement information;

[0150] The pre-set bill data intelligent analysis module extracts and constructs features from the target bill data based on the bill settlement association information to obtain the bill settlement feature information;

[0151] The pre-set bill data intelligent analysis module sends the characteristic information of bill settlement to the pre-set deep learning module for learning and analysis to obtain abnormal bill data information;

[0152] The abnormal billing data is sent to the preset manual review module for review, and the review result information is obtained.

[0153] The audit results are sent to a pre-defined deep learning module for secondary learning and data analysis.

[0154] The characteristic information of the bill settlement includes the time information, amount information, settlement object information, and settlement reason information of the bill settlement.

[0155] It should be noted that bill-related information mainly refers to information related to bill settlement, involving monetary figures, time formats, and so on. The pre-set intelligent bill data analysis module performs detailed feature extraction on the target bill data based on the bill settlement related information, and rearranges the bill-related information according to settlement logic and rules to construct bill settlement feature information with settlement logic. The pre-set intelligent bill data analysis module sends the bill settlement feature information to the pre-set deep learning module for learning and analysis, removing duplicate bill data and invalid bill data to obtain abnormal bill data information, such as large income and expenditure information, income and expenditure information in abnormal time periods, and abnormal settlement data between counterparties. The abnormal bill data is then sent to the pre-set manual review module for review by professional financial personnel to obtain corresponding review results. The review results are then fed back to the pre-set deep learning module to back up, learn, and analyze the reasons for the abnormality of the abnormal bill data and the handling results of the review, and whether it possesses the corresponding characteristics, providing data support and assistance for future settlement decisions.

[0156] According to an embodiment of the present invention, the step of sending the review result information to a preset deep learning module for secondary learning and data analysis specifically includes:

[0157] Based on the preset manual review module, obtain abnormal billing data information;

[0158] The pre-set deep learning module learns and predicts the bill data for the following month based on the abnormal bill data information, and obtains the predicted value information of the bill data for the following month.

[0159] Get next month's billing data;

[0160] The difference between the bill data for the following month and the predicted bill data for the following month is obtained by analyzing and calculating the bill data for the following month.

[0161] Determine if the difference in the next month's bill data exceeds a preset difference threshold. If so, trigger the prediction result adjustment information and feed the prediction result back to the preset deep learning module to correct and adjust the bill data.

[0162] It should be noted that, through a preset manual review module, abnormal data information after review by professional financial personnel is obtained; the reasons for the abnormal bill data and the handling results of the review are fed back to a preset deep learning module for backup learning and analysis, and the bill data for the following month is predicted to obtain the predicted value information F of the bill data for the following month; when the settlement time of the following month arrives, the bill data information K of the following month is obtained, and the bill data information K of the following month is analyzed and calculated with the predicted value information F of the bill data for the following month to obtain the difference J of the bill data for the following month, J=|KF|. If the preset difference threshold is set to 2000, then when J≥2000, it indicates that there is a large deviation between the predicted settlement data and the actual settlement data, and the prediction method needs to be adjusted. The bill data prediction value information is fed back to the preset deep learning module to find the cause of the deviation, and the prediction methods used for the matching data and the mismatched data of the following month are saved, and the prediction methods used for the mismatched data are adjusted and corrected.

[0163] Figure 3 A block diagram of a billing settlement system according to the present invention is shown.

[0164] like Figure 3 As shown, a second aspect of the present invention provides a billing system 3, including a memory 31 and a processor 32. The memory stores a billing method program, which, when executed by the processor, performs the following steps:

[0165] Based on the preset image acquisition module, the bill data information is collected into the preset image processing module;

[0166] The preset image processing module preprocesses the bill data to obtain refined bill data.

[0167] By extracting key information from the refined data of the billing data, the pre-settlement information of the billing data can be obtained;

[0168] Obtain original billing and settlement information;

[0169] The pre-settlement information of the billing data is analyzed and compared with the original billing settlement information to obtain the billing settlement difference;

[0170] Determine whether the bill settlement difference is equal to the preset difference threshold. If so, send the pre-settlement information of the bill data to the preset bill data settlement module for settlement and upload it to the preset data storage module.

[0171] If not, trigger the bill reconciliation information to recalculate the pre-settlement information of the bill data.

[0172] It should be noted that the billing data is collected by a preset image acquisition module, such as a scanner, camera, mobile phone, or PC. The billing data is then transmitted in image format to a preset image processing module for processing to obtain refined billing data. Key information is then extracted using intelligent recognition technology to obtain the pre-settlement information x. The original billing data y is obtained, and the pre-settlement information x obtained through intelligent recognition is compared with the original billing data y to obtain the bill settlement difference information z, z = |xy|. If the preset difference threshold is set to 0, then when z = 0, it indicates that the pre-settlement information obtained through intelligent recognition is consistent with the original billing data, and the pre-settlement information obtained through intelligent recognition is sent to the preset billing settlement module for bill settlement. When z ≠ 0, bill recalculation is triggered, indicating that there is a discrepancy between the pre-settlement information obtained through intelligent recognition, and the pre-settlement information needs to be recalculated.

[0173] According to an embodiment of the present invention, it further includes:

[0174] Time information for obtaining billing data;

[0175] Based on the time information of the bill data, the bill data is sorted according to a preset time rule to obtain bill data information arranged in chronological order;

[0176] Based on preset time period nodes, the bill data information arranged in chronological order is classified and stored to obtain the bill data information within each time period node.

[0177] The bill data is then sent to a preset image acquisition module for data processing.

[0178] It should be noted that the time information for obtaining bill data generally refers to the time when the bill occurred. Based on this time information, the bill data is sorted and then categorized according to a preset time rule, namely, by the start and end times of the natural month. For example, the start time is 00:00 on the 1st of each month, and the end time is 24:00 on the 30th of the month (if the month has a 31st, then the end time is 24:00 on the 31st). This yields bill data categorized by month, and the monthly bill data is processed on a monthly cycle. The preset time cycle is set by professionals in the field based on actual conditions such as the volume of bill settlement tasks.

[0179] According to an embodiment of the present invention, the step of the preset image processing module preprocessing bill data information to obtain refined bill data information specifically includes:

[0180] Retrieve image information from billing data;

[0181] Based on the image information of the bill data, obtain the pixel information of the bill image data;

[0182] Determine whether the pixel information of the bill image data is greater than the preset pixel threshold. If so, set the pixel information of the bill image data that is greater than the preset pixel threshold to the preset first fixed pixel value.

[0183] If not, set the pixel information of bill image data that is less than the preset pixel threshold to the preset second fixed pixel value;

[0184] The determined bill image data is processed and saved to obtain refined bill data.

[0185] It should be noted that after the preset image processing module acquires the image information of the bill data, it obtains the pixel information α of the image based on the image information of the bill data. The preset pixel threshold is set to 120. Therefore, when the pixel information α of the bill image data is ≥ 120, the pixels of the bill image data with a pixel value greater than 120 are set to the preset first fixed pixel value, i.e., 255; when α < 120, the pixels of the bill image data with a pixel value less than 120 are set to the preset second fixed pixel value, i.e., 0. Through the above operation, the image information of the bill data is converted into a black and white binary image, thereby simplifying the bill image data to facilitate subsequent image processing and analysis. The preset pixel threshold is set by those skilled in the art according to the actual situation.

[0186] According to an embodiment of the present invention, the step of extracting key information from the refined information of the bill data to obtain the pre-settlement information of the bill data specifically includes:

[0187] Send the refined information of the bill data to the preset bill data intelligent analysis module;

[0188] Based on preset key information rules, set key information for billing data;

[0189] The pre-set bill data intelligent analysis module compares and analyzes key information in the refined information of the bill data to obtain the number of associations of key information.

[0190] Determine if the number of associations of key information exceeds the preset association threshold. If so, mark the refined information of the corresponding bill data to obtain the pre-settlement information of the bill data.

[0191] If not, trigger the manual review information and send the corresponding bill data refinement information to the preset manual review module for manual review and judgment.

[0192] It should be noted that the pre-set bill data intelligent analysis module is an intelligent module with functions such as data analysis, data processing, and deep learning. After receiving the refined information of the bill data, it sets keyword information related to bill settlement based on preset key information rules. The preset key information rules mainly include setting the time format to extract the bill settlement time, setting the amount value format to extract the bill settlement amount, and setting the company or unit name to extract the bill's counterparty information, etc. The pre-set bill data intelligent analysis module extracts key information based on the refined information of the bill data and records the number of times the key information is triggered, obtaining the association number N of the key information. The preset association number threshold is set to 2. When the association number N of the key information is ≥ 2, it means that there is a lot of keyword information triggered that is related to bill settlement, and it can be used as the pre-settlement information of the bill data. When the association number N of the key information is < 2, it triggers the manual review information, and the corresponding refined information of the bill data needs to be sent to the preset manual review module for manual review and judgment.

[0193] According to an embodiment of the present invention, it further includes:

[0194] Obtain the original billing documents;

[0195] By statistically analyzing each original invoice voucher, the original invoice data information is obtained;

[0196] The original billing data is reviewed to obtain the review results.

[0197] Determine whether the audit result information meets the preset audit threshold. If so, send the original bill data information to the preset bill data settlement module and wait for further bill settlement.

[0198] If not, the corresponding original bill will be marked as invalid and will not be settled.

[0199] It should be noted that in order to filter out original bill data that meets the bill settlement requirements, the original bill data needs to be reviewed; original bill vouchers are obtained, and each bill voucher is statistically calculated and reviewed, and the original bill data information is reviewed to obtain the review result; only if it meets the preset review threshold, that is, the review is passed, can the original bill data information be sent to the preset bill settlement module for settlement; if the review result is not passed, it means that the original bill does not meet the bill settlement requirements and is marked as an invalid bill.

[0200] According to an embodiment of the present invention, the step of triggering bill reconciliation information and recalculating the pre-settlement information of the bill data specifically includes:

[0201] Send the pre-settlement information of the billing data to the preset manual review module;

[0202] The preset manual review module reviews the pre-settlement information and obtains the review results.

[0203] If the review results are abnormal, the corresponding abnormal settlement information will be corrected.

[0204] If the review result is normal, the original billing data will be reviewed and entered to obtain the corrected original billing data.

[0205] It should be noted that the pre-settlement information of the bill data processed by the bill data intelligent analysis module is sent to the preset manual review module. The preset manual review module reviews the pre-settlement information of the bill data. The review mainly confirms the compliance and accuracy of the pre-settlement information of the bill data and obtains the review result. If the review result is abnormal, the corresponding pre-settlement information needs to be corrected. If the review result is normal, it means that the pre-settlement information of the bill data meets the requirements of bill settlement. Therefore, the original bill data is then reviewed and entered into the system to obtain the reviewed original bill data.

[0206] According to an embodiment of the present invention, it further includes:

[0207] Retrieve information on the number of times bill reconciliation information has been triggered;

[0208] Determine if the number of times the bill reconciliation information is triggered exceeds the preset threshold. If not, proceed with the bill settlement according to the normal process.

[0209] If so, a bill settlement warning will be triggered, and intelligent recognition of the current bill data will be paused. The preset manual review module will be activated to settle the bill data.

[0210] It should be noted that when the bill settlement difference z≠0, bill reconciliation information will be triggered, and the number of times the bill reconciliation information is triggered, T, will be obtained. The preset number threshold is set to 2. When T<2, the bill settlement steps will continue to be executed according to the original process. If T≥2, a bill settlement warning will be triggered, indicating that the bill data currently identified by the intelligent system has errors and the number of errors is too high. The intelligent system needs to be paused and the preset manual review module will be started to settle the bill data.

[0211] According to an embodiment of the present invention, it further includes:

[0212] Retrieve historical billing data;

[0213] Historical billing data is divided and calculated according to preset time period nodes to obtain billing data for each time period;

[0214] The average billing data within the preset time period is calculated based on the billing data information of each time period.

[0215] Retrieve bill settlement data information for the current time period;

[0216] By calculating and analyzing the average bill data within a preset time period and the bill settlement data at the current time period, the bill settlement data deviation value is obtained.

[0217] Determine if the deviation value of the bill settlement data is greater than the preset deviation threshold. If so, trigger the bill settlement anomaly information and send the bill data for the current month to the preset bill data intelligent analysis module for analysis.

[0218] It should be noted that historical billing data is obtained and divided according to preset time period nodes, such as dividing historical billing data into monthly billing data according to the natural month time period; the average monthly billing settlement data P is calculated; the difference between the average monthly billing settlement data P and the current month's billing settlement data Q is calculated to obtain the billing settlement data deviation value L, L=|QP|. The preset deviation threshold is set to 5000. When L≥5000, a billing settlement anomaly information is triggered, indicating that the current month's billing settlement data exceeds or falls significantly below the historical monthly billing average settlement data, indicating a significant anomaly. The current month's billing data needs to be sent to the preset billing data intelligent analysis module for analysis to find the reason for the excess or below-average level in the current month. The preset time period nodes, preset difference thresholds, etc., are all set by those skilled in the art according to the actual situation.

[0219] According to an embodiment of the present invention, the step of triggering abnormal bill settlement information and sending the monthly bill data to a preset intelligent bill data analysis module for analysis specifically includes:

[0220] Retrieve billing and settlement information;

[0221] The pre-set bill data intelligent analysis module extracts and constructs features from the target bill data based on the bill settlement association information to obtain the bill settlement feature information;

[0222] The pre-set bill data intelligent analysis module sends the characteristic information of bill settlement to the pre-set deep learning module for learning and analysis to obtain abnormal bill data information;

[0223] The abnormal billing data is sent to the preset manual review module for review, and the review result information is obtained.

[0224] The audit results are sent to a pre-defined deep learning module for secondary learning and data analysis.

[0225] The characteristic information of the bill settlement includes the time information, amount information, settlement object information, and settlement reason information of the bill settlement.

[0226] It should be noted that bill-related information mainly refers to information related to bill settlement, involving monetary figures, time formats, and so on. The pre-set intelligent bill data analysis module performs detailed feature extraction on the target bill data based on the bill settlement related information, and rearranges the bill-related information according to settlement logic and rules to construct bill settlement feature information with settlement logic. The pre-set intelligent bill data analysis module sends the bill settlement feature information to the pre-set deep learning module for learning and analysis, removing duplicate bill data and invalid bill data to obtain abnormal bill data information, such as large income and expenditure information, income and expenditure information in abnormal time periods, and abnormal settlement data between counterparties. The abnormal bill data is then sent to the pre-set manual review module for review by professional financial personnel to obtain corresponding review results. The review results are then fed back to the pre-set deep learning module to back up, learn, and analyze the reasons for the abnormality of the abnormal bill data and the handling results of the review, and whether it possesses the corresponding characteristics, providing data support and assistance for future settlement decisions.

[0227] According to an embodiment of the present invention, the step of sending the review result information to a preset deep learning module for secondary learning and data analysis specifically includes:

[0228] Based on the preset manual review module, obtain abnormal billing data information;

[0229] The pre-set deep learning module learns and predicts the bill data for the following month based on the abnormal bill data information, and obtains the predicted value information of the bill data for the following month.

[0230] Get next month's billing data;

[0231] The difference between the bill data for the following month and the predicted bill data for the following month is obtained by analyzing and calculating the bill data for the following month.

[0232] Determine if the difference in the next month's bill data exceeds a preset difference threshold. If so, trigger the prediction result adjustment information and feed the prediction result back to the preset deep learning module to correct and adjust the bill data.

[0233] It should be noted that, through a preset manual review module, abnormal data information after review by professional financial personnel is obtained; the reasons for the abnormal bill data and the handling results of the review are fed back to a preset deep learning module for backup learning and analysis, and the bill data for the following month is predicted to obtain the predicted value information F of the bill data for the following month; when the settlement time of the following month arrives, the bill data information K of the following month is obtained, and the bill data information K of the following month is analyzed and calculated with the predicted value information F of the bill data for the following month to obtain the difference J of the bill data for the following month, J=|KF|. If the preset difference threshold is set to 2000, then when J≥2000, it indicates that there is a large deviation between the predicted settlement data and the actual settlement data, and the prediction method needs to be adjusted. The bill data prediction value information is fed back to the preset deep learning module to find the cause of the deviation, and the prediction methods used for the matching data and the mismatched data of the following month are saved, and the prediction methods used for the mismatched data are adjusted and corrected.

[0234] This invention discloses a bill settlement method and system. The method includes: collecting bill data information and preprocessing it in a preset image processing module to obtain refined bill data information; extracting key information from the refined bill data information to obtain pre-settlement information; analyzing and comparing the pre-settlement information with the original bill settlement information to obtain a bill settlement difference; determining whether the bill settlement difference equals a preset difference threshold; if so, sending the pre-settlement information to a preset bill data settlement module for settlement and uploading it to a preset data storage module; if not, triggering bill verification information to recalculate the pre-settlement information. This invention improves the efficiency and accuracy of bill settlement by analyzing and using deep learning on the intelligently recognized bill data information and the original bill data information.

[0235] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods 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, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0236] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0237] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0238] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0239] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, 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 methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A bill settlement method, characterized in that, include: Based on the preset image acquisition module, the bill data information is collected into the preset image processing module; The preset image processing module preprocesses the bill data to obtain refined bill data. By extracting key information from the refined data of the billing data, the pre-settlement information of the billing data can be obtained; Obtain original billing and settlement information; The pre-settlement information of the billing data is analyzed and compared with the original billing settlement information to obtain the billing settlement difference; Determine whether the bill settlement difference is equal to the preset difference threshold. If so, send the pre-settlement information of the bill data to the preset bill data settlement module for settlement and upload it to the preset data storage module. If not, trigger the bill reconciliation information to recalculate the pre-settlement information of the bill data.

2. The bill settlement method according to claim 1, characterized in that, Also includes: Time information for obtaining billing data; Based on the time information of the bill data, the bill data is sorted according to a preset time rule to obtain bill data information arranged in chronological order; Based on preset time period nodes, the bill data information arranged in chronological order is classified and stored to obtain the bill data information within each time period node. The bill data is then sent to a preset image acquisition module for data processing.

3. The bill settlement method according to claim 1, characterized in that, The step of the preset image processing module preprocessing the bill data to obtain refined bill data specifically includes: Retrieve image information from billing data; Based on the image information of the bill data, obtain the pixel information of the bill image data; Determine whether the pixel information of the bill image data is greater than the preset pixel threshold. If so, set the pixel information of the bill image data that is greater than the preset pixel threshold to the preset first fixed pixel value. If not, set the pixel information of bill image data that is less than the preset pixel threshold to the preset second fixed pixel value; The determined bill image data is processed and saved to obtain refined bill data.

4. The bill settlement method according to claim 1, characterized in that, The step of extracting key information from the refined information of the bill data to obtain the pre-settlement information of the bill data specifically includes: Send the refined information of the bill data to the preset bill data intelligent analysis module; Based on preset key information rules, set key information for billing data; The pre-set bill data intelligent analysis module compares and analyzes key information in the refined information of the bill data to obtain the number of associations of key information. Determine if the number of associations of key information exceeds the preset association threshold. If so, mark the refined information of the corresponding bill data to obtain the pre-settlement information of the bill data. If not, trigger the manual review information and send the corresponding bill data refinement information to the preset manual review module for manual review and judgment.

5. A bill settlement method according to claim 1, characterized in that, Also includes: Obtain the original billing documents; By statistically analyzing each original invoice voucher, the original invoice data information is obtained; The original billing data is reviewed to obtain the review results. Determine whether the audit result information meets the preset audit threshold. If so, send the original bill data information to the preset bill data settlement module and wait for further bill settlement. If not, the corresponding original bill will be marked as invalid and will not be settled.

6. A bill settlement method according to claim 1, characterized in that, The step of triggering bill reconciliation information and recalculating the pre-settlement information of the bill data specifically includes: Send the pre-settlement information of the billing data to the preset manual review module; The preset manual review module reviews the pre-settlement information and obtains the review results. If the review results are abnormal, the corresponding abnormal settlement information will be corrected. If the review result is normal, the original billing data will be reviewed and entered to obtain the corrected original billing data.

7. A bill settlement method according to claim 1, characterized in that, Also includes: Retrieve information on the number of times bill reconciliation information has been triggered; Determine if the number of times the bill reconciliation information is triggered exceeds the preset threshold. If not, proceed with the bill settlement according to the normal process. If so, a bill settlement warning will be triggered, and intelligent recognition of the current bill data will be paused. The preset manual review module will be activated to settle the bill data.

8. A billing settlement system, characterized in that, The system includes a memory and a processor. The memory stores a billing method program, which, when executed by the processor, performs the following steps: Based on the preset image acquisition module, the bill data information is collected into the preset image processing module; The preset image processing module preprocesses the bill data to obtain refined bill data. By extracting key information from the refined data of the billing data, the pre-settlement information of the billing data can be obtained; Obtain original billing and settlement information; The pre-settlement information of the billing data is analyzed and compared with the original billing settlement information to obtain the billing settlement difference; Determine whether the bill settlement difference is equal to the preset difference threshold. If so, send the pre-settlement information of the bill data to the preset bill data settlement module for settlement and upload it to the preset data storage module. If not, trigger the bill reconciliation information to recalculate the pre-settlement information of the bill data.

9. A billing settlement system according to claim 8, characterized in that, Also includes: Time information for obtaining billing data; Based on the time information of the bill data, the bill data is sorted according to a preset time rule to obtain bill data information arranged in chronological order; Based on preset time period nodes, the bill data information arranged in chronological order is classified and stored to obtain the bill data information within each time period node. The bill data is then sent to a preset image acquisition module for data processing.

10. A billing settlement system according to claim 8, characterized in that, The step of the preset image processing module preprocessing the bill data to obtain refined bill data specifically includes: Retrieve image information from billing data; Based on the image information of the bill data, obtain the pixel information of the bill image data; Determine whether the pixel information of the bill image data is greater than the preset pixel threshold. If so, set the pixel information of the bill image data that is greater than the preset pixel threshold to the preset first fixed pixel value. If not, set the pixel information of bill image data that is less than the preset pixel threshold to the preset second fixed pixel value; The determined bill image data is processed and saved to obtain refined bill data.

Images