Method and device for generating approval scheme, equipment and storage medium

Abstract

The embodiment of the invention relates to a method and device for generating an approval scheme, equipment and a storage medium. The method provided by the invention comprises the following steps: in response to an approval request, determining to-be-approved approval content related to the approval request; determining a plurality of candidate approval configurations related to the approval request from a predetermined approval configuration set, wherein the approval configurations in the approval configuration set at least indicate an approval mode on at least one approval node; combining at least two candidate approval configurations in the plurality of candidate approval configurations to determine a group of target approval configurations for the approval content, the group of target approval configurations indicating an approval process related to the approval content and an approval mode on each approval node in the approval process; and generating an examination and approval scheme for the examination and approval request at least based on the examination and approval process and the examination and approval mode indicated by the group of target examination and approval configurations.

Description

Technical Field

[0001] Example embodiments of the present disclosure generally relate to the field of computer technology, and more particularly, to a method, apparatus, device, and storage medium for generating an approval solution. Background Art

[0002] In the current technological environment, many enterprises or organizations have established special systems for configuration management of approval plans in order to better meet internal business needs. These systems are often tailored to the specific needs of the enterprise, allowing approval plans to be customized for different departments or business units by writing business codes. This approach ensures the standardization and consistency of approval plans, thereby better supporting the operation and development of the enterprise. Summary of the invention

[0003] In a first aspect of the present disclosure, a method for generating an approval scheme is provided, the method comprising: in response to an approval request, determining approval content to be approved related to the approval request; determining a plurality of candidate approval configurations related to the approval request from a predetermined approval configuration set, wherein the approval configuration in the approval configuration set at least indicates an approval method on at least one approval node; combining at least two of the plurality of candidate approval configurations to determine a set of target approval configurations for the approval content, wherein the set of target approval configurations indicates an approval process related to the approval content and an approval method on each approval node in the approval process; and generating an approval scheme for the approval request based at least on the approval process and the approval method indicated by the set of target approval configurations.

[0004] In a second aspect of the present disclosure, there is provided an apparatus for generating an approval scheme, the apparatus comprising: an approval content determination module, configured to determine, in response to an approval request, approval content to be approved related to the approval request; a candidate approval configuration determination module, configured to determine a plurality of candidate approval configurations related to the approval request from a predetermined approval configuration set, wherein the approval configuration in the approval configuration set at least indicates an approval method on at least one approval node; a target approval configuration determination module, configured to combine at least two candidate approval configurations from the plurality of candidate approval configurations to determine a set of target approval configurations for the approval content, wherein the set of target approval configurations indicates an approval process related to the approval content and an approval method on each approval node in the approval process; and an approval scheme determination module, configured to generate an approval scheme for the approval request based at least on the approval process and the approval method indicated by the set of target approval configurations.

[0005] In a third aspect of the present disclosure, an electronic device is provided. The device includes at least one processing unit; and at least one memory, the at least one memory is coupled to the at least one processing unit and stores instructions for execution by the at least one processing unit. When the instructions are executed by the at least one processing unit, the device executes the method of the first aspect.

[0006] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, wherein a computer program is stored on the computer-readable storage medium, and the computer program can be executed by a processor to implement the method of the first aspect.

[0007] In a fifth aspect of the present disclosure, a computer program product is provided, which includes computer executable instructions, and when the instructions are executed by a processor, the method according to the first aspect of the present disclosure is implemented.

[0008] It should be understood that the contents described in this content section are not intended to limit the key features or important features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0010] Figure 1 A schematic diagram showing an example environment according to an embodiment of the present disclosure is shown;

[0011] Figure 2 A schematic diagram showing an example of a workflow model according to some embodiments of the present disclosure;

[0012] Figure 3 A flowchart showing an example process of a method for generating an approval scheme according to some embodiments of the present disclosure;

[0013] Figure 4 A schematic diagram showing an example of determining multiple candidate approval configurations according to some embodiments of the present disclosure;

[0014] Figure 5 A schematic diagram showing an example of preprocessing approval content according to some embodiments of the present disclosure;

[0015] Figure 6 A flowchart illustrating an example process for determining a set of target approval configurations according to some embodiments of the present disclosure;

[0016] Figure 7A flowchart showing an example process of searching for at least one approval configuration for approval content from multiple groups of candidate approval configurations according to some embodiments of the present disclosure;

[0017] Figure 8 A flowchart illustrating an example process of determining ignore rules related to approving content according to some embodiments of the present disclosure;

[0018] Fig. 9 A schematic diagram showing an example of a termination rule of a tabu search algorithm according to some embodiments of the present disclosure;

[0019] Fig.10 A schematic structural block diagram of an apparatus for generating an approval scheme according to some embodiments of the present disclosure is shown; and

[0020] Fig.11 A block diagram of an electronic device is shown in which one or more embodiments of the present disclosure may be implemented. DETAILED DESCRIPTION

[0021] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0022] It should be noted that the titles of any sections / subsections provided herein are not restrictive. Various embodiments are described throughout this article, and any type of embodiment may be included under any section / subsection. In addition, the embodiments described in any section / subsection may be combined in any manner with any other embodiments described in the same section / subsection and / or different sections / subsections.

[0023] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The term "some embodiments" should be understood as "at least some embodiments". Other explicit and implicit definitions may be included below. The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may be included below.

[0024] The embodiments of the present disclosure may involve user data, data acquisition and / or use, etc. These aspects are subject to the corresponding laws, regulations and relevant provisions. In the embodiments of the present disclosure, all data collection, acquisition, processing, processing, forwarding, use, etc. are carried out on the premise that the user knows and confirms. Accordingly, when implementing each embodiment of the present disclosure, the type, scope of use, usage scenario, etc. of the data or information that may be involved should be informed to the user and the user's authorization should be obtained in an appropriate manner in accordance with the relevant laws and regulations. The specific notification and / or authorization method can vary according to the actual situation and application scenario, and the scope of the present disclosure is not limited in this respect.

[0025] In this specification and the embodiments, if personal information processing is involved, it will be processed on the premise of having a legal basis (such as obtaining the consent of the subject of personal information, or it is necessary to perform a contract, etc.), and will only be processed within the scope of regulations or agreements. If a user refuses to process personal information other than the necessary information for basic functions, it will not affect the user's use of basic functions.

[0026] As briefly described above, it is currently possible to customize approval plans for different departments or business units by writing business codes. However, this approach has a significant disadvantage: each approval plan needs to be maintained independently. Once the approval logic of an approval plan needs to be adjusted, it must be manually updated one by one in each related approval plan. This approach not only greatly increases the complexity of approval system maintenance, but is also likely to lead to differences in the approval logic of different approval plans, thereby affecting the accuracy and overall efficiency of approval.

[0027] In view of this, an embodiment of the present disclosure provides a scheme for generating an approval scheme. According to the scheme, first, in response to an approval request, the approval content to be approved related to the approval request is determined. Then, multiple candidate approval configurations related to the approval request are determined from a predetermined approval configuration set, and the approval configuration in the approval configuration set at least indicates the approval method on at least one approval node. Next, at least two candidate approval configurations among the multiple candidate approval configurations are combined to determine a set of target approval configurations for the approval content, wherein the set of target approval configurations indicates the approval process related to the approval content and the approval method on each approval node in the approval process. Finally, based at least on the approval process and approval method indicated by the set of target approval configurations, an approval scheme for the approval request is generated.

[0028] It will be more clearly understood from the description below that the scheme of the present disclosure is intended to improve the flexibility and configurability of the formulation of approval schemes. Specifically, the scheme of the present disclosure responds to approval requests and determines the relevant approval content, and then selects suitable candidate approval configurations from a predetermined approval configuration set for combination. This method can easily adapt to the unique needs of different departments and business units, and customize the optimal approval process for each approval request. In addition, the scheme of the present disclosure effectively reduces the complexity and cost of maintaining approval schemes. As described above, in traditional schemes, each approval scheme needs to be maintained independently. Once the approval logic needs to be adjusted, it must be manually updated one by one in each associated approval scheme, which is not only time-consuming and labor-intensive, but also prone to errors. The implementation of this scheme, through the centralized management of approval configurations, is conducive to achieving logical consistency, thereby greatly reducing the complexity of maintenance.

[0029] Various example implementations of the solution will be described in detail below in conjunction with the accompanying drawings.

[0030] Figure 1 A schematic diagram of an example environment 100 is shown in accordance with an embodiment of the present disclosure.

[0031] Reference Figure 1 , the example environment 100 may include an electronic device 110. As an example, the electronic device 110 may be applied to a workflow model related to approval business. A workflow model is an abstract and general description of the business rules between the workflow and its various operation steps. Workflow modeling is a description of the business logic between the overall workflow and the operation steps of each link. The essence of workflow modeling is to organize the various links in the workflow in an orderly manner according to its internal logic and rules, and then build an appropriate model in the computer to achieve automated processing. In the example environment 100, the core goal of the workflow model is to automatically transfer approval documents, approval information or approval tasks between multiple participants based on the needs of the approval business according to preset rules.

[0032] Continue to refer to Figure 1 In the example environment 100, the workflow model may include a process definition module 120, a workflow service module 111 (implemented by the electronic device 110), a workflow management module 130, an application module 140-1, an application module 140-2, other workflow service modules 150, and multiple interfaces for connecting the above modules. As an example, the application module 140-1 may be a client application, and the application module 140-2 may be an application other than the client application.

[0033] As an example, one or more interfaces in the workflow model can be implemented in the form of http interfaces. In this way, workflow integration capabilities can be easily realized, helping enterprises to host workflow services in the cloud, thereby improving work efficiency and convenience.

[0034] As an example, the workflow model may be a lightweight model written in Java language, which may comply with the Apache V2 license open source protocol to provide enterprises with flexible and efficient workflow solutions.

[0035] As an example, the workflow model can cover case management (CMMN) and decision automation (DMN) in addition to supporting business process modeling notation (BPMN). In addition, it is equipped with a powerful tool set, including modeling, task management, operation monitoring, and user management functions.

[0036] As an example, the workflow model may use Hibernate as an object-relational mapping (ORM) tool for data persistence.

[0037] As an example, the workflow model can implement rich data tables through, for example, a form engine, and support integration as a middleware pattern.

[0038] As an example, a workflow model can be constructed based on the state machine principle. The workflow constituent elements provided by such a workflow model can include steps, conditions, loops, branches, merges, etc., which is conducive to improving flexibility and controllability.

[0039] Figure 2 A schematic diagram of an example 200 of a workflow model according to some embodiments of the present disclosure is shown.

[0040] Combined with reference Figure 1 and Figure 2 As an example, the process definition module 120 is configured to allow an administrator to define one or more approval configurations to form an approval configuration set. For example, for a leave approval process, the approval configuration may indicate which steps in the leave approval process require which personnel to approve, as well as the order and conditions of the approval, etc.

[0041] As an example, the approval configuration set may be stored in the storage module 210 (e.g., a database, etc.). For example, the process definition module 120 may send information related to the newly added approval configuration to the parser 220 in response to the administrator's operation. The parser 220 parses the received information to identify the newly added approval configuration, and then stores the newly added approval configuration in the storage module 210.

[0042] As an example, the application module 140-1 (eg, client application) may be an entry for the user 230 to submit an approval request. The user 230 fills in the approval content to be approved through the application module 140-1 and starts the corresponding approval process.

[0043] As an example, the workflow service module 111 is an execution engine for approval requests. The workflow service module 111 can respond to the approval request initiated by the user 230 through the application module 140-1, and determine the candidate approval configuration from the approval configuration set in the storage module 210 with the help of the storage management module 240. Then, the workflow service module 111 can determine a set of target approval configurations based on the candidate approval configurations, and automatically generate an approval plan for the approval request based on the set of target approval configurations. In addition, the workflow service module 111 can also automatically send the approval request to the corresponding approval personnel according to the approval plan, and after obtaining the approval result, decide the next step according to the approval plan.

[0044] As an example, user 230 may initiate an approval request through an office automation system (OA), and application module 140 - 1 may implement interaction between the OA system and the workflow service module 111 , thereby sending the approval request initiated by user 230 in the OA system to the workflow service module 111 .

[0045] As an example, the workflow management module 130 can monitor and manage the approval process. The workflow management module 130 can provide the real-time status of the approval process, thereby prompting relevant personnel to handle abnormal situations in a timely manner, such as approval timeout or unreachable approval personnel.

[0046] As an example, the application module 140 - 2 (other applications except the client application) may be an application used to view approval statistics, analyze approval efficiency, etc., or an interface integrated with other enterprise systems (such as a human resources system).

[0047] As an example, other workflow service modules 150 may be modules that can provide additional services, such as notification services (sending notifications when approval status changes), archiving services (saving approval records and results), and the like.

[0048] It should be noted that the above descriptions of the modules are only exemplary. According to actual needs, the above modules can realize more functions, and the workflow model can also include more modules.

[0049] In some embodiments, the application module 140-1 can be implemented on a terminal device. The terminal device can be any type of mobile terminal, fixed terminal or portable terminal, including a mobile phone, a desktop computer, a laptop computer, a notebook computer, a netbook computer, a tablet computer, a media computer, a multimedia tablet, a personal communication system (PCS) device, a personal navigation device, a personal digital assistant (PDA), an audio / video player, a digital camera / camcorder, a positioning device, a television receiver, a radio broadcast receiver, an e-book device, a gaming device or any combination of the foregoing, including accessories and peripherals of these devices or any combination thereof. In some embodiments, the terminal device can also support any type of interface for the user 230 (such as a "wearable" circuit, etc.).

[0050] The electronic device 110 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks, and big data and artificial intelligence platforms. The electronic device 110 may include, for example, a computing system / server, such as a mainframe, an edge computing node, a computing device in a cloud environment, and the like.

[0051] The electronic device 110 and Figure 1 A communication connection may be established between the modules shown. The communication connection may be established in a wired manner or a wireless manner. The communication connection may include, but is not limited to, a Bluetooth connection, a mobile network connection, a universal serial bus connection, a wireless fidelity connection, etc., and the embodiments of the present disclosure are not limited in this respect. In the embodiments of the present disclosure, the electronic device 110 and the Figure 1 The modules shown can implement data and / or signaling interaction via a communication connection between them.

[0052] It should be understood that the structure and function of the various elements in the environment 100 are described for exemplary purposes only and do not imply any limitation on the scope of the present disclosure.

[0053] Figure 3 1 is a flowchart showing an example process 300 of a method for generating an approval scheme according to some embodiments of the present disclosure. The process 300 may be implemented at the electronic device 110. Figure 1 and Figure 2 The depiction illustrates process 300 .

[0054] The method for generating an approval scheme provided by the embodiments of the present disclosure is intended to achieve dynamic acquisition of approval configuration and automatic generation of approval schemes. In this way, the electronic device 110 can dynamically generate approval schemes with different approval levels, thereby abandoning the previous fixed two-level approval rule restrictions. This method makes it possible to process approvals under different businesses, different systems, and different rules, thereby greatly improving the adaptability and efficiency of approval processing.

[0055] Specifically, in block 310 , the electronic device determines approval content to be approved related to the approval request in response to the approval request.

[0056] As an example, the approval request may include basic information about the approval content to be approved, such as the approval type, applicant information, application time, etc. After receiving the approval request, the electronic device will parse it to extract key information. Based on the parsed results, the electronic device can determine the specific content to be approved. For example, approval requests related to documents, contracts, project plans, and budgets, etc.

[0057] In some embodiments, the electronic device may pre-process the approval content in one or more ways. Specifically, this includes two main aspects: correcting abnormal items and completing missing items. As an example, the electronic device corrects the abnormal item in response to at least one abnormal item in the approval content. As another example, the electronic device completes the missing item based on the approval reference information related to the approval request in response to the missing item in the approval content.

[0058] As examples, the above two examples can be used alone or in combination, which depends on the actual situation and requirements of the approval content, and the embodiments of the present disclosure do not limit this.

[0059] As an example, the electronic device can check whether there are any abnormal items in the approval content. These abnormal items may include data errors, format problems, or content that does not comply with regulations. Once an abnormal item is found, the electronic device will automatically correct it. For example, if an incorrect value is entered in a data field, the electronic device can automatically correct it to a reasonable value based on a preset rule or algorithm.

[0060] As an example, the electronic device can also check whether there are missing items in the approval content. Missing items may mean that some necessary information is not filled in or omitted. In this case, the electronic device can complete these missing items based on the approval reference information related to the approval request. For example, if a field is blank, but the electronic device can infer the content that should be filled in based on other relevant information, then the electronic device will automatically fill in this field.

[0061] In addition, the electronic device can perform special processing for approval requests initiated by special users. For example, the bills of lading of certain leaders may involve higher security levels or more complex approval processes, and the electronic device can identify this and process it accordingly. As an example, the processed approval request can be applied to the ignore rules in the taboo search algorithm in subsequent steps, thereby realizing a personalized approval process for the initiator.

[0062] In this way, electronic devices can process approval requests more efficiently, helping to reduce the risk of human errors and omissions, thereby improving the reliability and security of the entire approval process.

[0063] In block 320 , the electronic device determines a plurality of candidate approval configurations related to the approval request from a predetermined approval configuration set, where the approval configurations in the approval configuration set at least indicate an approval method on at least one approval node.

[0064] As an example, an approval configuration set may refer to a set of predefined approval configurations, which includes multiple possible approval configurations. The approval configurations in the approval configuration set may indicate the approval methods that should be adopted at one or more approval nodes.

[0065] As an example, an approval node can refer to a key link in the approval process, such as department manager approval, financial review, general manager approval, etc. Each approval node will have one or more corresponding approval methods, which can include online approval, offline signature, meeting resolution, etc. The approval configuration will clearly indicate which approval method should be adopted at each approval node.

[0066] In addition, the approval configuration in the approval configuration set can also indicate more content. For example, the approval configuration in the approval configuration set can also indicate the approval level (such as the number and order of approval nodes in the approval process), etc. This can be determined according to actual needs, and the embodiments of the present disclosure do not limit this.

[0067] As an example, the electronic device may analyze the approval request, such as the type of the approval request or the submitter. Then, the electronic device may search for candidate approval configurations that match the information in the approval configuration set. These candidate configurations may indicate different approval processes, approvers, approval time limits, and other elements related to the current approval request.

[0068] Figure 4 A schematic diagram of an example 400 of determining multiple candidate approval configurations according to some embodiments of the present disclosure is shown.

[0069] Reference Figure 4In some embodiments, the electronic device may determine the candidate approval configuration from the predetermined approval configuration set in one or more ways. As an example, the electronic device determines the target information related to the initiator of the approval request from the approval content. Then, in box 430, the electronic device uses a second machine learning model based on a clustering algorithm to determine one or more candidate approval configurations related to the target information from at least the predetermined approval configuration set. As another example, in box 440, the electronic device uses a second machine learning model to determine one or more candidate approval configurations related to the approval request from at least the predetermined approval configuration set.

[0070] As examples, the above two examples can be used alone or in combination, which depends on the actual situation and requirements of the approval content, and the embodiments of the present disclosure do not limit this.

[0071] As an example, when the above two examples are combined to obtain candidate approval configurations, in block 450 , the electronic device may aggregate multiple candidate approval configurations obtained in two ways to obtain multiple final candidate approval configurations.

[0072] In some embodiments, the electronic device can obtain information related to the initiator of the approval request from the approval content. This information may include the initiator's position, department, historical approval records, etc. Then, the electronic device uses a second machine learning model based on a clustering algorithm to conduct an in-depth analysis of these target information and the approval configurations in the approval configuration set. The clustering algorithm can group the approval configurations in the approval configuration set according to similarity or correlation based on the target information, thereby obtaining the candidate approval configurations that are most relevant to the target information (using the initiator's department and business unit).

[0073] In some embodiments, the electronic device may also directly use the second machine learning model to group the approval configurations in the approval configuration set according to similarity or relevance based on the initiator of the approval request, etc. This method can obtain the bottom-line approval configuration related to the approval request, thereby helping to determine a more comprehensive approval configuration.

[0074] As an example, the electronic device may process key features in the approval content and / or the approval request and key features in the approval configuration in the approval configuration set into vector features, and then implement clustering based on methods such as calculating the Euclidean distance.

[0075] As an example, the clustering algorithm uses the error square sum criterion function to evaluate the performance of clustering. Given a data set V (for example, a data set composed of the vector features described above), it contains K cluster subsets C1, C2, ..., Ck, the number of samples in each cluster is n1, n2, ..., nk, and the cluster center is M1, M2, ..., Mk. The error square sum criterion function E can be expressed by formula 1:

[0076]

[0077] As an example, the electronic device randomly selects K vector features from the data set V as the initial cluster centers M = (M1, M2, ..., Mk). For the i-th vector feature, its distance dij to the j-th cluster center is calculated and assigned to the nearest cluster. This process can be expressed by formula 2,

[0078]

[0079] Among them, v i represents the i-th vector feature in the dataset V, M j represents the jth cluster center, x i Represents the horizontal coordinate related to the i-th vector feature, y i Represents the ordinate associated with the i-th vector feature, x j Represents the horizontal coordinate related to the jth cluster center, y j represents the ordinate associated with the jth cluster center, n is the total number of vector features in the data set V, and k is the total number of cluster centers.

[0080] After completing a clustering, K clusters C = (C1, C2, ..., Ck) are formed. Then the center Mi of each cluster is recalculated. This process can be expressed by formula 3:

[0081]

[0082] Next, the electronic device determines whether the cluster center has changed. If it has changed, the previous steps are repeated; otherwise, the current clustering result is determined to be the final clustering result. Based on the clustering result, multiple candidate approval configurations related to the current approval request can be determined.

[0083] In this way, the electronic device can initially screen the approval configurations in the approval configuration set, which is beneficial to reducing the search scope of the subsequent taboo search algorithm, thereby speeding up the processing speed of the taboo search algorithm.

[0084] In block 330, the electronic device combines at least two candidate approval configurations from the plurality of candidate approval configurations to determine a set of target approval configurations for the approval content, wherein the set of target approval configurations indicates an approval process related to the approval content and an approval method at each approval node in the approval process.

[0085] In some embodiments, the electronic device screens multiple candidate approval configurations based on the supplementary information, wherein the supplementary information at least indicates the geographical information related to the initiator of the approval request. Then, the electronic device combines at least two of the screened multiple candidate approval configurations to determine a set of target approval configurations for the approval content.

[0086] In some embodiments, the electronic device may further screen the previously determined multiple candidate approval configurations based on supplementary information. Such supplementary information is essential for accurately matching the approval configurations, and includes at least geographical information related to the initiator of the approval request. The geographical information may involve the country, region, city, or specific work location of the initiator.

[0087] After obtaining this supplementary information, the electronic device will compare and analyze it with the candidate approval configurations. Based on the analysis results, the electronic device can identify which approval configurations match or are most suitable for the initiator's regional information. For example, some regions may have specific approval requirements or restrictions, which will be taken into account during the screening process.

[0088] After the screening is completed, the electronic device will provide the screened candidate approval configurations to the first machine learning model based on the taboo search algorithm to be mentioned below. The first machine learning model will, based on these inputs, find one or more approval configurations that are most relevant to the approval request as a set of target approval configurations.

[0089] As an example, after the screening is completed, the electronic device can add the screened candidate approval configurations to a taboo table of the taboo search algorithm. The taboo table records which approval configurations will be restricted for selection in the taboo search algorithm.

[0090] In this way, the electronic device significantly optimizes the input content that will be provided to the first machine learning model, thereby helping to increase the processing speed of the taboo search algorithm and speed up the determination of a set of target approval configurations.

[0091] Figure 5 A schematic diagram of an example 500 of pre-processing approval content according to some embodiments of the present disclosure is shown.

[0092] Reference Figure 5In some embodiments, the electronic device may pre-process the approval content to extract key features required by the first machine learning model. These key features may be, for example, information related to the approval level in the approval content.

[0093] The following takes the process of determining the approval level as an example. As an example, in box 501, the electronic device obtains information related to the initiator of the approval request from the approval content.

[0094] In block 502 , the electronic device determines whether the approval level indicated by the information related to the initiator of the approval request is greater than a pre-configured maximum approval level related to the approval request.

[0095] If the approval level indicated by the information related to the initiator of the approval request is greater than the pre-configured maximum approval level related to the approval request, the electronic device uses the pre-configured maximum approval related to the approval request as a candidate approval level in block 503. If the approval level indicated by the information related to the initiator of the approval request is less than the pre-configured maximum approval level related to the approval request, the electronic device uses the approval level indicated by the information related to the initiator of the approval request as a candidate approval level in block 504.

[0096] In block 505 , the electronic device determines whether the initiator of the approval request is a special initiator.

[0097] If the initiator of the approval request is a special initiator, in block 506, the electronic device simplifies (or prunes) the candidate approval levels based on a method related to the special initiator to reduce the approval levels. Then in block 507, the electronic device determines the target approval level based on the simplified candidate approval levels. If the initiator of the approval request is not a special initiator, block 508 is executed, and the electronic device determines the target approval level based on the current candidate approval levels.

[0098] After the target approval hierarchy is determined, this information will be used as part of the input to the first machine learning model to participate in the process of determining a set of target approval configurations.

[0099] In this way, the electronic device performs various optimization processes on the input content of the first machine learning model (such as the supplementation, correction, and pruning described above), which is beneficial to improving the calculation accuracy and speed of the first machine learning model.

[0100] In some embodiments, the electronic device, based on the approval content, uses a first machine learning model based on a taboo search algorithm to iteratively perform the following process to determine a set of target approval configurations for the approval content: determining at least one set of approval configurations for the approval content from multiple candidate approval configurations.

[0101] The taboo search algorithm is mainly used to solve integer programming problems. Its core idea is to avoid repeated searches of the solution space that has been explored, thereby preventing falling into local optimality and achieving global search. The embodiment of the present disclosure introduces a taboo table to store the optimal approval configurations that have been searched, and ignores these approval configurations in subsequent searches.

[0102] As an example, in the tabu search algorithm, the following is first defined.

[0103] Local minimum solution: In a given neighborhood, if the evaluation values ​​of all adjacent solutions are not lower than the current solution, the current solution is called the local minimum solution.

[0104] Global minimum solution: In the entire solution space, if the evaluation values ​​of all solutions are not lower than a certain solution, then the solution is called the global minimum solution.

[0105] Adaptive value function: used to evaluate the quality of the solution during the search process. The selection of new solutions is based on the taboo criterion and the contempt criterion.

[0106] Taboo objects: refer to elements added to the taboo table to avoid repeated searches.

[0107] Changes in the solution: include the overall change and component changes of the solution vector, as well as the resulting changes in the target value.

[0108] Taboo length: defines the maximum number of iterations that a taboo object can be prohibited from searching without being violated by the defiance criteria.

[0109] Candidate solutions: To improve efficiency, when the solution space is large, only some excellent solutions are selected as candidate solutions for calculation.

[0110] Disregarding the rules: Used to break taboo restrictions in certain situations to ensure that good solutions are not missed.

[0111] Neighborhood function: used to generate a new solution space.

[0112] Termination criteria: including the maximum number of iterations, the maximum taboo frequency, and the degree of deviation of the adaptation value, etc., are used to determine when the algorithm terminates.

[0113] Figure 6 A flow chart is shown of an example process 600 of determining a set of target approval configurations according to some embodiments of the present disclosure.

[0114] Reference Figure 6 The following is an exemplary description of the process of determining a set of target approval configurations using a taboo search algorithm in an embodiment of the present disclosure.

[0115] In box 601, the electronic device selects an initial set of candidate approval configurations from multiple candidate approval configurations (for clarity, this set of candidate approval configurations will also be referred to as an initial solution below) as a starting point. This solution can undergo changes in the whole or in components to explore different solution spaces. For example, let the solution vector of the initial solution be (x1,…,xi-1,x,xi+1,…,xn), and the most basic change of the vector components is (x1,…,xi,xi,…xn)→(x1,…,xi-1,yi,…,xn), that is, only the i-th component changes. In addition, it also includes the situation where multiple components change.

[0116] Figure 7 A flowchart of an example process 700 of searching for at least one approval configuration for approval content from multiple groups of candidate approval configurations according to some embodiments of the present disclosure is shown.

[0117] In some embodiments, the electronic device may use additional judgment logic for approval content including special purposes, etc., to search for a more accurate approval configuration. Figure 7 , taking approval content with target purposes as an example.

[0118] In block 701 , the electronic device determines whether the approval content has a target purpose.

[0119] If there is a target purpose (e.g., a purchasing purpose), in box 702, the electronic device selects one or more candidate approval configurations from multiple candidate approval configurations related to the target purpose (the multiple candidate approval configurations here specifically refer to the multiple candidate approval configurations after the clustering results are screened again as described above) based on the selection rules related to the target purpose as the initial solution (or part of the solution) of the current iteration. If there is no target purpose, in box 703, the electronic device selects one or more candidate approval configurations from multiple candidate approval configurations related to the predetermined purpose (e.g., no purchasing path) (the multiple candidate approval configurations here specifically refer to the multiple candidate approval configurations after the clustering results are screened again as described above) as the initial solution of the current iteration.

[0120] As an example, the selection rules related to the target purpose (e.g., procurement purpose) may include: if multiple candidate approval configurations include all approval elements required for approval, then select the candidate approval configuration with the smallest amount. If multiple candidate approval configurations include some approval elements required for approval, then select the candidate approval configuration with the smallest amount for procurement purpose and the candidate approval configuration with the smallest amount for no procurement purpose, and so on.

[0121] In block 704 , the electronic device determines whether there is a candidate approval configuration among the plurality of candidate approval configurations that is related to the target usage and satisfies the selection rule.

[0122] If there is a candidate approval configuration that satisfies the selection rule among the multiple candidate approval configurations related to the target purpose, the candidate approval configuration is returned as an initial solution (or part of the solution) in block 705. If there is no candidate approval configuration that satisfies the selection rule among the multiple candidate approval configurations related to the target purpose, the electronic device determines in block 706 whether the multiple candidate approval configurations related to the target purpose are empty.

[0123] As an example, if the multiple candidate approval configurations related to the target purpose are empty, the electronic device can execute block 703. If the multiple candidate approval configurations related to the target purpose are not empty, in block 707, the electronic device determines whether there is a candidate approval configuration related to the target purpose and meeting the selection rule among the multiple candidate approval configurations before screening.

[0124] If there is a candidate approval configuration related to the target purpose and meeting the selection rule among the plurality of candidate approval configurations before screening, block 705 may be executed. If there is no candidate approval configuration related to the target purpose and meeting the selection rule among the plurality of candidate approval configurations before screening, the electronic device executes block 703.

[0125] As an example, in block 708, the electronic device determines whether there is a candidate approval configuration applicable to the current approval content among the multiple candidate approval configurations related to the intended purpose (e.g., no procurement channel). If there is a candidate approval configuration applicable to the current approval content among the multiple candidate approval configurations related to the intended purpose, block 705 is executed. If there is no candidate approval configuration applicable to the current approval content among the multiple candidate approval configurations related to the intended purpose, in block 709, the electronic device searches for a candidate approval configuration related to the intended purpose and satisfying the selection rule among the multiple candidate approval configurations before screening.

[0126] In some embodiments, the first machine learning model updates the taboo table related to the taboo search algorithm in each iteration. As an example, the electronic device obtains at least one set of first approval configurations for the approval content determined by the first machine learning model before the current iteration. Then, the electronic device changes at least one first approval configuration in the at least one set of first approval configurations to obtain at least one set of second approval configurations.

[0127] Return to reference Figure 6 In block 602, the electronic device determines one or more new groups of approval configurations (e.g., at least one group of second approval configurations) adjacent to the initial group of candidate approval configurations (e.g., a group of first approval configurations) through a neighborhood function. These adjacent new groups of approval configurations constitute neighborhood solutions of the initial solution to explore a wider solution space.

[0128] Then, the electronic device evaluates the relevance of a set of first approval configurations and at least one set of second approval configurations to the approval content through an adaptive value function, and selects a candidate optimal solution for the current iteration from the set of first approval configurations and at least one set of second approval configurations.

[0129] As an example, through the taboo table, the electronic device can take the optimal solutions determined in the iterations before the current iteration (for example, a set of candidate approval configurations that have been determined as the optimal solution) as taboo objects and add them to the taboo table. In this way, in the current iteration, the electronic device will avoid repeatedly searching for these solutions, thereby preventing the algorithm from falling into a local optimum.

[0130] In addition, in order to ensure the real-time performance of the taboo table, the electronic device also updates the taboo table in the current iteration based on at least one set of first approval configurations and at least one set of second approval configurations.

[0131] In some embodiments, the electronic device screens at least one set of first approval configurations and at least one set of second approval configurations based on at least the taboo table in the current iteration to obtain at least one set of candidate restriction approval configurations. Then, the electronic device determines a set of target restriction approval configurations from the at least one set of candidate restriction approval configurations based on relevance information related to the approval content of the at least one set of candidate restriction approval configurations. Next, the electronic device updates the set of target restriction approval configurations to the taboo table in the current iteration.

[0132] In some embodiments, the electronic device filters at least one set of first approval configurations and at least one set of second approval configurations based on the taboo table in the current iteration and the ignore rule related to the approval content, wherein the ignore rule indicates that in the screening process of the at least one set of first approval configurations and the at least one set of second approval configurations, one or more approval configurations in the at least one set of first approval configurations and the at least one set of second approval configurations ignore the restrictions of the taboo table.

[0133] Return to reference Figure 6 As an example, in box 603, the electronic device determines whether there is an ignore rule related to the approval content.

[0134] Figure 8 A flow chart is shown of an example process 800 for determining ignore rules related to approving content, according to some embodiments of the present disclosure.

[0135] Reference Figure 8 In box 801, the electronic device can determine whether the approval content meets the ignore condition based on information such as the approval purpose in the approval content.

[0136] If the approval content meets the ignore condition, the electronic device determines whether there is an ignore rule related to the approval content in block 802. As an example, if the approval content does not meet the ignore condition, the electronic device may end process 800 in block 803.

[0137] As an example, if there is an ignore rule related to the approval content, in block 804, the electronic device determines whether the approval level related to the approval content meets the minimum approval level. If there is an ignore rule related to the approval content, block 803 may be executed.

[0138] As an example, if the approval level associated with the approval content meets the minimum approval level, in block 805, it is determined that the ignore rule associated with the approval content is valid, thereby allowing at least part of the approval configuration to ignore the restrictions of the taboo table. If the approval level associated with the approval content does not meet the minimum approval level, block 803 is executed.

[0139] Return to reference Figure 6 , if there is an ignore rule related to the approval content, in box 604, the electronic device makes one or more approval configurations in at least one set of first approval configurations and at least one set of second approval configurations ignore the restrictions of the taboo table. For example, the electronic device can directly use the candidate optimal solution described above as the optimal solution for the current iteration. If there is no ignore rule related to the approval content, in box 605, the electronic device uses the group of approval configurations in at least one set of first approval configurations and at least one set of second approval configurations that are not in the taboo table and have the highest evaluation as the optimal solution for the current iteration.

[0140] In some embodiments, the electronic device determines a first feature representation related to the approval content. Then, the electronic device determines at least one second feature representation related to at least one set of candidate restriction approval configurations. Next, the electronic device determines relevance information based on the first feature representation and the at least one second feature representation. Then, the electronic device determines a set of target restriction approval configurations from at least one set of candidate restriction approval configurations based on the relevance information.

[0141] In some embodiments, the electronic device may analyze a first feature representation that is closely related to the approval content. Subsequently, it determines a second feature representation associated with multiple sets of candidate restriction approval configurations. Based on these two feature representations, the electronic device calculates the correlation information between them. Using this information, the device accurately selects a set of the most suitable target restriction approval configurations from multiple sets of candidate configurations.

[0142] In some embodiments, the first and second feature representations may exist in the form of multidimensional vectors. For example, the dimensions of the first feature representation may cover key information such as the department, business unit, amount, group, and approver of the approval initiator. Similarly, each dimension in these feature representations is usually encoded using natural numbers for ease of processing and calculation. In addition, the system also supports flexible path cutting based on preset constraints to meet the actual needs of different approval scenarios.

[0143] In some embodiments, the first feature representation and the second feature representation may be multidimensional vector features. As an example, in the first feature representation, each dimension may represent the department, business unit, amount, group, approver, etc. of the initiator of the approval content. As an example, each dimension in the first feature representation and the second feature representation is encoded with natural numbers. And path cutting may be allowed according to constraints.

[0144] In some embodiments, the electronic device modifies the first feature representation based on first weight information related to the first feature representation, wherein the first weight information indicates a weight value of at least one element in the first feature representation. Then, the electronic device modifies at least one second feature representation based on second weight information related to each of the at least one second feature representations, wherein the second weight information indicates a weight value of at least one element in the second feature representation related thereto. Next, the electronic device determines the correlation information based on the similarity between the modified first feature representation and the modified at least one second feature representation.

[0145] In some embodiments, the electronic device may first modify the first feature representation based on first weight information associated with the first feature representation, where the weight information specifically indicates the weight value of each element in the first feature representation. Subsequently, the device may also use second weight information associated with multiple second feature representations to make similar modifications to these second feature representations, where the second weight information reflects the weights of the elements in each second feature representation. Next, the electronic device may determine the correlation information between them based on the similarity between the modified first feature representation and the modified second feature representation. In this way, weight information is innovatively introduced in the process of calculating the similarity. In this way, the relevance of more appropriate approval configurations can be improved, thereby accelerating the convergence speed of the algorithm.

[0146] Return to reference Figure 6 As an example, in block 606, the electronic device uses the optimal solution of the current iteration as a set of target restriction approval configurations and adds them to the taboo table.

[0147] After each iteration, in box 607, the electronic device determines whether the termination criteria are met, such as reaching the maximum number of iteration steps, the maximum taboo frequency, the degree of deviation of the adaptation value is less than a certain threshold, or finding the optimal solution that meets specific conditions. If the termination criteria are met, in box 608, the electronic device determines a set of candidate approval configurations indicated by the optimal solution with the highest score as a set of target approval configurations. If the termination criteria are not met, in box 609, the electronic device performs the next iteration.

[0148] Fig. 9 A schematic diagram showing an example of a termination rule 900 for a tabu search algorithm according to some embodiments of the present disclosure is shown.

[0149] Reference Fig. 9 In box 901, the electronic device determines whether a set of candidate approval configurations indicated by the optimal solution determined in the current iteration includes all approval elements for the approval content, that is, whether this set of candidate approval configurations can successfully complete the approval of the approval content (for example, the maximum approval amount indicated in a set of candidate approval configurations is greater than or equal to the amount required for approval in the approval content).

[0150] If the set of candidate approval configurations indicated by the optimal solution determined in the current iteration includes all approval elements for the approval content, it is determined in block 902 that the iteration is terminated. Otherwise, in block 903, the next iteration is executed.

[0151] In some embodiments, during the above process, the electronic device may encounter a local minimum solution, that is, within a given neighborhood, the evaluation values ​​of all adjacent solutions are not lower than the current solution. However, the goal of the electronic device is to find a global minimum solution, that is, in the entire solution space, the evaluation values ​​of all solutions are not lower than the solution. Therefore, the electronic device needs to constantly jump out of the local minimum solution and continue searching.

[0152] In this way, with the help of the taboo search algorithm, the electronic device can quickly screen and determine a set of target approval configurations, thereby greatly improving the efficiency and accuracy of configuration determination. At the same time, using the ignore rule, the limitations of the taboo table can be broken to ensure that the approval configuration in a good state is not missed.

[0153] In some embodiments, the above process should be able to meet the following four constraints: first, each approval request can obtain the corresponding approval configuration; second, each configuration can only be used once by the same approval request; third, each approval request has a maximum limit on the approval level; fourth, a single approval request is subject to different approval amount constraints.

[0154] In block 340 , the electronic device generates an approval solution for the approval request based at least on the approval process and approval method indicated by a set of target approval configurations.

[0155] In some embodiments, the electronic device intelligently generates customized approval solutions for various approval requests based on the approval process and method indicated by one or more groups of target approval configurations. At the same time, the electronic device can correct the information in the generated approval solution through a post-processor.

[0156] It can be clearly understood from the various embodiments described above that the embodiments of the present disclosure are intended to improve the flexibility and configurability of the formulation of approval schemes. Specifically, the embodiments of the present disclosure respond to approval requests and determine the relevant approval content, and then select suitable candidate approval configurations from a predetermined approval configuration set for combination. This method can easily adapt to the unique needs of different departments and business units, and customize the optimal approval process for each approval request. In addition, the embodiments of the present disclosure effectively reduce the complexity and cost of maintaining approval schemes. As described above, in traditional schemes, each approval scheme needs to be maintained independently. Once the approval logic needs to be adjusted, it must be manually updated one by one in each associated approval scheme, which is not only time-consuming and labor-intensive, but also prone to errors. The embodiments of the present disclosure, by centrally managing approval configurations, are conducive to achieving logical consistency, thereby greatly reducing the complexity of maintenance.

[0157] The embodiments of the present disclosure also provide corresponding devices for implementing the above methods or processes. Fig.10 A schematic structural block diagram of an apparatus 1000 for generating an approval scheme according to some embodiments of the present disclosure is shown. The apparatus 1000 may be implemented as or included in an electronic device 110. Each module / component in the apparatus 1000 may be implemented by hardware, software, firmware, or any combination thereof.

[0158] Reference Fig.10 , the device 1000 includes an approval content determination module 1010, a candidate approval configuration determination module 1020, a target approval configuration determination module 1030, and an approval scheme determination module 1040. In some embodiments, the approval content determination module 1010 is configured to determine the approval content to be approved related to the approval request in response to the approval request. The candidate approval configuration determination module 1020 is configured to determine a plurality of candidate approval configurations related to the approval request from a predetermined approval configuration set, wherein the approval configurations in the approval configuration set at least indicate an approval method on at least one approval node. The target approval configuration determination module 1030 is configured to combine at least two candidate approval configurations from the plurality of candidate approval configurations to determine a set of target approval configurations for the approval content, wherein the set of target approval configurations indicates an approval process related to the approval content and an approval method on each approval node in the approval process. The approval scheme determination module 1040 is configured to generate an approval scheme for the approval request based at least on the approval process and the approval method indicated by the set of target approval configurations.

[0159] In some embodiments, the target approval configuration determination module 1030 is further configured to: based on the approval content, iteratively perform the following process using a first machine learning model based on a taboo search algorithm to determine a set of target approval configurations for the approval content: determine at least one set of approval configurations for the approval content from multiple candidate approval configurations.

[0160] In some embodiments, the first machine learning model updates the taboo table related to the taboo search algorithm in each iteration. And the target approval configuration determination module 1030 is further configured to: obtain at least one set of first approval configurations for approval content determined by the first machine learning model before the current iteration; change at least one first approval configuration in the at least one set of first approval configurations to obtain at least one set of second approval configurations; and update the taboo table in the current iteration based on the at least one set of first approval configurations and the at least one set of second approval configurations.

[0161] In some embodiments, the target approval configuration determination module 1030 is further configured to: screen at least one group of first approval configurations and at least one group of second approval configurations based at least on the taboo table in the current iteration to obtain at least one group of candidate restriction approval configurations; determine a group of target restriction approval configurations from at least one group of candidate restriction approval configurations based on relevance information related to the approval content of at least one group of candidate restriction approval configurations; and update a group of target restriction approval configurations to the taboo table in the current iteration.

[0162] In some embodiments, the target approval configuration determination module 1030 is further configured to: screen at least one group of first approval configurations and at least one group of second approval configurations based on the taboo table in the current iteration and the ignore rule related to the approval content, wherein the ignore rule indicates that in the screening process of the at least one group of first approval configurations and the at least one group of second approval configurations, one or more approval configurations in the at least one group of first approval configurations and the at least one group of second approval configurations ignore the restrictions of the taboo table.

[0163] In some embodiments, the target approval configuration determination module 1030 is further configured to: determine a first feature representation related to the approval content; determine at least one second feature representation related to at least one set of candidate restriction approval configurations; determine relevance information based on the first feature representation and at least one second feature representation; and determine a set of target restriction approval configurations from at least one set of candidate restriction approval configurations based on the relevance information.

[0164] In some embodiments, the target approval configuration determination module 1030 is further configured to: modify the first feature representation based on first weight information related to the first feature representation, wherein the first weight information indicates the weight value of at least one element in the first feature representation; modify at least one second feature representation based on second weight information related to each of the at least one second feature representations, wherein the second weight information indicates the weight value of at least one element in the second feature representation related thereto; and determine the relevance information based on the similarity between the modified first feature representation and the modified at least one second feature representation.

[0165] In some embodiments, the candidate approval configuration determination module 1020 is further configured to: determine target information related to the initiator of the approval request from the approval content, and use a second machine learning model based on a clustering algorithm to determine one or more candidate approval configurations related to the target information from at least a predetermined approval configuration set; and / or use a second machine learning model to determine one or more candidate approval configurations related to the approval request from at least a predetermined approval configuration set.

[0166] In some embodiments, the target approval configuration determination module 1030 is further configured to: filter multiple candidate approval configurations based on supplementary information, wherein the supplementary information at least indicates geographical information related to the initiator of the approval request; and combine at least two of the filtered multiple candidate approval configurations to determine a set of target approval configurations for the approval content.

[0167] In some embodiments, the approval content determination module 1010 is further configured to: in response to at least one abnormal item in the approval content, correct the abnormal item; or in response to missing items in the approval content, complete the missing items based on approval reference information related to the approval request.

[0168] Fig.11 1 shows a block diagram of an electronic device 1100 in which one or more embodiments of the present disclosure may be implemented. The electronic device 1100 may be used to implement, for example, Figure 1 The electronic device 110 shown. It should be understood that Fig.11 The electronic device 1100 shown is merely exemplary and should not constitute any limitation on the functionality and scope of the embodiments described herein.

[0169] Reference Fig.11, the electronic device 1100 is in the form of a general electronic device. The components of the electronic device 1100 may include, but are not limited to, one or more processors or processing units 1110, a memory 1120, a storage device 1130, one or more communication units 1140, one or more input devices 1150, and one or more output devices 1160. The processing unit 1110 may be an actual or virtual processor and is capable of performing various processes according to a program stored in the memory 1120. In a multi-processor system, multiple processing units execute computer executable instructions in parallel to improve the parallel processing capability of the electronic device 1100.

[0170] The electronic device 1100 typically includes a plurality of computer storage media. Such media may be any available media accessible to the electronic device 1100, including but not limited to volatile and non-volatile media, removable and non-removable media. The memory 1120 may be a volatile memory (e.g., a register, a cache, a random access memory (RAM)), a non-volatile memory (e.g., a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. The storage device 1130 may be a removable or non-removable medium, and may include a machine-readable medium, such as a flash drive, a disk, or any other medium, which may be capable of being used to store information and / or data and may be accessed within the electronic device 1100.

[0171] The electronic device 1100 may further include additional removable / non-removable, volatile / non-volatile storage media. Fig.11 As shown in , a disk drive for reading or writing from a removable, non-volatile disk (e.g., a "floppy disk") and an optical drive for reading or writing from a removable, non-volatile optical disk may be provided. In these cases, each drive may be connected to the bus (not shown) by one or more data media interfaces. Memory 1120 may include a computer program product 1125 having one or more program modules that are configured to perform various methods or actions of various embodiments of the present disclosure.

[0172] The communication unit 1140 implements communication with other electronic devices through a communication medium. Additionally, the functions of the components of the electronic device 1100 can be implemented in a single computing cluster or multiple computing machines that can communicate through a communication connection. Therefore, the electronic device 1100 can operate in a networked environment using a logical connection with one or more other servers, a network personal computer (PC), or another network node.

[0173] The input device 1150 may be one or more input devices, such as a mouse, a keyboard, a tracking ball, etc. The output device 1160 may be one or more output devices, such as a display, a speaker, a printer, etc. The electronic device 1100 may also communicate with one or more external devices (not shown) through the communication unit 1140 as needed, such as a storage device, a display device, etc., communicate with one or more devices that allow a user to interact with the electronic device 1100, or communicate with any device that allows the electronic device 1100 to communicate with one or more other electronic devices (e.g., a network card, a modem, etc.). Such communication may be performed via an input / output (I / O) interface (not shown).

[0174] According to an exemplary implementation of the present disclosure, a computer-readable storage medium is provided, on which computer-executable instructions are stored, wherein the computer-executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the method described above.

[0175] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices, equipment, and computer program products implemented according to the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0176] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0177] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, so that the instructions executed on the computer, other programmable data processing apparatus, or other device implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0178] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple implementations of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of a module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some implementations as replacements, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be realized by a special hardware-based system that performs the function or action of the specification, or can be realized by a combination of special hardware and computer instructions.

[0179] The above descriptions of various implementations of the present disclosure are exemplary, non-exhaustive, and not limited to the disclosed implementations. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described implementations. The determination of the terms used herein is intended to best explain the principles of the implementations, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the various implementations disclosed herein.

Claims

1. A method for generating an approval scheme, comprising: In response to the approval request, determining the approval content to be approved related to the approval request; Determining a plurality of candidate approval configurations related to the approval request from a predetermined approval configuration set, wherein the approval configurations in the approval configuration set at least indicate an approval method on at least one approval node; Combining at least two candidate approval configurations among the multiple candidate approval configurations to determine a set of target approval configurations for the approval content, wherein the set of target approval configurations indicates an approval process related to the approval content and an approval method at each approval node in the approval process; as well as An approval plan for the approval request is generated based at least on the approval process and the approval method indicated by the set of target approval configurations.

2. The method according to claim 1, wherein combining at least two candidate approval configurations among the plurality of candidate approval configurations to determine a set of target approval configurations for the approval content comprises: Based on the approval content, the following process is iteratively performed using a first machine learning model based on a taboo search algorithm to determine a set of target approval configurations for the approval content: At least one group of approval configurations for the approval content is determined from the multiple candidate approval configurations.

3. The method according to claim 2, wherein the first machine learning model updates a tabu table related to the tabu search algorithm in each iteration; and Wherein updating the taboo table comprises: Obtain at least one set of first approval configurations for the approval content determined by the first machine learning model before the current iteration; Changing at least one first approval configuration in the at least one set of first approval configurations to obtain at least one set of second approval configurations; and The taboo table in the current iteration is updated based on the at least one set of first approval configurations and the at least one set of second approval configurations.

4. The method according to claim 3, wherein updating the taboo table in the current iteration process based on the at least one set of first approval configurations and the at least one set of second approval configurations comprises: Based at least on the taboo table in the current iteration, the at least one set of first approval configurations and the at least one set of second approval configurations are screened to obtain at least one set of candidate restricted approval configurations; determining a set of target restriction approval configurations from the at least one set of candidate restriction approval configurations based on relevance information related to the approval content of the at least one set of candidate restriction approval configurations; as well as The set of target restriction approval configurations is updated into the taboo table in the current iteration.

5. The method according to claim 4, wherein the at least one set of first approval configurations and the at least one set of second approval configurations are screened based on at least the taboo table in the current iteration, comprising: Based on the taboo table in the current iteration and the ignore rule related to the approval content, the at least one group of first approval configurations and the at least one group of second approval configurations are screened, wherein the ignore rule indicates that in the screening process of the at least one group of first approval configurations and the at least one group of second approval configurations, one or more approval configurations of the at least one group of first approval configurations and the at least one group of second approval configurations ignore the restriction of the taboo table.

6. The method according to claim 4, wherein determining a set of target restriction approval configurations from the at least one set of candidate restriction approval configurations based on relevance information related to the at least one set of candidate restriction approval configurations for the approval content comprises: Determining a first characteristic representation related to the approval content; determining at least one second characteristic representation associated with the at least one set of candidate restricted approval configurations; determining the relevance information based on the first feature representation and the at least one second feature representation; as well as Based on the relevance information, the set of target restriction approval configurations is determined from the at least one set of candidate restriction approval configurations.

7. The method according to claim 6, wherein determining the relevance information based on the first feature representation and the at least one second feature representation comprises: modifying the first feature representation based on first weight information related to the first feature representation, wherein the first weight information indicates a weight value of at least one element in the first feature representation; Modifying the at least one second feature representation based on second weight information associated with each of the at least one second feature representations, wherein the second weight information indicates a weight value of at least one element in the second feature representation associated therewith; as well as The relevance information is determined based on a similarity between the modified first feature representation and the modified at least one second feature representation.

8. The method according to claim 1, wherein determining a plurality of candidate approval configurations related to the approval request from a predetermined approval configuration set comprises: Determine target information related to the initiator of the approval request from the approval content, and determine one or more candidate approval configurations related to the target information from at least the predetermined approval configuration set using a second machine learning model based on a clustering algorithm; and / or Using the second machine learning model, determine one or more candidate approval configurations related to the approval request from at least the predetermined approval configuration set.

9. The method according to claim 1, wherein combining at least two candidate approval configurations among the plurality of candidate approval configurations to determine a set of target approval configurations for the approval content comprises: Screening the plurality of candidate approval configurations based on supplementary information, wherein the supplementary information at least indicates regional information related to an initiator of the approval request; as well as At least two of the screened candidate approval configurations are combined to determine a set of target approval configurations for the approval content.

10. The method according to claim 1, wherein in response to the approval request, determining the approval content to be approved related to the approval request comprises at least one of the following: In response to at least one abnormal item in the approval content, the abnormal item is corrected; or In response to missing items in the approval content, the missing items are completed based on approval reference information related to the approval request.

11. A device for generating an approval scheme, comprising: An approval content determination module, configured to determine approval content to be approved related to the approval request in response to the approval request; A candidate approval configuration determination module is configured to determine a plurality of candidate approval configurations related to the approval request from a predetermined approval configuration set, wherein the approval configurations in the approval configuration set at least indicate an approval method on at least one approval node; a target approval configuration determination module, configured to combine at least two of the plurality of candidate approval configurations to determine a set of target approval configurations for the approval content, wherein the set of target approval configurations indicates an approval process related to the approval content and an approval method at each approval node in the approval process; as well as The approval scheme determination module is configured to generate an approval scheme for the approval request based at least on the approval process and the approval method indicated by the set of target approval configurations.

12. An electronic device comprising: at least one processing unit; as well as At least one memory, the at least one memory is coupled to the at least one processing unit and stores instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the electronic device to perform the method according to any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program stored thereon, wherein the computer program can be executed by a processor to implement the method according to any one of claims 1 to 10.

14. A computer program product comprising computer executable instructions, wherein the computer executable instructions, when executed by a processor, implement the method according to any one of claims 1 to 10.

Images