Data processing method and device, electronic equipment and computer readable medium

Abstract

The invention discloses a data processing method and device, electronic equipment and a computer readable medium, and relates to the technical field of big data processing.The specific implementation mode comprises the steps that in response to a reimbursement request, reimbursement application data is obtained; if the display form of the reimbursement application data is an image, identifying partition edges of the reimbursement application data, partitioning the reimbursement application data based on the partition edges to obtain partition data, identifying key fields in the partition data, and classifying the reimbursement application data according to the key fields to obtain categories; marking the reimbursement application data according to each category to obtain marked data; generating a result tree based on the mark data; and according to the result tree and the pre-recorded business data, executing a verification process, and in response to verification failure, generating an offline labeling task and sending the offline labeling task to an offline labeling node. Therefore, the data processing efficiency and accuracy during reimbursement application processing are improved, and the labor cost is reduced.

Description

Technical Field

[0001] The present application relates to the field of big data processing technology, and in particular to a data processing method, device, electronic device and computer-readable medium. Background Art

[0002] At present, there are a large number of budget institutions in financial service enterprises. In the enterprise financial expenditure management, various types of expenditure voucher data are involved in reimbursement approval processing and voucher management. In order to establish an efficient and convenient processing mechanism, it is necessary to extract the information of paper or electronic expenditure vouchers into electronic data to facilitate the subsequent reimbursement process and voucher verification. There is a large amount of expenditure voucher data in a single reimbursement application. It is very cumbersome to manually extract and enter it into electronic data. There is a lot of information content that needs to be entered. At the same time, the continuous identification and entry of number information with a large number of digits is prone to manual identification and entry errors, and the accuracy cannot be guaranteed. It requires multiple audits and verifications to see if it is correct. It relies heavily on a large investment in labor costs. Manually entering various types of expenditure voucher data into electronic data is inefficient, costly, and has low accuracy. Summary of the invention

[0003] In view of this, the embodiments of the present application provide a data processing method, device, electronic device and computer-readable medium, which can solve the problem that there is a large amount of expense voucher data in the existing single reimbursement application, and the manual entry of various expense voucher data into electronic data is inefficient, costly and low in accuracy.

[0004] To achieve the above-mentioned purpose, according to one aspect of an embodiment of the present application, a data processing method is provided, including: in response to a reimbursement request, obtaining reimbursement application data; if the reimbursement application data is displayed in the form of an image, identifying the partition edges of the reimbursement application data, partitioning the reimbursement application data based on the partition edges to obtain each partition data, identifying the key fields in each partition data, and classifying the reimbursement application data according to the key fields to obtain each category; marking the reimbursement application data according to each category to obtain marked data; generating a result tree based on the marked data; executing a verification process according to the result tree and pre-recorded business data, and generating an offline labeling task in response to a verification failure and sending it to an offline labeling node.

[0005] Optionally, the reimbursement application data is classified according to key fields to obtain various categories, including: similarity matching the key fields with preset category descriptions to determine various categories corresponding to the reimbursement application data based on the matched category descriptions.

[0006] Optionally, the reimbursement application data is marked according to each category to obtain marked data, including: for each business application, the reimbursement application data of the category of reimbursement form is marked as the main node, and the reimbursement application data of the category of bill is marked as the child node, so as to obtain marked data based on the main node and the child node.

[0007] Optionally, generating a result tree includes: associating each child node in the tag data with the main node to generate the result tree.

[0008] Optionally, executing a verification process includes: summarizing the number of master nodes of each result tree, and then comparing the number of master nodes with a preset number of business applications, and determining that the verification has failed if the comparison is inconsistent.

[0009] Optionally, executing the verification process includes: for each business application, summarizing the number of child nodes in the corresponding result tree, and then comparing the number of child nodes with the number of corresponding preset bills. If the comparison is inconsistent, it is determined that the verification has failed.

[0010] In addition, the present application also provides a data processing device, including: an acquisition unit, configured to acquire reimbursement application data in response to a reimbursement request; a classification unit, configured to identify the partition edges of the reimbursement application data if the reimbursement application data is displayed in the form of an image, partition the reimbursement application data based on the partition edges to obtain each partition data, identify the key fields in each partition data, and classify the reimbursement application data according to the key fields to obtain each category; a marking unit, configured to mark the reimbursement application data according to each category to obtain marked data; a result tree generation unit, configured to generate a result tree based on the marked data; a verification unit, configured to execute a verification process based on the result tree and pre-recorded business data, and in response to a verification failure, generate an offline annotation task and send it to an offline annotation node.

[0011] Optionally, the classification unit is further configured to: perform similarity matching between the key fields and preset category descriptions, so as to determine the categories corresponding to the reimbursement application data based on the matched category descriptions.

[0012] Optionally, the marking unit is further configured to: for each business application, mark the reimbursement application data of the reimbursement form category as a main node, and mark the reimbursement application data of the bill category as a sub-node, so as to obtain marked data based on the main node and the sub-node.

[0013] Optionally, the result tree generating unit is further configured to: associate each child node in the mark data with the main node to generate a result tree.

[0014] Optionally, the verification unit is further configured to: summarize the number of main nodes of each result tree, and then compare the number of main nodes with a preset number of business applications, and determine that the verification has failed if the comparison is inconsistent.

[0015] Optionally, the verification unit is further configured to: for each business application, summarize the number of child nodes in the corresponding result tree, and then compare the number of child nodes with the number of corresponding preset tickets. If the comparison is inconsistent, it is determined that the verification has failed.

[0016] In addition, the present application also provides a data processing electronic device, including: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by one or more processors, the one or more processors implement the data processing method as described above.

[0017] In addition, the present application also provides a computer-readable medium on which a computer program is stored, and when the program is executed by a processor, the data processing method as described above is implemented.

[0018] To achieve the above objective, according to another aspect of the embodiments of the present application, a computer program product is provided.

[0019] A computer program product according to an embodiment of the present application includes a computer program, and when the program is executed by a processor, the data processing method provided by the embodiment of the present application is implemented.

[0020] One embodiment of the above invention has the following advantages or beneficial effects: the present application obtains reimbursement application data by responding to a reimbursement request; if the reimbursement application data is displayed in the form of an image, identifies the partition edge of the reimbursement application data, partitions the reimbursement application data based on the partition edge to obtain each partition data, identifies the key fields in each partition data, and classifies the reimbursement application data according to the key fields to obtain each category; marks the reimbursement application data according to each category to obtain marked data; generates a result tree based on the marked data; executes a verification process based on the result tree and pre-recorded business data, and generates an offline annotation task in response to a verification failure and sends it to an offline annotation node. This improves the data processing efficiency and accuracy when processing reimbursement applications and reduces labor costs.

[0021] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings are used to better understand the present application and do not constitute an improper limitation on the present application.

[0023] Figure 1is a schematic diagram of the main flow of a data processing method according to an embodiment of the present application;

[0024] Figure 2 is a schematic diagram of the main flow of a data processing method according to an embodiment of the present application;

[0025] Figure 3 It is a main flow chart of a data processing method according to an embodiment of the present application;

[0026] Figure 4 is a schematic diagram of main units of a data processing device according to an embodiment of the present application;

[0027] Figure 5 is an exemplary system architecture diagram to which the embodiments of the present application can be applied;

[0028] Figure 6 It is a structural diagram of a computer system of a terminal device or server suitable for implementing an embodiment of the present application. DETAILED DESCRIPTION

[0029] The following is a description of the exemplary embodiments of the present application in conjunction with the accompanying drawings, including various details of the embodiments of the present application to facilitate understanding, and they should be considered as merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present application. Similarly, for the sake of clarity and conciseness, the description of known functions and structures is omitted in the following description. It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of the present application are in compliance with the relevant provisions of national laws and regulations. It should be noted that in the embodiments of the present application, certain software, components, models, and other existing solutions in the industry may be mentioned, which should be considered as exemplary, and their purpose is only to illustrate the feasibility of the implementation of the technical solution of the present application, but it does not mean that the applicant has or must have used the solution. In the technical solution of the present application, the collection, analysis, use, transmission, storage, and other aspects of the user's personal information involved are in compliance with the provisions of relevant laws and regulations, and are used for legal and reasonable purposes, and are not shared, leaked, or sold outside of these legal uses, and are subject to supervision and management by regulatory authorities. Take necessary measures to prevent illegal access to such personal information, ensure that persons with access to personal information comply with relevant laws and regulations, and ensure the security of personal information. Once such personal information is no longer needed, risks should be minimized by limiting or even prohibiting data collection and / or deleting data.

[0030] When used, including in certain related applications, user privacy is protected by de-identifying data, such as by removing specific identifiers when used, controlling the amount or specificity of data stored, controlling how data is stored, and / or other methods of de-identification.

[0031] Figure 1 is a schematic diagram of the main flow of a data processing method according to an embodiment of the present application, such as Figure 1 As shown, the data processing method mainly includes the following steps S101 to S105.

[0032] Step S101, in response to a reimbursement request, obtaining reimbursement application data.

[0033] In this embodiment, the execution subject of the data processing method (for example, it can be a server) can receive the reimbursement request through a wired connection or a wireless connection. After receiving the reimbursement request, the execution subject can jump to the target location corresponding to the link based on the link carried in the reimbursement request to obtain the reimbursement application data. For example, the reimbursement application data can be the reimbursement data of various expense vouchers involved in the financial expense management of the enterprise by the budget agency of the financial service enterprise, or it can be the reimbursement data generated by the business trip of the enterprise employees. The embodiment of this application does not specifically limit the content of the reimbursement application data.

[0034] Step S102, if the display form of the reimbursement application data is an image, identify the partition edges of the reimbursement application data, partition the reimbursement application data based on the partition edges to obtain each partition data, identify the key fields in each partition data, and classify the reimbursement application data according to the key fields to obtain each category.

[0035] The execution entity may identify noun words in the reimbursement application data, and determine each identified noun word as a key field in the reimbursement application data.

[0036] Specifically, when the display behavior of the reimbursement application data is an image, the execution entity can identify the partition edge that initially exists in the image. The partition edge can be the edge that appears in the image to distinguish different content areas. The execution entity can partition the reimbursement application data based on the partition edge to obtain each partition, and obtain the key fields in each partition data by identifying each partition data, thereby improving the speed and accuracy of identifying the key fields in the reimbursement data. At the same time, when the same key field as that in another partition data is identified in one partition data, it can automatically remove duplicates and retain one of them, effectively avoiding data redundancy.

[0037] Specifically, the reimbursement application data is classified according to key fields to obtain various categories, including: similarity matching of the key fields with preset category descriptions to determine various categories corresponding to the reimbursement application data based on the matched category descriptions.

[0038] The preset category description may include, for example, reimbursement forms, bills, and other attachments. The embodiments of the present application do not specifically limit the preset category description. Specifically, the execution subject may calculate the cosine similarity between the key field and the preset category description, and then determine the preset category description corresponding to each key field based on the calculated cosine similarity, thereby dividing the reimbursement application data into categories corresponding to the determined preset category description.

[0039] Step S103, marking the reimbursement application data according to each category to obtain marked data.

[0040] According to each obtained category, such as reimbursement form, receipt, and other attachments, the corresponding data in the reimbursement application data are marked respectively to obtain marked data, that is, the reimbursement application data marked with reimbursement form, receipt, and other attachments are obtained.

[0041] Step S104: Generate a result tree based on the labeled data.

[0042] The execution entity can store the data corresponding to each category in the tree nodes or associate them with the tree nodes according to the subordinate relationships of the data corresponding to each category input by the user and the obtained marking data, determine the subordinate relationships of the corresponding tree nodes according to the subordinate relationships of the data corresponding to each category input by the user, and generate a result tree according to the determined subordinate relationships of the tree nodes.

[0043] Step S105, executing a verification process according to the result tree and the pre-recorded business data, and generating an offline annotation task in response to a verification failure and sending it to an offline annotation node.

[0044] The pre-recorded business data may include: a preset number of business applications and a preset number of bills. The embodiment of the present application does not specifically limit the content of the pre-recorded business data.

[0045] According to the number of main nodes and child nodes in the result tree, the actual number of business applications and the actual number of bills (including invoices) can be determined, and the determined actual number of business applications and the preset number of business applications in the pre-recorded business data are verified. The determined actual number of bills and the preset number of bills in the pre-recorded business data are verified. If one of the verification results of the actual number of business applications and the actual number of bills fails, the verification is determined to have failed. When the verification fails, an offline annotation task is generated based on the data that failed the verification and the offline annotation task is sent to the offline annotation node. By way of example, the offline annotation node can be a manual annotation operator, and the embodiments of the present application do not specifically limit the offline annotation node.

[0046] This embodiment obtains reimbursement application data by responding to a reimbursement request; if the display form of the reimbursement application data is an image, the partition edge of the reimbursement application data is identified, the reimbursement application data is partitioned based on the partition edge to obtain each partition data, the key fields in each partition data are identified, and the reimbursement application data is classified according to the key fields to obtain each category; the reimbursement application data is marked according to each category to obtain marked data; a result tree is generated based on the marked data; a verification process is executed according to the result tree and pre-recorded business data, and in response to a verification failure, an offline annotation task is generated and sent to an offline annotation node. This improves the data processing efficiency and accuracy when processing reimbursement applications and reduces labor costs.

[0047] Figure 2 is a schematic diagram of the main flow of a data processing method according to an embodiment of the present application, such as Figure 2 As shown, the data processing method mainly includes the following steps S201 to S205.

[0048] Step S201, in response to a reimbursement request, obtaining reimbursement application data.

[0049] When receiving a reimbursement request, the execution entity may obtain corresponding reimbursement application data from the corresponding data source according to the data source identifier carried in the reimbursement request.

[0050] Step S202, if the display form of the reimbursement application data is an image, identify the partition edges of the reimbursement application data, partition the reimbursement application data based on the partition edges to obtain each partition data, identify the key fields in each partition data, and classify the reimbursement application data according to the key fields to obtain each category.

[0051] The partition edge of the reimbursement application data may be an explicit partition edge, an implicit partition edge, a regular partition edge, or an irregular partition edge. The embodiments of the present application do not specifically limit the explicit form and regularity of the partition edge of the reimbursement application data. The execution subject may also identify the boundary between data of different formats in the image corresponding to the reimbursement application data, and use the boundary as the partition edge of the reimbursement application data. The reimbursement application data is partitioned based on the partition edge to obtain each partition data, the key fields in each partition data are identified, and the reimbursement application data is classified according to the key fields to accurately obtain each category.

[0052] The executing entity can obtain the type identifier carried in the reimbursement request, which is used to characterize the reimbursement type. According to the type identifier, the relevant key fields in the reimbursement application data are identified and extracted to improve the personalization and accuracy of the reimbursement processing. The identified and extracted key fields are matched with the historical reimbursement data corresponding to the preset categories for similarity, and the categories corresponding to the historical reimbursement data corresponding to the similarities greater than the preset threshold are determined as the categories corresponding to the corresponding key fields. The categories corresponding to the extracted key fields can be accurately determined using the historical reimbursement data. The reimbursement application data is classified according to the key fields of the corresponding categories to obtain the categories corresponding to the reimbursement application data.

[0053] Step S203, for each business application, mark the reimbursement application data of the reimbursement form category as a main node, and mark the reimbursement application data of the bill category as a sub-node, so as to obtain marked data based on the main node and the sub-node.

[0054] Since the reimbursement form is the main part, the reimbursement application data of the category of reimbursement form is marked as the main node, and the bill is the sub-item of the reimbursement form, so the reimbursement application data of the category of bill is marked as the sub-node. The marked data can be the reimbursement application data marked with the main node and the sub-node.

[0055] Step S204: Generate a result tree based on the labeled data.

[0056] According to the main node and sub-node marked in the reimbursement application data, a result tree consisting of the main node and sub-node is generated. By generating a result tree, the situation of the reimbursement application data can be clearly seen, which helps to improve the data processing speed and accuracy.

[0057] Specifically, generating a result tree includes associating each child node in the tag data with a main node to generate a result tree.

[0058] Each child node is connected to the main node as a subordinate node of the main node to generate a result tree. The number of main nodes and the number of child nodes of the main node can be marked in the result tree to speed up data processing.

[0059] Step S205, executing a verification process according to the result tree and the pre-recorded business data, and in response to a verification failure, generating an offline annotation task and sending it to the offline annotation node.

[0060] Specifically, the verification process is executed, including: summarizing the number of main nodes of each result tree, and then comparing the number of main nodes with the preset number of business applications. If the comparison is inconsistent, it is determined that the verification has failed.

[0061] The number of main nodes of each result tree is the actual number of business applications. The actual number of business applications is compared with the preset number of business applications in the pre-recorded business data to determine whether they are consistent. If they are inconsistent, it indicates that there is a recognition error or an image with incomplete information recognition, and the verification is determined to have failed.

[0062] Specifically, executing the verification process includes: for each business application, summarizing the number of child nodes in the corresponding result tree, and then comparing the number of child nodes with the number of corresponding preset bills. If the comparison is inconsistent, it is determined that the verification has failed.

[0063] Each business application can correspond to a result tree. For each result tree, the number of child nodes in the corresponding result tree is summarized, that is, the actual number of bills corresponding to the business application. The actual number of bills is compared with the preset number of bills corresponding to the business application in the pre-recorded business data to determine whether they are consistent. If they are inconsistent, it indicates that there is an image with recognition error or incomplete information recognition, and the verification is determined to have failed.

[0064] When the verification fails, an offline annotation task is generated and sent to the offline annotation node for timely correction. The data processing method of the embodiment of the present application can improve the data processing efficiency and accuracy when processing reimbursement applications and reduce labor costs.

[0065] Figure 3 FIG. 1 is a schematic diagram of the main flow of a data processing method according to an embodiment of the present application. Figure 3 As shown, the data processing method of the embodiment of the present application involves pre-recording information - scanning, uploading images - image recognition - verification of recognition results - if the verification passes, executing subsequent links - if the verification fails, manual marking and re-recording are performed, and the manually marked and re-recorded data are verified again until the verification passes, and subsequent links are executed.

[0066] Taking the processing of expense voucher data as an example, the method for extracting content from expense voucher data according to an embodiment of the present application includes the following steps 1-5.

[0067] 1. Pre-recorded information: Asset or expense staff pre-enters the number of transactions for which reimbursement applications are initiated, as well as the total number of bill images in each transaction.

[0068] 2. Scan images: Scan paper expense vouchers, or upload electronic images through files, and import all expense voucher images into the processing system.

[0069] 3. Reimbursement form recognition: Image recognition technology is used to identify the front and back images of the reimbursement form to identify the reimbursement form number and amount. Images that meet this feature (i.e., the presence of the reimbursement form number and amount) are marked as reimbursement forms. Each reimbursement form and subsequent non-reimbursement form images are bound as a business application.

[0070] 4. Bill recognition: Image recognition technology is used to identify the front and back images of invoices or bills, identify and mark their types, identify and extract key information corresponding to the bill type, and mark images that are not identified as reimbursement forms or any type of bills as attachments.

[0071] 5. Verification of recognition results: Verify the number of reimbursement forms against the pre-recorded number of transactions. Also, verify the number of bills bound to the image under each reimbursement form against the pre-recorded number. If there are any inconsistent results, manual annotation is required for adjustment and supplement.

[0072] 6. Main item marking: The main item is the reimbursement form. The main item marking is the marking of the reimbursement form. When the number of business transactions is inconsistent with the number of reimbursement forms, the task is pushed to the manual marking operator to display all images and whether they are marked as the main item. The operator manually checks and confirms, adds or deletes the main item marking, and adjusts the number of main items on the reimbursement form to be consistent with the number of business transactions.

[0073] 7. Image annotation re-recording: When the number of bills identified in a business is inconsistent with the pre-recorded number, there may be an image with an identification error or incomplete information. The system displays the image with incomplete information or inconsistent type in the task and pushes it to the manual annotation operator. The operator checks the image to confirm whether the image type needs to be adjusted and re-records the incomplete information.

[0074] 8. Repeat the verification in step 5 until the verification is passed, and summarize the extracted information and images one by one into each business transaction, and continue with subsequent image verification and other reimbursement approval links.

[0075] The embodiment of the present application provides a system solution for quickly and accurately extracting information on expense vouchers: first, the total number of bills needs to be entered, and all voucher images, including the front and back of reimbursement forms, invoices or bills, and attachments, are scanned and imported into the system; reimbursement forms, invoices or bills are automatically identified and classified using artificial intelligence image recognition technology, and unidentified images are classified as attachments, and key information of reimbursement forms, invoices or bills is extracted through image recognition; the number and sequence of reimbursement forms are verified, and each reimbursement application must have and only have one reimbursement form as the first main item, and at the same time, the total number of entered bills is verified to match the results of the identification and classification; if the main item of the reimbursement form fails to be judged, or the number of bills fails to be verified, it may be that the identification is not completely accurate, and the main item identified by the expense voucher and the front and back images of other voucher images are pushed to fixed personnel for confirmation, and the main item mark is adjusted and marked; the images of the bills that do not match the pre-entered number are pushed to the input personnel at the same time, and the classification of the images is corrected at the same time, and the key information of the images is checked; until the main item of the reimbursement form and the number of bills are checked, the system automatically summarizes the information extraction results for subsequent verification and processing. The embodiment of the present application adopts image recognition technology for identification and classification. On the one hand, it automatically identifies the image information of standardized layout and extracts the key content of printed text to reduce a large number of manual identification and entry operations. On the other hand, it verifies the classification results to confirm the potential risk of recognition failure. After multiple manual corrections, the accuracy of information extraction can be guaranteed.

[0076] The embodiment of the present application utilizes artificial intelligence image recognition technology to simultaneously identify and classify the uploaded expense vouchers, and extracts data in combination with manual review and annotation, which can achieve fast and accurate extraction of expense voucher data. The expense voucher image information extraction method of the embodiment of the present application utilizes image recognition technology, which greatly reduces the reliance on manual operations, improves efficiency and reduces errors in manual processing; at the same time, by identifying the comparison results of the image type, timely intervention in manual review, annotation, and supplementary recording can be achieved, ensuring the integrity and reliability of information extraction, and realizing an efficient, accurate, and low-manpower-cost expense voucher information extraction solution.

[0077] The embodiment of the present application makes the image information extraction process of multiple expense vouchers online, reducing the manual workload of data transmission and processing by personnel of the expense center or related institutions. Intelligent image recognition technology is used to quickly extract the printed information of a large amount of expense voucher images, complete the automatic extraction of the main information content, and improve the overall efficiency. The recognition type and information integrity check are used to avoid the recognition errors and omissions that may occur in image recognition, and the completeness and accuracy of the recognition are ensured through manual annotation and supplementation.

[0078] Figure 4 Schematic diagram of the main units of the data processing device according to an embodiment of the present application. Figure 4 As shown, the data processing device 400 includes an acquisition unit 401 , a classification unit 402 , a marking unit 403 , a result tree generation unit 404 and a verification unit 405 .

[0079] The acquisition unit 401 is configured to acquire reimbursement application data in response to a reimbursement request.

[0080] The classification unit 402 is configured to identify the partition edges of the reimbursement application data if the display form of the reimbursement application data is an image, partition the reimbursement application data based on the partition edges to obtain individual partition data, identify key fields in each partition data, and classify the reimbursement application data according to the key fields to obtain individual categories.

[0081] The marking unit 403 is configured to mark the reimbursement application data according to various categories to obtain marked data.

[0082] The result tree generating unit 404 is configured to generate a result tree based on the marked data.

[0083] The verification unit 405 is configured to execute a verification process according to the result tree and the pre-recorded business data, and in response to a verification failure, generate an offline annotation task and send it to the offline annotation node.

[0084] In some embodiments, the classification unit 402 is further configured to: perform similarity matching between the key fields and the preset category descriptions, so as to determine the categories corresponding to the reimbursement application data based on the matched category descriptions.

[0085] In some embodiments, the marking unit 403 is further configured to: for each business application, mark the reimbursement application data of the category of reimbursement form as the main node, and mark the reimbursement application data of the category of bill as the child node, so as to obtain marked data based on the main node and the child node.

[0086] In some embodiments, the result tree generating unit 404 is further configured to: associate each child node in the tag data with the main node to generate a result tree.

[0087] In some embodiments, the verification unit 405 is further configured to: summarize the number of master nodes of each result tree, and then compare the number of master nodes with the preset number of business applications, and determine that the verification has failed if the comparison is inconsistent.

[0088] In some embodiments, the verification unit 405 is further configured to: for each business application, summarize the number of child nodes in the corresponding result tree, and then compare the number of child nodes with the number of corresponding preset tickets. If the comparison is inconsistent, it is determined that the verification has failed.

[0089] It should be noted that the data processing method and the data processing device of the present application have a corresponding relationship in terms of specific implementation contents, so the repeated contents will not be described again.

[0090] Figure 5 An exemplary system architecture 500 to which the data processing method or data processing device according to the embodiment of the present application can be applied is shown.

[0091] like Figure 5 As shown, system architecture 500 may include terminal devices 501, 502, 503, a network 504 and a server 505. Network 504 is used to provide a medium for communication links between terminal devices 501, 502, 503 and server 505. Network 504 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0092] Users can use terminal devices 501, 502, 503 to interact with server 505 through network 504 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 501, 502, 503, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only as examples).

[0093] The terminal devices 501 , 502 , and 503 may be various electronic devices having a data processing screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0094] Server 505 can be a server that provides various services, such as a background management server that provides support for reimbursement requests submitted by users using terminal devices 501, 502, and 503 (only as an example). The background management server can obtain reimbursement application data in response to the reimbursement request; if the display form of the reimbursement application data is an image, identify the partition edge of the reimbursement application data, partition the reimbursement application data based on the partition edge to obtain each partition data, identify the key fields in each partition data, and classify the reimbursement application data according to the key fields to obtain each category; mark the reimbursement application data according to each category to obtain marked data; generate a result tree based on the marked data; execute the verification process according to the result tree and the pre-recorded business data, and generate an offline annotation task in response to the failure of the verification and send it to the offline annotation node. This improves the data processing efficiency and accuracy when processing reimbursement applications and reduces labor costs.

[0095] It should be noted that the data processing method provided in the embodiment of the present application is generally executed by the server 505 , and accordingly, the data processing device is generally set in the server 505 .

[0096] It should be understood that Figure 5 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0097] Reference below Figure 6 , which shows a schematic diagram of the structure of a computer system 600 of a terminal device suitable for implementing an embodiment of the present application. Figure 6 The terminal device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0098] like Figure 6 As shown, the computer system 600 includes a central processing unit (CPU) 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 into a random access memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the computer system 600 are also stored. The CPU 601, the ROM 602, and the RAM 603 are connected to each other via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0099] The following components are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage section 608 as needed.

[0100] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the central processing unit (CPU) 601, the above-mentioned functions defined in the system of the present application are executed.

[0101] It should be noted that the computer-readable medium shown in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. Computer-readable storage media may include, for example, but are not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, 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 above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0102] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the above-mentioned module, program segment or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0103] The units involved in the embodiments described in the present application may be implemented by software or hardware. The units described may also be set in a processor, for example, it may be described as: a processor includes an acquisition unit, a classification unit, a marking unit, a result tree generation unit, and a verification unit. The names of these units do not constitute limitations on the units themselves in some cases.

[0104] As another aspect, the present application also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently and not be assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device responds to the reimbursement request and obtains the reimbursement application data; if the display form of the reimbursement application data is an image, the partition edge of the reimbursement application data is identified, the reimbursement application data is partitioned based on the partition edge to obtain each partition data, the key fields in each partition data are identified, and the reimbursement application data is classified according to the key fields to obtain each category; the reimbursement application data is marked according to each category to obtain marked data; based on the marked data, a result tree is generated; according to the result tree and the pre-recorded business data, a verification process is executed, and in response to the failure of the verification, an offline annotation task is generated and sent to the offline annotation node.

[0105] The computer program product of the present application includes a computer program, and the computer program implements the data processing method in the embodiment of the present application when executed by a processor.

[0106] According to the technical solution of the embodiment of the present application, the data processing efficiency and accuracy when processing reimbursement applications can be improved and the labor cost can be reduced.

[0107] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A data processing method, characterized in that: include: Responding to the reimbursement request, obtaining reimbursement application data; If the display form of the reimbursement application data is an image, identifying the partition edge of the reimbursement application data, partitioning the reimbursement application data based on the partition edge to obtain each partition data, identifying the key field in each partition data, and classifying the reimbursement application data according to the key field to obtain each category; Marking the reimbursement application data according to the various categories to obtain marked data; Based on the labeled data, a result tree is generated; A verification process is executed according to the result tree and the pre-recorded business data, and in response to a verification failure, an offline annotation task is generated and sent to an offline annotation node.

2. The method according to claim 1, characterized in that: The reimbursement application data is classified according to the key fields to obtain various categories, including: The key fields are matched with preset category descriptions for similarity, so as to determine the categories corresponding to the reimbursement application data based on the matched category descriptions.

3. The method according to claim 1, characterized in that The step of marking the reimbursement application data according to the various categories to obtain marked data includes: For each business application, the reimbursement application data of the reimbursement form category is marked as a main node, and the reimbursement application data of the bill category is marked as a sub-node, so as to obtain marked data based on the main node and the sub-node.

4. The method according to claim 3, characterized in that The generated result tree includes: Each of the child nodes in the tag data is associated with the main node to generate a result tree.

5. The method according to claim 4, characterized in that The execution verification process includes: The number of main nodes of each result tree is summarized, and then the number of main nodes is compared with the preset number of business applications. If the comparison is inconsistent, it is determined that the verification has failed.

6. The method according to claim 4, characterized in that The execution verification process includes: For each business application, the number of child nodes in the corresponding result tree is summarized, and then the number of child nodes is compared with the number of corresponding preset bills. If the comparison is inconsistent, it is determined that the verification has failed.

7. A data processing device, characterized in that: include: an acquisition unit, configured to acquire reimbursement application data in response to a reimbursement request; a classification unit configured to, if the display form of the reimbursement application data is an image, identify a partition edge of the reimbursement application data, partition the reimbursement application data based on the partition edge to obtain individual partition data, identify a key field in each of the partition data, and classify the reimbursement application data according to the key field to obtain individual categories; a marking unit configured to mark the reimbursement application data according to the respective categories to obtain marked data; A result tree generating unit, configured to generate a result tree based on the marked data; The verification unit is configured to execute a verification process according to the result tree and pre-recorded business data, and in response to a verification failure, generate an offline annotation task and send it to an offline annotation node.

8. The device according to claim 7, characterized in that The classification unit is further configured to: The key fields are matched with preset category descriptions for similarity, so as to determine the categories corresponding to the reimbursement application data based on the matched category descriptions.

9. The device according to claim 7, characterized in that The marking unit is further configured to: For each business application, the reimbursement application data of the reimbursement form category is marked as a main node, and the reimbursement application data of the bill category is marked as a sub-node, so as to obtain marked data based on the main node and the sub-node.

10. The device according to claim 9, characterized in that The result tree generation unit is further configured to: Each of the child nodes in the tag data is associated with the main node to generate a result tree.

11. The device according to claim 10, characterized in that The verification unit is further configured to: The number of main nodes of each result tree is summarized, and then the number of main nodes is compared with the preset number of business applications. If the comparison is inconsistent, it is determined that the verification has failed.

12. The device according to claim 10, characterized in that The verification unit is further configured to: For each business application, the number of child nodes in the corresponding result tree is summarized, and then the number of child nodes is compared with the number of corresponding preset bills. If the comparison is inconsistent, it is determined that the verification has failed.

13. A data processing electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

14. A computer readable 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 6 is implemented.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images