Accessory approval method and device applied to approval process, storage medium and computer equipment

Abstract

The invention discloses an attachment approval method and device applied to an approval process, a storage medium and computer equipment. The method comprises the steps of receiving an attachment file associated with an approval process submitted by a user; extracting structured data in the attachment file; inputting the structured data into a pre-trained attachment examination and approval model, and performing normalization judgment on a to-be-examined and approved target data field by the attachment examination and approval model to generate an examination and approval result; if the approval result indicates that all the target data fields meet the filling specification, generating an approval passing instruction; and if the approval result indicates that the target data field which does not meet the filling specification exists, generating a modification feedback instruction containing a problem field identifier. When the user submits the examination and approval process, the associated attachment file is subjected to pre-examination and approval, and the next examination and approval link is carried out after successful examination and approval, so that the condition that modification is returned at a subsequent node can be avoided, the time consumption of the examination and approval process can be effectively reduced, and the examination and approval efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of document processing, and in particular to an attachment approval method and device applied to an approval process, a storage medium and a computer device. BACKGROUND

[0002] In the existing enterprise management system, when a user initiates an approval process involving an attachment, the user usually needs to fill in relevant information and upload an attachment file in a process initiation interface, and then submit an approval application. A process engine inside the system will push the approval task to the to-do list of the first approver according to the preset process node configuration. The approver needs to log in to the system, check the to-do task and the corresponding attachment content, manually judge whether the attachment content meets the standard requirements, and if it is determined that the attachment does not meet the standard requirements, the approver needs to perform a return operation, and the entire process will be terminated and returned to the initiator, requiring the initiator to modify the attachment and re-initiate the entire approval process.

[0003] The existing technology highly depends on manual review of the standardization of attachments in the approval process. However, due to subjective differences in standards and experience levels of different approvers, the approval results lack consistency and accuracy. Secondly, when the approval process involves multiple nodes, each link needs to be manually reviewed in sequence, which not only introduces waiting time but also prolongs the overall approval cycle. Once it is found that the attachment does not meet the standard in the approval link, the entire process is returned to the starting point, and the initiator needs to modify and re-execute the complete approval process, resulting in a long time-consuming and low efficiency of the entire approval process. SUMMARY

[0004] To this end, embodiments of the present application provide an attachment approval method and device applied to an approval process, a storage medium and a computer device, which realize pre-positioned automatic review of associated attachment files when a user submits an approval process, and then transfer to the next approval link after successful review, which can avoid the situation of being returned to re-execute the approval process in subsequent nodes, effectively reducing the time-consuming of the approval process and improving the efficiency of the approval process.

[0005] In a first aspect, the present application provides an attachment approval method applied to an approval process.

[0006] The present application is achieved by the following technical solutions: An attachment approval method applied to an approval process, comprising: receiving an attachment file associated with a submitted approval process by a user; extracting structured data in the attachment file; inputting the structured data into a pre-trained attachment approval model, the attachment approval model performing standardization determination on a target data field to be approved to generate an approval result; If the approval result indicates that all target data fields meet the filling specifications, an approval approval instruction is generated; If the approval result indicates that there are target data fields that do not meet the filling specifications, a modification feedback instruction containing the identifier of the problematic field will be generated.

[0007] In a preferred embodiment of this application, the structured data in the attachment file can be further extracted, including: The text content of the attached file is identified, and the structured data of the attached file is obtained based on the text content.

[0008] In a preferred example of this application, the training process of the attachment approval model may further include: obtaining negative samples that were manually rejected and positive samples that were manually approved in the historical approval process; Label the error types of the verification items in negative samples and the compliance labels of the verification items in positive samples; Negative and positive samples and their corresponding data labels are input into a deep learning model for training.

[0009] In a preferred example of this application, the method for generating a modification feedback instruction containing an identifier for the problem field may be further configured to include: Locate the fields in the attachment file that do not meet the filling specifications; The user interface highlights the question field and displays the correct filling specifications in conjunction with it.

[0010] In a preferred embodiment of this application, before extracting the structured data from the attachment file, the method may further include: Perform structural verification of the attachment file, compare the attachment file with the pre-stored standard template using the document object model tree, and calculate the tree structure similarity between the attachment file and the standard template; If the tree structure similarity is lower than the preset threshold, it is judged as a structural error, and the user is prompted to modify and resubmit.

[0011] In a preferred example of this application, if the evaluation result indicates that all target data fields meet the filling rules, then after generating the approval instruction, the following further steps are included: The approved attachment documents are used as positive samples and input into the attachment approval model for incremental training.

[0012] Secondly, this application provides an attachment approval device for use in the approval process.

[0013] This application is achieved through the following technical solution: An attachment approval device for an approval process, used to execute the attachment approval method for an approval process described in the first aspect above, the device comprising: The file acquisition module is used to receive attachment files related to the approval process submitted by the user; The data extraction module is used to extract structured data from the attached files; The attachment approval module is used to input the structured data into a pre-trained attachment approval model, which performs a standardization judgment on the target data fields to be approved and generates an approval result. The result feedback module is used to generate an approval instruction if the approval result indicates that all target data fields meet the filling specifications; and to generate a modification feedback instruction containing the identifier of the problematic field if the approval result indicates that there are target data fields that do not meet the filling specifications.

[0014] In a preferred embodiment of this application, the data extraction module is further configured to: identify the text content of the attachment file and obtain the structured data of the attachment file based on the text content.

[0015] Thirdly, this application provides a computer-readable storage medium.

[0016] This application is achieved through the following technical solution: A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of any of the above-described methods for approving attachments in an approval process.

[0017] Fourthly, this application provides a computer device.

[0018] This application is achieved through the following technical solution: A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of any of the above-described appendix approval methods applied to the approval process.

[0019] In summary, compared with the prior art, the beneficial effects of the technical solution provided by the embodiments of this application include at least the following: This application receives attachment files associated with the approval process submitted by the user; extracts structured data from the attachment files; inputs the structured data into a pre-trained attachment approval model, the attachment approval model performs a standardization judgment on the target data fields to be approved, and generates an approval result; if the approval result indicates that all target data fields meet the filling specifications, an approval pass instruction is generated; if the approval result indicates that there are target data fields that do not meet the filling specifications, a modification feedback instruction containing the problem field identifier is generated. This application automatically reviews attachments when users submit their applications for approval, allowing the process to proceed to subsequent stages only after the initial review is passed. This directly eliminates situations where applications are returned for resubmission due to attachment issues in the later stages of the process, significantly shortening the overall approval cycle. Simultaneously, by using AI to replace manual review for basic compliance judgments, it completely eliminates inconsistencies in standards caused by subjective differences and experience levels among approvers, greatly improving the accuracy and consistency of the review. Furthermore, by intercepting compliance risks at the beginning of the process, it avoids the waste of resources caused by invalid transfers of non-compliant applications and frees up approvers to focus on higher-value decision-making tasks, thereby comprehensively improving the efficiency, quality, and business response speed of the approval process. Attached Figure Description

[0020] Figure 1 A flowchart illustrating an attachment approval method applied to an approval process, as provided in an embodiment of this application; Figure 2 A schematic diagram of an attachment approval device applied to an approval process according to an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application; Explanation of reference numerals in the attached figures: Module 10 for document acquisition, Module 20 for data extraction, Module 30 for attachment approval, and Module 40 for result feedback. Detailed Implementation

[0021] This specific embodiment is merely an explanation of this application and is not intended to limit it. After reading this specification, those skilled in the art can make modifications to this embodiment without contributing any inventive step, but such modifications are protected by patent law as long as they fall within the scope of the claims of this application.

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] Furthermore, the term "and / or" in this application is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this application, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0024] In this application, the terms "first," "second," etc., are used to distinguish identical or similar items with essentially the same function. It should be understood that there is no logical or temporal dependency between "first," "second," and "nth," nor are there any restrictions on quantity or execution order.

[0025] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0026] In current enterprise management systems, approval processes involving attachments heavily rely on manual review: after a user submits an application, the first approver must manually check the attachments for compliance; if they do not meet the requirements, the application must be returned to the initiator for revision and the entire process must be restarted. This model has significant drawbacks: first, manual review, due to subjective differences in the approver's experience and standards, leads to inconsistent and inaccurate results; second, in multi-node sequential approval processes, manual checks at each stage introduce waiting time, lengthening the overall cycle; and third, if attachment issues are discovered in the later stages of the process, the entire process will be returned to the starting point, resulting in ineffective workflow and wasted time, ultimately making the approval process time-consuming and inefficient.

[0027] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0028] The first exemplary embodiment of this application provides a method for approving attachments in an approval process, applicable to scenarios such as process approval in enterprise management. The executing entity of this method can be the data detection device provided in the embodiments of this application, or an electronic device integrating the aforementioned detection device. The aforementioned electronic devices include, but are not limited to: wearable devices, handheld devices, personal computers, tablets, in-vehicle devices, smartphones, computing devices, or other processing devices connected to a wireless modem. In different networks, electronic devices can be called by different names, such as: user equipment, access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, cellular phone, cordless phone, personal digital assistant (PDA), electronic devices in 5G networks or future evolved networks, etc.

[0029] The server can be a standalone server device, such as a rack-mount, blade, tower, or cabinet server device, or a workstation, mainframe, or other hardware device with strong computing power; or it can be a server cluster composed of multiple servers. The servers in the service cluster can be composed in a symmetrical manner, wherein each server is functionally and equally important in the business chain, and each server can provide services to the outside world independently. Providing services independently can be understood as not requiring the assistance of other servers.

[0030] See Figure 1 As shown, the specific steps of an attachment approval method applied to the approval process include: S1: Receives attachments related to the approval process submitted by the user.

[0031] In practice, when users submit their approval applications, they can upload attachment files through the "Add Attachment" button or drag-and-drop area on the approval form of the electronic device.

[0032] After the attachment file is uploaded, it is stored in a secure location on the server or in a cloud storage server, and the server records the information of the attachment file, including at least: the unique file identifier of the attachment file, the original file name, the storage path, the upload time, the uploading user, and the associated approval process ID.

[0033] Attachment files can be common file types, including documents (.doc, .docx, .pdf, or .txt), images (e.g., .jpg, .png, or .gif), etc.

[0034] S2: Extract structured data from the attachment file.

[0035] In practice, the attachment file is input into the multimodal document parsing module for parsing, resulting in structured JSON data of the attachment file. This structured data typically includes metadata fields, key content fields, and logical relationship fields. Metadata fields include a unique document identifier, document type classification, total word count, and original document format; key content fields include text fragments and fragment coordinates; and logical relationship fields detail the logical relationships between text fragments.

[0036] S3: Input structured data into the pre-trained attachment approval model. The attachment approval model performs a standardization judgment on the target data fields to be approved and generates the approval result.

[0037] S4: If the approval result indicates that all target data fields meet the filling specifications, then an approval approval instruction will be generated; S5: If the approval result indicates that there are target data fields that do not meet the filling specifications, a modification feedback instruction containing the problem field identifier will be generated.

[0038] This application automatically reviews attachments when users submit their applications for approval, allowing the process to proceed to subsequent stages only after the initial review is passed. This directly eliminates situations where applications are returned for resubmission due to attachment issues in the later stages of the process, significantly shortening the overall approval cycle. Simultaneously, by using AI to replace manual review for basic compliance judgments, it completely eliminates inconsistencies in standards caused by subjective differences and experience levels among approvers, greatly improving the accuracy and consistency of the review. Furthermore, by intercepting compliance risks at the beginning of the process, it avoids the waste of resources caused by invalid transfers of non-compliant applications and frees up approvers to focus on higher-value decision-making tasks, thereby comprehensively improving the efficiency, quality, and business response speed of the approval process.

[0039] In some embodiments, step S2 involves extracting structured data from the attachment file, including: Identify the text content of the attachment file and obtain the structured data of the attachment file based on the text content.

[0040] In practical implementation, a document parsing model is used to identify the text content of attachment files. Preferably, a multimodal document parsing model integrating image recognition and text recognition is used, which can match attachment files of different formats for text content recognition.

[0041] Typically, to ensure that attachments conform to the processing format of document parsing models, they need to be preprocessed and standardized before being input into these models. Different document parsing models are then used to process the attachments according to their document format.

[0042] For example, when the attachment file is in PDF or image format, the OCR engine is used to recognize the text content in the attachment file; When the attachment file is in Word or Excel format, first standardize the attachment file and convert it into XML format, and then recognize the text content within it. All the text content extracted from the attachment file is combined and cleaned, including regularization and stop word processing. The preprocessed text content is then input into the layout analysis model to obtain the chapter and page number mapping of the attachment file. Finally, the elements of the attachment file are obtained, and the structured data of the attachment file in JSON format is formed based on the extracted elements.

[0043] In some embodiments, the training process of the attachment approval model includes: Obtain negative samples that were manually rejected and positive samples that were manually approved in the historical approval process; Label the error types of the verification items in negative samples and the compliance labels of the verification items in positive samples; Negative and positive samples and their corresponding data labels are input into a deep learning model for training.

[0044] In practice, the approval data from the past two years is first extracted from the review system. The attachments in the approval data are divided into two types of samples: attachments that are manually approved are positive samples, and attachments that are manually returned are negative samples.

[0045] When labeling positive samples, each key field is labeled with a "compliance label" and its location within the document. When labeling negative samples, the error type and location within the document are indicated for the problematic key field. Error types include formatting errors, logical conflicts, or missing data, which can be represented by different coded numbers. Location information includes the page number and coordinate frame of the key field. The labeled positive and negative samples are then converted into structured data.

[0046] Structured data, including both negative and positive samples, is used as training data to train a deep learning model based on BERT. This deep learning model includes an input layer, a feature extraction layer, a feature fusion layer, and a result output layer. The input layer receives the structured input data; the feature extraction layer includes a rule-based feature channel and a semantic feature channel. The rule-based feature channel encodes business rules into feature vectors, while the semantic feature channel understands the contextual semantics of fields using a BERT-like model and outputs semantic features; the feature fusion layer fuses the feature vectors and semantic features based on an attention mechanism; and the result output layer outputs the validation results, including the validity classification of validation items and the location coordinates of the problematic fields.

[0047] To address the imbalanced sample problem during training, Focal Loss was used to construct the field classification loss, cross-entropy loss was used as the misclassification loss function, and smoothing L1 loss was used as the location prediction loss function. The weights of the three loss functions were adjusted during training until the prediction accuracy of the deep learning model reached the preset target. This model was then approved as an attachment model.

[0048] In some embodiments, generating a modification feedback instruction that includes a problem field identifier further includes: Locate the fields in the attachment file that do not meet the filling specifications; The user interface highlights the question field and displays the correct filling specifications in conjunction with it.

[0049] In practice, the location of the issue field is obtained, including the document page number where the issue field is located and the coordinate box of the issue field. For example, it can be represented as {"page":3,"bbox":[120,340,280,380]}. Furthermore, the position coordinates of the issue field in the original document are converted into the front-end rendering coordinate system, establishing a coordinate mapping relationship: Actual coordinates = Original position coordinates × Page scaling factor + Page offset.

[0050] The system uses Canvas or PDF.js to render the document content based on the actual coordinates obtained from the conversion, overlaying a semi-transparent highlight layer over the positioned area. A floating card is then displayed over the highlighted area, showing the correct filling specification for the incorrect field. The correct filling specification can be retrieved from the rule base based on the field name.

[0051] Through a smart feedback mechanism that combines spatial positioning and highlighting, users who submit approval requests can more intuitively understand the issues and modify the specifications, making adjustments easier and improving the efficiency of the approval process.

[0052] In one feasible implementation, AR technology can be used to mark question fields on the user interface and associate them with the correct filling specifications. AR assistance can more intuitively mark the parts that need to be modified on the user interface. First, a smart device (such as a mobile phone or tablet) is needed in conjunction with sensors (such as cameras and gyroscopes / accelerometers) to capture the real environment and track the device's posture. Through technologies such as SLAM (Simultaneous Localization and Mapping), a 3D model of the environment is built, and the device's position and posture in the environment are tracked in real time. Using image recognition or machine learning algorithms, various elements on the user interface (such as buttons, text boxes, or icons) are identified, and the identified elements are matched with a predefined interface model to determine the function and attributes of each element. According to the question fields that the user needs to modify, AR markers are overlaid on the user interface. For example, the area that needs to be modified is highlighted with color, and / or arrows are used to point to the element that needs to be operated, and / or virtual labels are displayed next to the element to provide modification instructions or prompts, and / or the modification operation is demonstrated to the user through animation. Meanwhile, AR assistance also supports users to interact with AR tags through gestures, voice, or touch. Based on the user's interaction, it provides real-time feedback to the user. For example, it receives the user's interaction through the sensor module, responds to the interaction, identifies the AR tag associated with the interaction, and displays the user's detailed information about the AR tag.

[0053] In some embodiments, before extracting the structured data from the attachment file, the method further includes: Before extracting the structured data from the attachment file, the following steps are also included: Perform structural verification of the attachment file, compare the attachment file with the pre-stored standard template using the document object model tree, and calculate the tree structure similarity between the attachment file and the standard template; If the tree structure similarity is lower than the preset threshold, it is judged as a structural error, and the user is prompted to modify and resubmit.

[0054] In actual implementation, standard modules need to be created according to the approval type, and a standard document object model (DOM) tree is generated through the document parsing engine. The standard document DOM tree includes the following features: node type (heading, table, paragraph, signature area); node hierarchy relationship that defines heading levels and nesting relationships; and page coordinates, which record the coordinate range of each node on the page.

[0055] The attachment file to be verified is parsed to generate a DOM tree to be verified, including node type, node hierarchy and page coordinate triples; Using a tree structure comparison algorithm, the DOM tree to be verified is compared with the standard document DOM tree, and the node type matching degree is calculated: Node type matching rate = |N_template∩N_test| / |N_template|; Calculate node hierarchical similarity: Level score = 1 - Σ|L_template(i) - L_test(i)| / (maximum level difference × number of matching nodes); Calculate spatial similarity: Position score = Σ[IOU(B_template(i),B_test(i))] / number of matched nodes; Where IOU(Box1,Box2) = intersection area / union area; The similarity of the tree structure is calculated by combining node type matching rate, level score, and position score: The similarity of the tree structure = α × matching rate + β × level score + γ × position score, where α, β, and γ can be selected according to actual needs. In this application, α = 0.4, β = 0.3, and γ = 0.3.

[0056] The structure is deemed acceptable when the similarity of the tree structure exceeds a first threshold. The first threshold can be set according to actual needs; in this application, it can be 90%.

[0057] In some embodiments, the submitted attachments and modification process during the approval process, as well as the approved attachments, are all recorded on the blockchain to ensure the integrity and immutability of the documents.

[0058] If the verification result indicates that all target data fields meet the filling rules, the process after generating the approval instruction also includes: using the approved attachments as positive samples to input into the attachment approval model for incremental training.

[0059] The approved attachments are re-entered into the training queue, and the attachment approval model is optimized at preset time intervals to continuously improve the model's recall. Over time, the model continuously learns new standard cases, improving its accuracy and processing efficiency.

[0060] Another embodiment of this application also provides an attachment approval device for an approval process, used to execute the above-described attachment approval method for an approval process, such as... Figure 2 As shown, the detection device includes: The document acquisition module 10 is used to receive attachment files related to the approval process submitted by the user; Data extraction module 20 is used to extract structured data from the attachment file; The attachment approval module 30 is used to input the structured data into the pre-trained attachment approval model, which performs a standardization judgment on the target data fields to be approved and generates an approval result. The result feedback module 40 is used to generate an approval instruction if the approval result indicates that all target data fields meet the filling specifications; and to generate a modification feedback instruction containing the problem field identifier if the approval result indicates that there are target data fields that do not meet the filling specifications.

[0061] The data extraction module 20 is specifically used to: identify the text content of the attachment file and obtain the structured data of the attachment file based on the text content.

[0062] The specific limitations of the detection device for the attachments in the approval process provided in this embodiment can be found in the embodiments of the detection method described above, and will not be repeated here. Each module in the above-mentioned detection device can be implemented entirely or partially through software, hardware, or a combination thereof. Each module can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0063] This specification also provides a computer-readable storage medium storing a computer program that, when executed by a processor, can be used to perform the above-described method for approving attachments in an approval process.

[0064] The embodiments in this specification also provide Figure 3 The diagram shows the structure of the electronic device. Figure 3 At the hardware level, the electronic device includes a processor, an internal bus, a network interface, memory, and non-volatile storage, and may also include other hardware required for business operations. The processor reads the corresponding computer program from the non-volatile storage into memory and then runs it to implement the attachment approval method for the approval process described in the first embodiment above.

[0065] Of course, in addition to software implementation, this specification does not exclude other implementation methods, such as logic devices or a combination of hardware and software. In other words, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0066] Improvements in a technology can be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many improvements to the methodology can now be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that an improvement in methodology cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (e.g., a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog are the most commonly used. Those skilled in the art should understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0067] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0068] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0069] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0070] Those skilled in the art will understand that embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0071] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0072] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0073] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0074] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0075] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0076] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0077] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0078] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0079] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0080] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0081] The above description is merely an embodiment of this specification and is not intended to limit this specification. Various modifications and variations can be made to this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of the claims of this specification.

Claims

1. A method for approving attachments in an approval process, characterized in that, The method includes: Receive attachments related to the approval process submitted by the user; Extract the structured data from the attached file; The structured data is input into a pre-trained attachment approval model, which performs a standardization judgment on the target data fields to be approved and generates an approval result. If the approval result indicates that all target data fields meet the filling specifications, an approval approval instruction is generated; If the approval result indicates that there are target data fields that do not meet the filling specifications, a modification feedback instruction containing the identifier of the problematic field will be generated.

2. The attachment approval method applied to the approval process according to claim 1, characterized in that, Extract the structured data from the attached file, including: The text content of the attached file is identified, and the structured data of the attached file is obtained based on the text content.

3. The attachment approval method applied to the approval process according to claim 1, characterized in that, The training process of the attachment approval model includes: Obtain negative samples that were manually rejected and positive samples that were manually approved in the historical approval process; Label the error types of the verification items in negative samples and the compliance labels of the verification items in positive samples; Negative and positive samples and their corresponding data labels are input into a deep learning model for training.

4. The attachment approval method applied to the approval process according to claim 1, characterized in that, After generating the modification feedback instruction containing the issue field identifier, it also includes: Locate the problematic fields in the attached file that do not meet the filling specifications, highlight the problematic fields in the user interface, and display the correct filling specifications in conjunction with them.

5. The attachment approval method applied to the approval process according to any one of claims 1 to 4, characterized in that, Before extracting the structured data from the attached file, the process also includes: Perform structural verification of the attachment file, compare the attachment file with the pre-stored standard template using the document object model tree, and calculate the tree structure similarity between the attachment file and the standard template; If the tree structure similarity is lower than the preset threshold, it is judged as a structural error, and the user is prompted to modify and resubmit.

6. The attachment approval method applied to the approval process according to claim 5, characterized in that, If the evaluation results indicate that all target data fields meet the filling rules, the approval instruction will also include: The approved attachment documents are used as positive samples and input into the attachment approval model for incremental training.

7. An attachment approval device applied in an approval process, characterized in that, The apparatus for performing the attachment approval method for an approval process as described in claim 1, the apparatus comprising: The file acquisition module is used to receive attachment files related to the approval process submitted by the user; The data extraction module is used to extract structured data from the attached files; The attachment approval module is used to input the structured data into a pre-trained attachment approval model, which performs a standardization judgment on the target data fields to be approved and generates an approval result. The result feedback module is used to generate an approval instruction if the approval result indicates that all target data fields meet the filling specifications; and to generate a modification feedback instruction containing the identifier of the problematic field if the approval result indicates that there are target data fields that do not meet the filling specifications.

8. The attachment approval device applied to the approval process according to claim 7, characterized in that, The data extraction module is specifically used to: identify the text content of the attachment file, and obtain the structured data of the attachment file based on the text content.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method according to any one of claims 1 to 7.

10. A computer device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 7.

Images