Return auditing method, system and equipment based on many-to-many matching and medium

Abstract

The invention relates to a goods return auditing method, system and device based on many-to-many matching and a medium, and the method comprises the steps: carrying out the image recognition of an error application form uploaded by an initiator, and obtaining first information; according to the uploaded error application form, the error handling personnel performs manual input to obtain second information; the first information and the second information are matched according to the retrieval reference number field, if multiple pieces of data are matched, many-to-many matching is carried out on the first information and the second information according to the retrieval reference number, the merchant number and the error amount combination field, if the comparison result is consistent, goods return auditing is passed, and otherwise modification is returned. Compared with the prior art, the method has the advantages of high accuracy, high efficiency and the like.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a return audit method, system, device and medium based on many-to-many matching. Background Art

[0002] At present, when processing error application forms for return business in a centralized manner, the following methods are often used:

[0003] Method 1: One person manually enters information based on the error application form, and another person visually checks the consistency between the error application form content and the manually entered information. This method has a high error rate and low business efficiency when faced with a scenario where there are multiple business data items in an error application form.

[0004] Method 2: Centralized data entry by image slicing and information fragmentation. In the centralized data entry process, each fragment needs to be manually entered by two fragment entry personnel. If they are consistent, they pass. If they are inconsistent, a third person needs to arbitrate as an arbitrator. Each fragment entry requires at least two people to manually enter, which has low business efficiency and high labor costs. In addition, image slicing is difficult to complete the slicing process in the scenario of multiple business data for one error application form.

[0005] In response to the above problems, it is urgent to design a return audit solution with high accuracy and efficiency. Summary of the invention

[0006] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a return audit method, system, equipment and medium based on many-to-many matching with high accuracy and efficiency.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] According to a first aspect of the present invention, a return audit method based on many-to-many matching is provided, the method comprising:

[0009] S1. Perform image recognition on the error application form uploaded by the initiator to obtain first information;

[0010] S2. The error handling personnel manually enters the second information according to the uploaded error application form;

[0011] S3, matching the first information and the second information according to the search reference number field, if multiple pieces of data are matched, go to S4, if unique data is matched, go to S5, if no data is matched, go to S6;

[0012] S4, performing a many-to-many match on the first information and the second information according to the search reference number, the merchant number and the error amount combination field: if a unique data is matched, go to S5; if multiple data are matched, randomly select fields for matching, if the random fields match consistently, it is considered that the comparison results are consistent and go to S7; if the random fields do not match consistently, go to S6; if no data is matched, it is considered that the comparison results are inconsistent and go to S6;

[0013] S5, matching the remaining fields. If the remaining fields are matched, the matching results are considered to be consistent, and the process goes to S7; otherwise, the process goes to S6;

[0014] S6. Conduct third-party verification and return the document for modification after confirming the error.

[0015] S7. Return review is approved.

[0016] Preferably, the performing image recognition on the error application form uploaded by the initiator specifically comprises: performing image recognition on the error application form uploaded by the initiator using an OCR recognition model.

[0017] Preferably, the method further comprises: marking the error application form with recognition errors, and optimizing and training the OCR recognition module at each set time or after reaching a set recognition error number threshold.

[0018] Preferably, the image recognition also includes format checking of each field of the first information.

[0019] Preferably, the field information in the error application form includes merchant number, search reference number, original transaction date, original transaction amount, error amount, and sub-order number.

[0020] Preferably, a third party verification is performed in S6, and the information is returned for modification after the erroneous party is confirmed. Specifically, if the comparison result is inconsistent, it is sent to a third party for review and verification, and the details of the inconsistent fields are displayed. The third party verifies the inconsistent fields with reference to the error application form, and returns the information for modification after confirming that the information is entered in an erroneous party.

[0021] Preferably, if multiple pieces of data are matched, fields are randomly selected for matching. If the random fields match consistently, the comparison results are considered consistent and the process goes to S7. If the random fields do not match consistently, the process goes to S6. Specifically,

[0022] If multiple data are matched, fields are randomly selected for matching. When some fields in the randomly selected fields are inconsistent or no data is matched, it is determined that the random field matches are inconsistent and the process goes to S6. Otherwise, the process goes to S7.

[0023] According to a second aspect of the present invention, a return audit system based on many-to-many matching is provided, the system comprising:

[0024] The initiation upload module is used by the initiator to initiate the return review process and upload the error application form attachments;

[0025] An identification module, used for performing image recognition on the error application form uploaded by the initiator to obtain first information;

[0026] An input module is used for the error handling personnel to manually input the second information according to the uploaded error application form;

[0027] A first matching module is used to match the first information and the second information according to the search reference number field, and if multiple pieces of data are matched, it is transferred to the second matching module, if only one piece of data is matched, it is transferred to the third matching module, and if no data is matched, it is transferred to the return and modification module;

[0028] The second matching module is used to perform a many-to-many match between the first information and the second information according to the search reference number, the merchant number and the error amount combination field: if a unique data is matched, it is transferred to the third matching module; if multiple data are matched, the fields are randomly selected for matching, and if the random fields match consistently, the comparison result is considered consistent and transferred to the output module; if the random fields do not match consistently, it is transferred to the return and modification module; if no data is matched, the comparison result is considered inconsistent and transferred to the return and modification module;

[0029] The third matching module is used to match the remaining fields. If the remaining fields are matched, the matching results are considered to be consistent and transferred to the output module; otherwise, the matching results are transferred to the return modification module;

[0030] The return and modification module is used to send the comparison results to a third party for review and proofreading when they are inconsistent, and display the details of the inconsistent fields. The third party will refer to the error application form to check the inconsistent fields, and return the modification after confirming that the information is entered incorrectly;

[0031] Output module, used to output the return review results.

[0032] According to a third aspect of the present invention, there is provided an electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the program, any one of the methods described above is implemented.

[0033] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, any one of the methods described above is implemented.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] (1) Use key fields for many-to-many matching, establish a matching relationship between image recognition information and manually entered information, and introduce a third party for verification when the comparison is inconsistent, effectively ensuring the accuracy of business information and avoiding erroneous returns that have adverse effects on merchants and companies.

[0036] (2) Combining image recognition and manual data entry and cross-checking can effectively avoid image recognition errors or manual data entry errors, and further improve the accuracy of information collection and business review.

[0037] (3) An OCR recognition model is used to perform image recognition on the error application form and extract the field information in the error application form, further improving the applicable scenarios and collection efficiency of information collection. In addition, the error application form with recognition errors is marked, and the OCR recognition module is optimized and trained after each set time or when the set recognition error number threshold is reached, thereby specifically improving the recognition accuracy of the OCR recognition model.

[0038] (4) Perform format verification on the fields recognized by OCR according to business requirements to further improve the accuracy and usability of the information recognized by OCR. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a process diagram of return audit of the present invention;

[0040] Figure 2 A detailed flow chart of the method of the present invention;

[0041] Figure 3 A schematic diagram of the manual input interface for the handling personnel;

[0042] Figure 4 This is a schematic diagram of the page with inconsistent comparison results;

[0043] Figure 5 This is a schematic diagram of the error return page;

[0044] Figure 6 This is a schematic diagram of the return approval page. DETAILED DESCRIPTION

[0045] 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 part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0046] Example

[0047] like Figure 1 and Figure 2 As shown, this embodiment provides a return review method based on many-to-many matching, the method comprising:

[0048] S1. Perform image recognition on the error application form uploaded by the initiator to obtain first information; wherein the field information in the error application form includes merchant number, search reference number, original transaction date, original transaction amount, error amount, and sub-order number;

[0049] S2. The error handling personnel manually enters the second information according to the uploaded error application form;

[0050] S3, matching the first information and the second information according to the search reference number field, if multiple pieces of data are matched, go to S4, if unique data is matched, go to S5, if no data is matched, go to S6;

[0051] S4, performing a many-to-many match on the first information and the second information according to the search reference number, the merchant number and the error amount combination field: if a unique data is matched, go to S5; if multiple data are matched, randomly select fields for matching, if the random fields match consistently, it is considered that the comparison results are consistent and go to S7; if the random fields do not match consistently, go to S6; if no data is matched, it is considered that the comparison results are inconsistent and go to S6;

[0052] S5, matching the remaining fields. If the remaining fields are matched, the matching results are considered to be consistent, and the process goes to S7; otherwise, the process goes to S6;

[0053] S6. Conduct third-party verification and return for modification after confirming the erroneous party: If the comparison results are inconsistent, they will be sent to a third party for review and verification, and the details of the inconsistent fields will be displayed. The third party will refer to the error application form to verify the inconsistent fields. After confirming that the information has been entered in an erroneous party, it will be returned for modification.

[0054] S7. Return review is approved.

[0055] As another preferred embodiment, an OCR recognition model is used to perform image recognition on the error application form uploaded by the initiator. The OCR recognition model supports multiple application form templates, as well as multiple formats of files such as jpg, png, jpeg, and pdf. OCR recognition supports submitting multiple files of different formats in a single process, as well as supporting scenarios with multiple pages in a single file, which can greatly improve the applicable scenarios and collection efficiency of information collection.

[0056] As another preferred embodiment, the error application form with recognition error is marked, and after each set time or after reaching a set recognition error number threshold, the OCR recognition module is optimized and trained to further improve the accuracy of OCR recognition.

[0057] As another preferred embodiment, when OCR image recognition is performed, the format of each field of the first information is also checked. For example, the length and format of key elements such as date, amount, and business code are restricted according to business requirements, and a fault tolerance mechanism is added to filter some invalid information in key fields (for example, yyyy-mm-month-dd is output as yyyymmdd after recognition), so as to further improve the accuracy and usability of the information after OCR recognition.

[0058] If multiple data are matched, fields are randomly selected for matching. If the random fields match consistently, the comparison results are considered consistent and the process goes to S7. If the random fields do not match consistently, the process goes to S6. Specifically,

[0059] If multiple data are matched, fields are randomly selected for matching. When some fields in the randomly selected fields are inconsistent or no data is matched, it is determined that the random field matches are inconsistent and the process goes to S6. Otherwise, the process goes to S7.

[0060] This embodiment also provides a return review system based on many-to-many matching, including:

[0061] The initiation upload module is used by the initiator to initiate the return review process and upload the error application form attachments;

[0062] An identification module, used for performing image recognition on the error application form uploaded by the initiator to obtain first information;

[0063] An input module is used for the error handling personnel to manually input the second information according to the uploaded error application form;

[0064] A first matching module is used to match the first information and the second information according to the search reference number field, and if multiple pieces of data are matched, it is transferred to the second matching module, if only one piece of data is matched, it is transferred to the third matching module, and if no data is matched, it is transferred to the return and modification module;

[0065] The second matching module is used to perform a many-to-many match between the first information and the second information according to the search reference number, the merchant number and the error amount combination field: if a unique data is matched, it is transferred to the third matching module; if multiple data are matched, the fields are randomly selected for matching, and if the random fields match consistently, the comparison result is considered consistent and transferred to the output module; if the random fields do not match consistently, it is transferred to the return and modification module; if no data is matched, the comparison result is considered inconsistent and transferred to the return and modification module;

[0066] The third matching module is used to match the remaining fields. If the remaining fields are matched, the matching results are considered to be consistent and transferred to the output module; otherwise, the matching results are transferred to the return modification module;

[0067] The return and modification module is used to send the comparison results to a third party for review and proofreading when they are inconsistent, and display the details of the inconsistent fields. The third party will refer to the error application form to check the inconsistent fields, and return the modification after confirming that the information is entered incorrectly;

[0068] Output module, used to output the return review results.

[0069] The specific implementation process is as follows:

[0070] 101. The OCR recognition model is pre-established and trained based on the error application form template. The OCR recognizes the merchant number, search reference number, original transaction date, original transaction amount, error amount, sub-order number and other information based on the error application form content. At the same time, the OCR limits the length and format of the date, amount and merchant code according to business requirements.

[0071] 102. The initiator initiates the return process and uploads the error application form attachment during the process. The system performs OCR recognition on the error application form content and converts it into text information to display in the process and record it in the background.

[0072] 103. After the process is initiated, it enters the error handling review stage. The system shields the OCR recognition information of the error handling personnel to ensure the independence and objectivity of the manually entered data. Figure 3 shown.

[0073] 104. The error handling personnel shall complete the manual entry of information such as merchant number, search reference number, original transaction date, original transaction amount, error amount, sub-order number, etc. based on the contents of the error application form, display it in the process and record it in the background.

[0074] 105. The system establishes a correspondence between the initiator's OCR identification information and the manually entered information. The "search reference number" is used as the first factor for matching, and the corresponding relationship is established after the matching is successful; if there are multiple data with the same search reference number, the system will perform many-to-many matching according to "search reference number + merchant number + error amount" to ensure that the final data corresponds one to one. Among them, for invalid details with the status of "revoked" or "deleted", the system filters them during comparison to improve the efficiency of the system comparison.

[0075] 106. The system compares the consistency of OCR recognition information and manually entered information one by one for merchant number, search reference number, original transaction date, original transaction amount, error amount, sub-order number and other information. If all fields are consistent, the system marks the comparison result as "consistent"; if there is an inconsistency in the field comparison, the system marks the comparison result as "inconsistent".

[0076] 107. For processes with inconsistent comparison results, the system assigns a third person as a reviewer to proofread. The system displays the details of the specific fields that are inconsistent with the comparison. The error reviewer refers to the error application form to proofread the inconsistent fields between the two parties. Figure 4 shown.

[0077] 108. The error reviewer confirms that the party who entered the information incorrectly will return the process to the corresponding person for modification. Figure 5 shown.

[0078] 109. When the comparison results of OCR recognition information and manually entered information are consistent, the system allows the process to take effect. Figure 6 shown.

[0079] The electronic device of the present invention includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) or loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for device operation can also be stored. The CPU, ROM and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0080] Multiple components in the device are connected to the I / O interface, including: input units, such as keyboards, mice, etc.; output units, such as various types of displays, speakers, etc.; storage units, such as disks, optical disks, etc.; and communication units, such as network cards, modems, wireless communication transceivers, etc. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunication networks.

[0081] The processing unit performs the various methods and processes described above, such as methods S1 to S7. For example, in some embodiments, methods S1 to S7 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via a ROM and / or a communication unit. When the computer program is loaded into the RAM and executed by the CPU, one or more steps of methods S1 to S7 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S7 in any other appropriate manner (e.g., by means of firmware).

[0082] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip systems (SOCs), load programmable logic devices (CPLDs), and the like.

[0083] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0084] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0085] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A return audit method based on many-to-many matching, characterized in that: The method includes: S1. Perform image recognition on the error application form uploaded by the initiator to obtain first information; S2. The error handling personnel manually enters the second information according to the uploaded error application form; S3, matching the first information and the second information according to the search reference number field, if multiple pieces of data are matched, go to S4, if unique data is matched, go to S5, if no data is matched, go to S6; S4, performing a many-to-many match on the first information and the second information according to the search reference number, the merchant number and the error amount combination field: if a unique data is matched, go to S5; if multiple data are matched, randomly select fields for matching, if the random fields match consistently, it is considered that the comparison results are consistent and go to S7; if the random fields do not match consistently, go to S6; if no data is matched, it is considered that the comparison results are inconsistent and go to S6; S5, matching the remaining fields. If the remaining fields are matched, the matching results are considered to be consistent, and the process goes to S7; otherwise, the process goes to S6; S6. Conduct third-party verification and return the document for modification after confirming the error. S7. Return review is approved.

2. A return audit method based on many-to-many matching according to claim 1, characterized in that: The performing image recognition on the error application form uploaded by the initiator specifically includes: performing image recognition on the error application form uploaded by the initiator using an OCR recognition model.

3. A return audit method based on many-to-many matching according to claim 2, characterized in that: The method further includes: marking the error application form with recognition errors, and optimizing and training the OCR recognition module at each set time or after reaching a set recognition error number threshold.

4. A return audit method based on many-to-many matching according to claim 1, characterized in that: The image recognition also includes format checking of each field of the first information.

5. A return audit method based on many-to-many matching according to claim 1, characterized in that: The field information in the error application form includes the merchant number, the search reference number, the original transaction date, the original transaction amount, the error amount, and the sub-order number.

6. A return audit method based on many-to-many matching according to claim 1, characterized in that: In S6, a third party verifies and confirms the party with the error, and then returns the information for modification. Specifically, if the comparison results are inconsistent, the information is sent to a third party for review and verification, and the details of the inconsistent fields are displayed. The third party verifies the inconsistent fields with reference to the error application form, and returns the information for modification after confirming that the party with the error has entered the information.

7. A return audit method based on many-to-many matching according to claim 1, characterized in that: If multiple data are matched, fields are randomly selected for matching. If the random fields match consistently, the comparison results are considered consistent and the process goes to S7. If the random fields do not match consistently, the process goes to S6. Specifically, If multiple data are matched, fields are randomly selected for matching. When some fields in the randomly selected fields are inconsistent or no data is matched, it is determined that the random field matches are inconsistent and the process goes to S6. Otherwise, the process goes to S7.

8. A return audit system based on many-to-many matching, characterized in that: The system comprises: The initiation upload module is used by the initiator to initiate the return review process and upload the error application form attachments; An identification module, used for performing image recognition on the error application form uploaded by the initiator to obtain first information; An input module is used for the error handling personnel to manually input the second information according to the uploaded error application form; A first matching module is used to match the first information and the second information according to the search reference number field, and if multiple pieces of data are matched, it is transferred to the second matching module, if only one piece of data is matched, it is transferred to the third matching module, and if no data is matched, it is transferred to the return and modification module; The second matching module is used to perform a many-to-many match between the first information and the second information according to the search reference number, the merchant number and the error amount combination field: if a unique data is matched, it is transferred to the third matching module; if multiple data are matched, the fields are randomly selected for matching, and if the random fields match consistently, the comparison result is considered consistent and transferred to the output module; if the random fields do not match consistently, it is transferred to the return and modification module; if no data is matched, the comparison result is considered inconsistent and transferred to the return and modification module; The third matching module is used to match the remaining fields. If the remaining fields are matched, the matching results are considered to be consistent and transferred to the output module; otherwise, the matching results are transferred to the return modification module; The return and modification module is used to send the comparison results to a third party for review and proofreading when they are inconsistent, and display the details of the inconsistent fields. The third party will refer to the error application form to check the inconsistent fields, and return the modification after confirming that the information is entered incorrectly; Output module, used to output the return review results.

9. An electronic device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images