An AI Financial Task Allocation Method and System Based on an Accounting Knowledge Graph

Abstract

The present invention relates to the technical field of financial data processing, and discloses an AI financial task allocation method based on an accounting knowledge graph. The method includes obtaining experience relevance data, skill relevance data, professional matching data of financial personnel, and enterprise financial business requirements; generating an accounting form graph according to the enterprise financial business requirements, and performing form integration and summarization operations to obtain a complete set of accounting forms; calculating based on the analytic hierarchy process to obtain a task matching degree; analyzing the task capabilities of financial personnel according to the task matching degree, and performing intelligent allocation operations to obtain a task allocation form; performing data entry and node feedback operations according to the complete set of accounting forms and the task allocation form to obtain a real-time task form; performing progress analysis and task reallocation operations according to the real-time task form to obtain a final task plan form. This method can realize intelligent allocation of tasks, improve the accounting efficiency, and reduce the error rate.

Description

Technical Field

[0001] The present invention relates to the field of financial data processing, and particularly to an AI financial task allocation method and system based on an accounting knowledge graph. Background Art

[0002] At present, financial management, as an important part of economic activities, records and analyzes economic activities through accounting and supervision to provide data support for decision-making. With the digital and intelligent development of the financial business process, higher requirements are put forward for the efficiency and accuracy of financial data processing. The rapid development of artificial intelligence technology has gradually deepened its application in the financial field, such as accounting robots, intelligent auditing, and intelligent processing of financial transactions. These technologies can simulate human intelligent thinking and effectively improve data processing capabilities. In addition, as a tool for constructing a knowledge network and revealing data relationships, the knowledge graph has become an important basis for realizing the application of artificial intelligence technology in the industry.

[0003] In an existing technology, the allocation of financial data relies on manual task decomposition and allocation operations. By manually analyzing financial scenarios and task requirements, tasks are allocated to financial personnel for execution. This method is feasible in small-scale financial scenarios, but in the case of complex tasks and large amounts of data, the efficiency of manual allocation is reduced, and data allocation errors are likely to occur due to human errors.

[0004] There are problems in the prior art such as low allocation efficiency, high error rate, and low intelligent allocation efficiency of tasks. Summary of the Invention

[0005] The present invention provides an AI financial task allocation method and system based on an accounting knowledge graph to achieve intelligent allocation of tasks, improve accounting efficiency, and reduce the error rate.

[0006] In a first aspect, to solve the above technical problems, the present invention provides an AI financial task allocation method based on an accounting knowledge graph, including:

[0007] Obtaining experience relevance data, skill relevance data, professional matching data of financial personnel, and enterprise financial business requirements;

[0008] Generating an accounting form graph according to the enterprise financial business requirements, and performing form integration and summary operations according to the accounting form graph to obtain a complete set of accounting forms;

[0009] Based on the analytic hierarchy process, calculating a matching degree according to the complete set of accounting forms, the experience relevance data, the skill relevance data, and the professional matching data to obtain a task matching degree;

[0010] Analyze the task capabilities of financial personnel according to the task matching degree, and perform intelligent allocation operations to obtain a task allocation table;

[0011] According to the complete set of accounting tables and the task allocation table, perform data entry and node feedback operations to obtain a real-time task table;

[0012] According to the real-time task table, perform progress analysis and task reallocation operations to obtain a final task schedule.

[0013] Preferably, according to the enterprise's financial business requirements, generate an accounting table atlas, and perform form integration and summary operations according to the accounting table atlas to obtain a complete set of accounting tables, including:

[0014] Determine the financial scenario according to the enterprise's financial business requirements;

[0015] According to the financial scenario, perform accounting node definition operations to obtain accounting nodes;

[0016] Configure the required form templates and data formats for the accounting nodes to obtain an accounting table atlas;

[0017] Perform associated form integration operations on the accounting nodes and corresponding associated forms defined in the accounting table atlas, and combine all node forms in the same process into a process accounting table set to obtain a process accounting table set;

[0018] According to the financial scenario and the accounting nodes, perform summary integration operations on the process accounting tables to obtain a complete set of accounting tables.

[0019] Preferably, based on the analytic hierarchy process, according to the complete set of accounting tables, the experience correlation data, the skill correlation data, and the professional matching data, perform matching degree calculations to obtain the task matching degree, including:

[0020] Based on the analytic hierarchy process, according to the complete set of accounting tables, perform task importance determination operations to obtain the first characteristic weight, the second characteristic weight, and the third characteristic weight;

[0021] According to the first characteristic weight, the second characteristic weight, the third characteristic weight, the experience correlation data, the skill correlation data, and the professional matching data, perform matching degree calculations to obtain the task matching degree;

[0022] Among them, the task matching degree is calculated according to the following formula:

[0023] ;

[0024] In the formula, is the task matching degree; is the first feature weight; is the second feature weight; is the third feature weight; is the experience relevance data; is the skill relevance data; is the professional matching degree data.

[0025] Preferably, based on the analytic hierarchy process, according to the complete accounting table set, perform a task importance determination operation to obtain the first feature weight, the second feature weight, and the third feature weight, including:

[0026] Based on the analytic hierarchy process, extract the task feature data in the complete accounting table set to obtain the urgency, complexity, and impact scope;

[0027] According to the urgency, the complexity, and the impact scope, perform a task importance analysis and feature weight calculation operation to obtain the first feature weight, the second feature weight, and the third feature weight.

[0028] Preferably, according to the task matching degree, analyze the task capabilities of financial personnel and perform an intelligent allocation operation to obtain a task allocation table, including:

[0029] Traverse and analyze the list of financial personnel, perform a descending order ranking according to the task matching degree to obtain a matching degree ranking table;

[0030] Sort the tasks according to the difficulty level and allocate them to the corresponding financial personnel in the matching degree ranking table to obtain a task allocation result;

[0031] Record and file the task allocation result to obtain a task allocation table.

[0032] Preferably, according to the real-time task table, perform a progress analysis and task reallocation operation to obtain a final task schedule, including:

[0033] Extract the current progress and expected progress of each task in the real-time task table, and perform a deviation calculation according to the current progress and the expected progress to obtain a task determination value;

[0034] When the task determination value is greater than a preset task determination threshold, perform a matching degree adjustment calculation according to the task matching degree to obtain an adjusted matching degree;

[0035] According to the adjusted matching degree, perform a task reallocation operation to obtain a final task table.

[0036] In a second aspect, the present invention provides an AI financial task allocation system based on an accounting knowledge graph, including:

[0037] A data acquisition module, configured to acquire experience relevance data, skill relevance data, professional matching data of financial personnel, and enterprise financial business requirements;

[0038] An accounting statement acquisition module, configured to generate an accounting statement graph according to the enterprise financial business requirements, and perform form integration and summarization operations according to the accounting statement graph to obtain a complete set of accounting statements;

[0039] A distribution degree calculation module, configured to perform matching degree calculation based on the analytic hierarchy process, according to the complete set of accounting statements, the experience relevance data, the skill relevance data, and the professional matching data, to obtain a task matching degree;

[0040] A distribution table generation module, configured to analyze the task capabilities of financial personnel according to the task matching degree, and perform intelligent distribution operations to obtain a task distribution table;

[0041] A task table generation module, configured to perform data entry and node feedback operations according to the complete set of accounting statements and the task distribution table to obtain a real-time task table;

[0042] A schedule generation module, configured to perform progress analysis and task re-distribution operations according to the real-time task table to obtain a final task schedule.

[0043] In a third aspect, the present invention further provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the AI financial task allocation method based on an accounting knowledge graph described in any one of the above is implemented.

[0044] In a fourth aspect, the present invention further provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute the AI financial task allocation method based on an accounting knowledge graph described in any one of the above.

[0045] Compared with the prior art, the present invention combines an accounting knowledge graph and an artificial intelligence algorithm, generates an accounting statement graph based on enterprise financial business requirements, forms a complete set of accounting statements through form integration and summarization operations, uses the analytic hierarchy process to determine task characteristics, generates a task matching degree, and finally realizes intelligent task allocation. By constructing an underlying logic network of financial data through a knowledge graph, associating enterprise business processes and accounting nodes, and combining a task matching degree model, tasks are matched with data such as the skills and experience of financial personnel, realizing task extraction to intelligent allocation.

[0046] The present invention applies the analytic hierarchy process to the calculation of task matching degree. By extracting task features (such as urgency, complexity, and scope of influence), calculating the corresponding feature weights, and combining the experience, skills, and professional matching degree of financial personnel, a task matching degree ranking table is generated. Through the feature weight calculation formula, the priority of task allocation is dynamically adjusted to match the most suitable financial personnel to execute complex or urgent tasks, thus solving the problem in the prior art that the task priority and allocation strategy cannot be dynamically adjusted, ensuring the fairness and scientific nature of task allocation, and optimizing the human resource allocation at the same time.

[0047] The present invention also records the accounting node feedback and task progress data through a real-time task table. Combining task deviation analysis and matching degree adjustment, a task reallocation operation is carried out, and finally an optimized task schedule is generated. Starting from the real-time node status and task progress, through deviation calculation (such as the difference between the current progress and the expected progress), the matching degree adjustment and task reallocation are triggered to ensure that task delays are resolved in a timely manner, thus dynamically responding to abnormal situations in task execution, avoiding the impacts caused by delays or uneven resource allocation, and realizing the dynamic optimization of tasks and the optimization of resource utilization.

[0048] Through the construction of the accounting table atlas and the summary of the process accounting table set, the present invention realizes the structured management of task data and forms a closed-loop management system by using the data entry of financial personnel and node feedback. Before task allocation, the complete accounting table set realizes the systematic sorting of tasks; during task execution, the task progress and status are monitored through the real-time task table; after task completion, the subsequent task allocation is optimized by using the final task schedule.

[0049] In summary, the present invention can realize the intelligent allocation of tasks, improve the accounting efficiency, and reduce the error rate. Brief Description of the Drawings

[0050] Figure 1 is a schematic flowchart of an AI financial task allocation method based on an accounting knowledge graph provided by the first embodiment of the present invention;

[0051] Figure 2 is a schematic structural diagram of an AI financial task allocation system based on an accounting knowledge graph provided by the second embodiment of the present invention. Detailed Embodiments

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0053] Refer toFigure 1 , the first embodiment of the present invention provides an AI financial task allocation method based on an accounting knowledge graph, including the following steps:

[0054] S11, obtain the experience relevance data, skill relevance data, professional matching data of financial personnel, and the enterprise's financial business requirements;

[0055] S12, generate an accounting form graph according to the enterprise's financial business requirements, and perform form integration and summary operations according to the accounting form graph to obtain a complete set of accounting forms;

[0056] S13, based on the analytic hierarchy process, calculate the matching degree according to the complete set of accounting forms, the experience relevance data, the skill relevance data, and the professional matching data to obtain the task matching degree;

[0057] S14, analyze the task capabilities of financial personnel according to the task matching degree, and perform intelligent allocation operations to obtain a task allocation form;

[0058] S15, perform data entry and node feedback operations according to the complete set of accounting forms and the task allocation form to obtain a real-time task form;

[0059] S16, perform progress analysis and task reallocation operations according to the real-time task form to obtain a final task plan form.

[0060] In step S11, it is necessary to obtain the experience relevance data, skill relevance data, professional matching data of financial personnel, and the enterprise's financial business requirements, including:

[0061] In a specific embodiment, the experience relevance data of financial personnel reflects their practical experience in similar tasks and is an important basis for evaluating proficiency in task allocation. The experience relevance data includes the number of task completions, task types, completion quality scores, and completion time lengths of financial personnel in different task scenarios. These data can be obtained by querying the enterprise's internal task management system or ERP system.

[0062] Specifically, the "Historical Task Record Table" of financial personnel can be retrieved to screen task data in relevant scenarios, such as procurement accounting tasks and expense reimbursement accounting tasks participated in over the past year, and the completion quantity and average score of each task are respectively counted. The score data is sourced from the supervisor's evaluation after task completion or the performance evaluation scores automatically generated by the system. At the same time, work efficiency is calculated in combination with the task completion time. The experience relevance data is used as the experience factor in subsequent matching degree calculations to evaluate the familiarity of financial personnel with specific tasks. For example, if a financial personnel has participated in 50 procurement accounting tasks in the past 12 months with an average score of 92 points, it indicates a high experience relevance and is suitable for priority allocation of such tasks.

[0063] In a specific embodiment, the skill relevance data of financial personnel measures whether their professional skills and technical capabilities meet the task requirements. The skill data includes professional qualification certification records (such as Certified Public Accountant CPA, junior or intermediate accounting titles), system operation proficiency (such as the usage of ERP systems and financial software), and participation in training courses. These data can be extracted through the enterprise's training management system or professional qualification certification system. For example, the "Training Record Table" can be retrieved to confirm whether financial personnel have completed relevant training courses, such as tax policy training and budget analysis method training, and at the same time, the "System Operation Log" is combined to analyze their proficiency in ERP systems or financial software. In addition, financial personnel with high-level qualifications can be screened from the certification records. For example, CPA holders are more suitable for handling complex tax accounting or financial analysis tasks. These skill data directly affect the skill factor in the matching degree calculation to ensure that tasks are assigned to personnel with corresponding capabilities. For example, a financial personnel who has completed the system training of advanced financial management software and holds an intermediate accounting title indicates that they are capable of handling multi-node and complex accounting tasks.

[0064] In a specific embodiment, the professional matching degree data of financial personnel is a key indicator for task allocation and is used to measure the suitability of the professional direction of financial personnel for task requirements. The professional matching degree data includes the professional specialties of financial personnel (such as being good at expense management, budget analysis, asset accounting, etc.), field task completion records, and relevant performance evaluations. These data can be obtained through the enterprise's personnel division records, project feedback reports, and performance appraisal forms. Specifically, querying the "Personnel Professional Classification Table" can confirm the main professional direction of each financial personnel, and the "Performance Appraisal Form" is combined to evaluate their performance in specific field tasks. For example, a person who has completed 20 projects in the field task of expense management and has an average performance score of 95 points is given priority for expense-related task allocation. The professional matching degree data is directly used as the professional factor in the matching degree calculation to improve the accuracy of task allocation.

[0065] In a specific embodiment, the enterprise's financial business requirements are the basic data source for all task allocations, defining specific financial scenarios, task nodes, and execution goals. The enterprise's financial business requirements can be obtained from the annual budget report, monthly task plan, or project implementation plan of the finance department. The specific content includes: clarifying the financial scenarios (such as purchasing goods, selling goods, expense reimbursement, etc.), dividing the task nodes within the scenarios (such as generating purchase orders, recording inventory acceptance, payment accounting, etc.), and setting the completion criteria for each task (such as an accuracy rate of not less than 95% and a completion time limit of 2 days). These requirements are obtained by querying the "Annual Task Schedule" or "Enterprise Project Requirements Specification" and are associated with specific task nodes, providing data support for the subsequent generation of the accounting form atlas. For example, the enterprise's monthly financial plan requires the completion of 50 accounting tasks for purchasing goods, including generating purchase orders, updating inventory records, and verifying payment vouchers, which provide a clear scope of operation for task allocation.

[0066] In step S12, it is necessary to generate an accounting form atlas according to the enterprise's financial business requirements, and perform form integration and summary operations according to the accounting form atlas to obtain a complete set of accounting forms, including:

[0067] First, determine the financial scenario according to the enterprise's financial business requirements. The enterprise's financial business requirements are derived from the annual budget plan, monthly task objectives, or special financial project plans. When determining the financial scenario, the business scope and task objectives should be clarified. For example, the scenario of purchasing goods includes tasks such as purchase order accounting, receiving goods into inventory, and payment verification; the scenario of expense reimbursement involves tasks such as travel expense reimbursement, direct payment, and salary and welfare accounting.

[0068] Specifically, the demand data can be extracted by analyzing the task planning module in the enterprise's financial system, and the scenario scope can be further refined in combination with the business requirements specification. Clarifying the scenario is the basis for generating the accounting form atlas, ensuring that the task division is targeted.

[0069] Next, perform the operation of defining accounting nodes according to the financial scenario to obtain accounting nodes. Accounting nodes are the execution units of specific tasks in the scenario, such as "generating purchase orders", "receiving goods into inventory", "verifying payment vouchers", etc. When defining accounting nodes, it is necessary to decompose the business processes within the scenario, extract key task points, and set inputs, outputs, and goals in combination with task requirements. For example, in the scenario of purchasing goods, the accounting nodes can be defined as the "purchase application node", the "acceptance node", and the "payment node", which are responsible for verifying purchase requirements, recording inventory acceptance information, and handling supplier payments respectively. Through the task process management module or the business scenario description form, the system can automatically extract or manually mark the accounting nodes.

[0070] Then, configure the required form templates and data formats for the accounting nodes to generate an accounting table atlas. The accounting table atlas is a structured definition of the data forms required for each accounting node.

[0071] Specifically, predefined form templates can be selected or custom templates can be created according to the node task requirements. For example, the "purchase application node" needs to configure a purchase order form template, including fields such as supplier information, commodity details, and amount; the "acceptance node" needs to configure an acceptance form template, including fields such as acceptance quantity and quality inspection results. The generation of form templates needs to combine the existing resources of the enterprise financial management system to ensure format standardization and data consistency. After configuration, associate the node with its corresponding form to form an accounting table atlas.

[0072] In a specific embodiment, perform an associated form integration operation on the accounting nodes defined in the accounting table atlas and the corresponding associated forms to obtain a set of process accounting tables. The associated form integration operation aims to combine multiple node forms in the same process into a whole. For example, in the commodity purchase process, the purchase order form, the acceptance record form, and the payment voucher form can be integrated into a "commodity purchase process accounting table set".

[0073] Specifically, the form can be associated one by one according to the node execution order through the process management module, and a process diagram can be generated to ensure that the integration operation meets the logical requirements. The integrated set of process accounting tables is stored in the form of a folder for subsequent calling and modification.

[0074] Finally, perform a summary integration operation on the process accounting tables according to the financial scenarios and accounting nodes to obtain a complete set of accounting tables. The complete set of accounting tables is a further summary of the set of process accounting tables for all financial scenarios.

[0075] Specifically, it is necessary to store the process accounting tables by scenario classification, and perform field matching and data mapping on the cross-tasks between different scenarios (such as procurement and asset accounting) to ensure the comprehensiveness and consistency of the complete set of accounting tables. For example, the "inventory acceptance" form is shared by the commodity purchase and asset management scenarios, and data merging needs to be performed during integration. The generation of the complete set of accounting tables marks the completion of the structured management of task data and provides basic support for subsequent task allocation and accounting operations.

[0076] In step S13, it is necessary to perform a matching degree calculation based on the analytic hierarchy process according to the complete set of accounting tables, the experience relevance data, the skill relevance data, and the professional matching degree data to obtain a task matching degree, including:

[0077] First, extract task feature data based on the Analytic Hierarchy Process. The task feature data includes urgency, complexity, and scope of influence, which directly determine the importance of the task. When extracting the urgency, data is obtained by analyzing the completion deadline field of tasks in the complete accounting form set. For example, "The procurement order accounting task needs to be completed within 2 days." Tasks are classified into high urgency (need to be completed within 1 day), medium urgency (need to be completed within 3 days), and low urgency (completed within 5 days or more) according to time requirements. When extracting the complexity, count the number of accounting nodes involved in the task and the number of dependencies between nodes , and calculate the complexity using the following formula:

[0078] ;

[0079] It should be noted that represents the influence weight of the number of accounting nodes on the task complexity, which is used to evaluate the specific number of steps that need to be processed for the task; represents the influence weight of node dependencies on the task complexity, which measures the logical coupling degree between task nodes. The weights and are set according to the opinions of enterprise experts or historical task data. The influence weight of the number of nodes on the complexity is greater than that of dependencies. and can be set. If there are many dependencies and it is easy to make mistakes, then the value of can be appropriately increased. The number of nodes is the number of accounting nodes of the task extracted from the complete accounting form set. For example, by analyzing the independent accounting nodes defined in each task process, the number of nodes can be extracted through the "Node Statistics Table" in the process management system. For example, in the commodity purchase task, there are three nodes: "Purchase order generation", "Receipt and warehousing", and "Payment accounting", then ; The number of node dependencies is obtained by analyzing the task flow chart or dependency graph to obtain the dependencies of the nodes. For example, by extracting the input and output fields of each node through the task logic definition table, the number of direct dependencies is counted. The "Purchase order generation" node has a sequential dependency with the "Receipt and warehousing" node, and the "Receipt and warehousing" node also has a dependency with the "Payment accounting" node, then .

[0080] In a specific embodiment, the higher the complexity, the greater the difficulty of task execution. The extraction of the scope of influence is based on the impact of the task on other processes and departments, and the relevance and coverage of the fields in the accounting form are counted. For example, the "Commodity purchase task" affects procurement, inventory, and financial statements at the same time, with a large scope of influence and a high weight.

[0081] Next, perform the task importance determination operation to calculate the weights. The task importance determination requires constructing a judgment matrix of the analytic hierarchy process to quantify the relative importance among the urgency, complexity, and impact scope. Through the historical data analysis method, a feature comparison matrix is formed. For example, if the urgency is more important for task completion and its relative weight is higher than that of the complexity and impact scope, the following matrix can be constructed:

[0082] ;

[0083] In this matrix, the first row and the first column correspond to the urgency, indicating that the weight of the urgency relative to the complexity (weight is 2) and the impact scope (weight is 4); the second row and the second column correspond to the complexity, indicating that the weight of the complexity relative to the impact scope is 3; the third row and the third column correspond to the impact scope.

[0084] Specifically, perform the normalization process on the matrix, that is, sum each column of the matrix and divide each element by the sum of the column where it is located to make the sum of each column element equal to 1, and obtain the normalized matrix:

[0085] ;

[0086] Specifically, calculate the average value of each row of the normalized matrix to obtain the weight vector:

[0087] ;

[0088] The weight vector indicates that the weight of the urgency is 61.3%, the complexity is 34%, and the impact scope is 4.7%.

[0089] Subsequently, perform the matching degree calculation by combining the relevant data of the financial personnel. Calculate the task matching degree through the following formula:

[0090] ;

[0091] In the formula, is the task matching degree; is the first feature weight; is the second feature weight; is the third feature weight; is the experience correlation data; is the skill correlation data; is the professional matching degree data.

[0092] It should be noted that , and are the feature weights corresponding to the urgency, complexity, and impact scope respectively. The higher the calculation result, the more suitable the financial personnel are for the task.

[0093] Finally, a task matching ranking table is generated and output. All finance personnel are sorted by their matching scores in descending order to generate a matching ranking table, prioritizing tasks for those with the highest matching scores. The matching results are then linked to the task list to establish a task assignment priority, providing a direct basis for subsequent task assignments. Furthermore, the task matching results are stored in a database for quick access during dynamic task adjustments.

[0094] In step S14, it is necessary to analyze the task capabilities of the financial personnel based on the task matching degree and perform intelligent allocation operations to obtain a task allocation table, including:

[0095] First, the list of financial personnel is analyzed to generate a matching ranking table. The system retrieves the basic information and historical task records of the financial personnel, then iterates through the list of financial personnel one by one. Combined with the task matching results, the list is sorted in descending order from high to low, generating a matching ranking table. For example, if Financial Personnel A's matching score is 85, Financial Personnel B's is 78, and Financial Personnel C's is 72, the system will generate a matching ranking table: ranking table = {A, B, C}. This ranking process is based on the task matching scores calculated in step S13. The system prioritizes the financial personnel with the highest matching scores to undertake tasks, ensuring a good match between tasks and capabilities.

[0096] Next, tasks are sorted and assigned according to their difficulty. The difficulty of a task is determined through a comprehensive assessment of task characteristics (such as urgency, complexity, and scope of impact). The system will classify tasks based on a complete set of accounting tables. For example, tasks with high urgency and high complexity will be marked as "priority tasks," and tasks with low urgency and low complexity will be marked as "ordinary tasks." The system will prioritize assigning "priority tasks" to financial personnel who rank higher in the matching ranking table. For example, for priority task T1, if A has the highest matching degree in the ranking table, the system will assign task T1 to A, and then assign ordinary task T2 to B who ranks second. Each time a task is assigned, the system will update the load of the financial personnel in real time to ensure balanced task distribution. For example, if the load of financial personnel A has reached the preset threshold, A will be skipped and assigned to the next qualified person.

[0097] In a specific embodiment, the allocation results are recorded and archived to generate a task allocation table. After the task allocation is completed, the system records the allocation situation of each task, including information such as task number, task type, allocated personnel, and estimated completion time. For example, task T1 is allocated to financial staff A, and task T2 is allocated to financial staff B. These allocation records will be uniformly archived in the task allocation table. The format of the task allocation table includes the following fields: task number, task description, allocated personnel, matching score, allocation time, and status flag (not started, in progress, completed). These records not only provide a basis for subsequent task progress tracking but also can be used as input for the performance evaluation data of financial staff to help optimize future task allocation strategies.

[0098] In step S15, data entry and node feedback operations need to be performed based on the complete accounting table set and the task allocation table to obtain a real-time task table, including:

[0099] First, based on the complete accounting table set, accounting node task extraction and form template initialization operations are performed. The system extracts the task information of each accounting node from the complete accounting table set, including data such as task number, task content, and expected output. Combining the attributes of the accounting nodes (such as task type, node role), a standardized form template is automatically loaded for each node. The initialization of the form template includes field definition (such as "amount", "date") and data format setting (such as the amount field requires a numeric type, and the date field is in the YYYY-MM-DD format). For example, for the "payment accounting" node in the procurement process, the system will generate a form template containing fields such as "supplier name", "payment amount", and "payment date". After completing this operation, the system will generate a dedicated task form for each accounting node as the basis for subsequent entry and feedback.

[0100] In a specific embodiment, according to the task allocation table, a data entry operation for financial staff is performed on the task form. The task allocation table specifies the financial staff corresponding to each task, so the system will directly distribute the generated task form to the accounts of the designated financial staff. The financial staff enters the actual financial data in the form according to the task requirements. For example, the form for the "goods receipt and warehousing" node requires filling in fields such as the actual arrival quantity and receipt time. The system will real-time verify the data format and integrity entered by the financial staff to ensure the accuracy of the data. For example, if the "payment amount" field is empty or a non-numeric value is entered, the system will prompt an error and prevent submission. After completing the data entry, the form will be automatically uploaded to the system and archived as a completed accounting form.

[0101] In a specific embodiment, based on the filled accounting forms, perform accounting node feedback operations to generate real-time node statuses. The system will automatically analyze the form completion status of each accounting node and update the node status according to data integrity and node processing results. For example, the status of the "Receipt and Warehousing" node changes from "Incomplete" to "Completed". At the same time, if an abnormal node status occurs (such as partial data missing or logical errors), the system will mark it as the "Pending Processing" status and send feedback information to relevant financial personnel for correction.

[0102] In a specific embodiment, calculate the task progress according to the process logic based on the accounting forms. The system calculates the progress of each task in sequence from the upstream node to the downstream node according to the task logic defined in the flowchart. For example, for the merchandise purchase process, if both the "Generate Purchase Order" and "Receipt and Warehousing" nodes are completed, but the "Payment Accounting" node is not completed, then the system will calculate the total process progress as: Progress ratio = (Number of completed nodes / Total number of nodes) × 100%. During this process, the system will also combine the expected completion time and actual completion time of the time nodes to mark the task nodes with delayed progress.

[0103] In a specific embodiment, based on the real-time node status and task progress, analyze the task completion status of each node and the overall task progress ratio to generate analysis results. The system will integrate the statuses (such as "Completed", "Incomplete", "Delayed") of each accounting node and task progress data to generate an analysis report on the task completion situation. For example, for a process with 5 accounting nodes, the system shows that 3 nodes are completed, 1 node is delayed, and 1 node is incomplete, and at the same time provides the overall progress ratio (such as 80%). These analysis results are automatically integrated by the system into a real-time task table, including task status, progress, delay information, etc., as a real-time monitoring tool for enterprise financial management.

[0104] In step S16, it is necessary to perform progress analysis and task reassignment operations based on the real-time task table to obtain a final task schedule, including:

[0105] First, extract the current progress and expected progress of each task in the real-time task table, perform deviation calculation, and obtain the task determination value. The system reads the relevant progress data of each task from the real-time task table. The current progress is calculated as the ratio of the number of completed task nodes to the total number of nodes. For example: current progress = (number of completed nodes / total number of nodes) × 100%. The expected progress is calculated based on the planned completion schedule of the task and the current time. For each task, the system calculates the deviation as the current progress minus the expected progress. For example, if the current progress of a task is 60% and the expected progress is 80%, the deviation value is -20%. The system compares the deviation value with a preset task determination threshold. If the deviation value is greater than the set threshold (such as -10%), it is marked as "needs to be reallocated".

[0106] In a specific embodiment, for tasks with deviation values exceeding the threshold, perform matching degree adjustment calculation based on task matching degree to obtain the adjusted matching degree. The system re-evaluates the matching degree of the affected tasks with financial personnel, considering the current priority of the task and the remaining workload of the financial personnel, and adjusts the matching degree calculation formula:

[0107] ;

[0108] In the formula, is the adjusted matching degree; is the task matching degree; is the adjustment coefficient; is the load factor, reflecting the current task load ratio of the financial personnel;

[0109] It should be noted that the load factor is the ratio of the current number of tasks to the maximum task capacity; the adjustment coefficient is based on statistical analysis of historical data and the actual situation of the enterprise. For example, statistically analyze the task completion efficiency of enterprise financial personnel at different load factor levels (such as low, medium, and high loads), calculate the corresponding change ratio of the matching degree, and obtain the influence degree of the load factor on the matching degree. Subsequently, use regression analysis to quantify this influence degree as the adjustment coefficient. For example, if the matching degree drops by an average of 30% in the high-load state, then λ = 0.3 can be set. The value of the adjustment coefficient is between 0 and 1, and the higher the value, the greater the influence of the load factor on the matching degree.

[0110] Specifically, after the matching degree adjustment is completed, the system re-ranks the financial personnel to generate a new matching degree ranking table. For example, if the original matching degree of financial personnel A is 85, but due to high task load, the adjusted matching degree drops to 72, and the priority will be lower than that of financial personnel B with an adjusted matching degree of 78.

[0111] In a specific embodiment, according to the adjusted matching degree, a reallocation operation is performed on the tasks to generate a final task schedule. The system, based on the new matching degree sorting table, preferentially assigns tasks to financial personnel with lower workloads and higher matching degrees, while taking the priority of the tasks as a key parameter. For example, high-priority tasks will be directly assigned to the personnel with the highest adjusted matching degree, and low-priority tasks can be assigned or rescheduled later. After the task reallocation is completed, the system generates a final task schedule, recording the new assignment of each task, including fields such as task number, reallocated financial personnel, adjusted matching degree, planned completion time, etc. The system will also compare the task reallocation record with the original plan and mark the reasons for reallocation (such as schedule delay, overload).

[0112] In summary, the present invention combines an accounting knowledge graph and artificial intelligence algorithms, generates an accounting form graph based on the financial business requirements of an enterprise, forms a complete set of accounting forms through form integration and summary operations, determines task characteristics using the analytic hierarchy process, generates a task matching degree, and finally realizes the intelligent allocation of tasks. By constructing the underlying logic network of financial data through the knowledge graph, associating the enterprise business process and accounting nodes, and combining with the task matching degree model, the tasks are matched with data such as the skills and experience of financial personnel, realizing the process from task extraction to intelligent allocation.

[0113] The present invention uses the analytic hierarchy process for calculating the task matching degree. By extracting task characteristics (such as urgency, complexity, and scope of influence), calculating the corresponding characteristic weights, and combining with the experience, skills, and professional matching degree of financial personnel, a task matching degree sorting table is generated. Through the characteristic weight calculation formula, the priority of task allocation is dynamically adjusted to match the most suitable financial personnel to execute complex or urgent tasks, thus solving the problem in the prior art that the task priority and allocation strategy cannot be dynamically adjusted, ensuring the fairness and scientific nature of task allocation, and optimizing the human resource allocation at the same time.

[0114] The present invention also records the accounting node feedback and task progress data through a real-time task table, combines task deviation analysis and matching degree adjustment, performs a task reallocation operation, and finally generates an optimized task schedule. Starting from the real-time node status and task progress, through deviation calculation (such as the difference between the current progress and the expected progress), the matching degree adjustment and task reallocation are triggered to ensure that task delays are resolved in a timely manner, thus dynamically responding to abnormal situations during task execution, avoiding the impacts caused by delays or uneven resource allocation, and realizing the dynamic optimization of tasks and the optimization of resource utilization.

[0115] Through the construction of the accounting form graph and the summary of the process accounting form set, the present invention realizes the structured management of task data, and forms a closed-loop management system by using the data entry and node feedback of financial personnel. Before task allocation, the complete accounting form set realizes the systematic sorting of tasks; during task execution, the task progress and status are monitored through the real-time task form; after task completion, the subsequent task allocation is optimized by using the final task schedule. The present invention can realize the intelligent allocation of tasks, improve the accounting efficiency and reduce the error rate.

[0116] Referring to Figure 2 , the second embodiment of the present invention provides an AI financial task allocation system based on an accounting knowledge graph, including:

[0117] A data acquisition module, configured to acquire the experience relevance data, skill relevance data, professional matching data of financial personnel, and the enterprise financial business requirements;

[0118] An accounting form acquisition module, configured to generate an accounting form graph according to the enterprise financial business requirements, and perform form integration and summary operations according to the accounting form graph to obtain a complete accounting form set;

[0119] An allocation degree calculation module, configured to perform matching degree calculation based on the analytic hierarchy process according to the complete accounting form set, the experience relevance data, the skill relevance data, and the professional matching data to obtain a task matching degree;

[0120] An allocation form generation module, configured to analyze the task capabilities of financial personnel according to the task matching degree, and perform intelligent allocation operations to obtain a task allocation form;

[0121] A task form generation module, configured to perform data entry and node feedback operations according to the complete accounting form set and the task allocation form to obtain a real-time task form;

[0122] A schedule generation module, configured to perform progress analysis and task reallocation operations according to the real-time task form to obtain a final task schedule.

[0123] Preferably, the data acquisition module is configured to acquire the experience relevance data, skill relevance data, professional matching data of financial personnel, and the enterprise financial business requirements;

[0124] Preferably, the accounting form acquisition module is configured to generate an accounting form graph according to the enterprise financial business requirements, and perform form integration and summary operations according to the accounting form graph to obtain a complete accounting form set, including:

[0125] The generating an accounting form graph according to the enterprise financial business requirements, and performing form integration and summary operations according to the accounting form graph to obtain a complete accounting form set, including:

[0126] Determine the financial scenario according to the enterprise's financial business requirements;

[0127] Perform an accounting node definition operation according to the financial scenario to obtain accounting nodes;

[0128] Configure the required form templates and data formats for the accounting nodes to obtain an accounting form atlas;

[0129] Perform an associated form integration operation on the accounting nodes and the corresponding associated forms defined in the accounting form atlas, and combine all the node forms in the same process into a process accounting form set to obtain a process accounting form set;

[0130] Perform a summary integration operation on the process accounting form according to the financial scenario and the accounting nodes to obtain a complete accounting form set.

[0131] Preferably, the allocation degree calculation module is used to perform a matching degree calculation based on the analytic hierarchy process according to the complete accounting form set, the experience correlation data, the skill correlation data, and the professional matching data to obtain a task matching degree, including:

[0132] The performing a matching degree calculation based on the analytic hierarchy process according to the complete accounting form set, the experience correlation data, the skill correlation data, and the professional matching data to obtain a task matching degree includes:

[0133] Based on the analytic hierarchy process, perform a task importance determination operation according to the complete accounting form set to obtain a first characteristic weight, a second characteristic weight, and a third characteristic weight;

[0134] Perform a matching degree calculation according to the first characteristic weight, the second characteristic weight, the third characteristic weight, the experience correlation data, the skill correlation data, and the professional matching data to obtain a task matching degree;

[0135] Among them, the task matching degree is calculated according to the following formula:

[0136] ;

[0137] In the formula, is the task matching degree; is the first characteristic weight; is the second characteristic weight; is the third characteristic weight; is the experience correlation data; is the skill correlation data; is the professional matching data.

[0138] Based on the analytic hierarchy process, perform an operation for determining the importance of tasks according to the complete accounting form set, and obtain a first feature weight, a second feature weight, and a third feature weight, including:

[0139] Based on the analytic hierarchy process, extract task feature data from the complete accounting form set to obtain the urgency level, complexity level, and influence scope;

[0140] According to the urgency level, the complexity level, and the influence scope, perform an operation for task importance analysis and feature weight calculation to obtain a first feature weight, a second feature weight, and a third feature weight.

[0141] Preferably, the allocation form generation module is used to analyze the task capabilities of financial personnel according to the task matching degree and perform an intelligent allocation operation to obtain a task allocation form, including:

[0142] The operation of analyzing the task capabilities of financial personnel according to the task matching degree and performing an intelligent allocation operation to obtain a task allocation form includes:

[0143] Traverse and analyze the list of financial personnel, perform a descending order ranking according to the task matching degree to obtain a matching degree ranking form;

[0144] Sort the tasks according to the difficulty level and allocate them to the corresponding financial personnel in the matching degree ranking form to obtain a task allocation result;

[0145] Record and file the task allocation result to obtain a task allocation form.

[0146] Preferably, the task form generation module is used to perform data entry and node feedback operations according to the complete accounting form set and the task allocation form to obtain a real-time task form, including:

[0147] The operation of performing data entry and node feedback operations according to the complete accounting form set and the task allocation form to obtain a real-time task form;

[0148] According to the complete accounting form set, perform an operation for extracting accounting node tasks and initializing form templates to obtain a task form corresponding to each accounting node;

[0149] According to the task allocation form, perform a data entry operation of financial personnel on the task form to obtain a filled accounting form;

[0150] According to the accounting form, perform an accounting node feedback operation to obtain a real-time node status;

[0151] According to the accounting form, calculate the task progress according to the process logic to obtain the task progress;

[0152] Analyze the task completion status of each node and the overall task progress ratio according to the real-time node status and the task progress, and obtain an analysis result;

[0153] Integrate the analysis result into a real-time task table.

[0154] Preferably, the schedule generation module is configured to perform progress analysis and task reallocation operations according to the real-time task table to obtain a final task schedule, including:

[0155] The performing progress analysis and task reallocation operations according to the real-time task table to obtain a final task schedule, including:

[0156] Extract the current progress and the expected progress of each task in the real-time task table, and perform deviation calculation according to the current progress and the expected progress to obtain a task determination value;

[0157] When the task determination value is greater than a preset task determination threshold, perform matching degree adjustment calculation according to the task matching degree to obtain an adjusted matching degree;

[0158] Perform a task reallocation operation according to the adjusted matching degree to obtain a final task table.

[0159] It should be noted that the AI financial task allocation method system based on an accounting knowledge graph provided in an embodiment of the present invention is used to execute all the process steps of the AI financial task allocation method based on an accounting knowledge graph in the above embodiment. The working principles and beneficial effects of the two correspond one by one, and thus will not be elaborated herein.

[0160] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as an allocation degree calculation program. When the processor executes the computer program, the steps in the above embodiments of the AI financial task allocation method based on an accounting knowledge graph are implemented, such as Figure 1 The step S11 shown. Alternatively, when the processor executes the computer program, the functions of each module / unit in the above device embodiments are implemented, such as the allocation degree calculation module.

[0161] Exemplarily, the computer program may be divided into one or more modules / units. The one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0162] The electronic device can be a computing device such as a desktop computer, notebook, palm computer, and smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above components are only examples of the electronic device and do not constitute a limitation on the electronic device. It may include more or fewer components than the above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, a bus, etc.

[0163] The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the electronic device and connects various parts of the entire electronic device through various interfaces and circuits.

[0164] The memory can be used to store the computer program and / or module. The processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory may include high-speed random access memory and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one magnetic disk storage device, flash device, or other volatile solid-state storage devices.

[0165] Among them, if the modules / units integrated in the electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above-described embodiment methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the content included in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium does not include electrical carrier signals and telecommunication signals.

[0166] It should be noted that the device embodiments described above are only illustrative. The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that they have a communication connection, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0167] The specific embodiments described above have further elaborated on the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. It is particularly pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An AI financial task allocation method based on an accounting knowledge graph, characterized in that Including: Obtain the experience relevance data, skill relevance data, professional matching data of financial personnel and the enterprise's financial business requirements; Determine the financial scenario according to the enterprise's financial business requirements; Perform an accounting node definition operation according to the financial scenario to obtain accounting nodes; Configure the required form templates and data formats for the accounting nodes to obtain an accounting table atlas; Perform an associated form integration operation on the accounting nodes and corresponding associated forms defined in the accounting table atlas, and combine all node forms in the same process into a process accounting table set to obtain a process accounting table set; Perform a summary integration operation on the process accounting table according to the financial scenario and the accounting nodes to obtain a complete accounting table set; Based on the analytic hierarchy process, perform a matching degree calculation according to the complete accounting table set, the experience relevance data, the skill relevance data and the professional matching data to obtain a task matching degree, including: Based on the analytic hierarchy process, extract the task feature data in the complete accounting table set to obtain the urgency, complexity and impact scope; Among them, the extraction of the urgency includes analyzing the data obtained from the completion deadline field of the tasks in the complete accounting table set; the extraction of the complexity includes counting the number of accounting nodes n involved in the task and the number of interdependence relationships m between nodes, and using the following formula to calculate the complexity: ； Among them, represents the influence weight of the number of accounting nodes on the task complexity, and is used to evaluate the specific number of steps that the task needs to process; represents the influence weight of the node dependency relationship on the task complexity, and measures the logical coupling degree between task nodes; The extraction of the impact scope includes counting the relevance and coverage of the fields in the accounting table based on the impact of the task on other processes and departments; According to the urgency, the complexity and the impact scope, construct a judgment matrix of the analytic hierarchy process, perform normalization processing on the matrix, and then calculate the average value of each row to obtain the first eigenweight, the second eigenweight and the third eigenweight; Perform a matching degree calculation according to the first eigenweight, the second eigenweight, the third eigenweight, the experience relevance data, the skill relevance data and the professional matching data to obtain a task matching degree; Among them, the task matching degree is calculated according to the following formula: ； Wherein, is the task matching degree; is the first feature weight; is the second feature weight; is the third feature weight; is the experience correlation data; is the skill correlation data; is the professional matching degree data; Analyze the task capabilities of financial personnel according to the task matching degree and perform an intelligent allocation operation to obtain a task allocation table; Perform an accounting node task extraction and form template initialization operation according to the complete accounting table set to obtain a task form corresponding to each accounting node; Perform a financial personnel data entry operation on the task form according to the task allocation table to obtain a filled accounting form; Perform an accounting node feedback operation according to the accounting form to obtain the real-time node status; Calculate the task progress according to the process logic of the accounting form to obtain the task progress; Analyze the task completion situation of each node and the overall task progress ratio according to the real-time node status and the task progress to obtain an analysis result; Integrate the analysis result into a real-time task table; Perform a progress analysis and task reallocation operation according to the real-time task table to obtain a final task schedule, including: Extract the current progress and expected progress of each task in the real-time task table, and perform a deviation calculation according to the current progress and the expected progress to obtain a task determination value; When the task determination value is greater than a preset task determination threshold, according to the task matching degree, perform matching degree adjustment calculation to obtain an adjusted matching degree; Among them, the matching degree adjustment calculation formula is as follows: ； Among them, is for adjusting the matching degree; is the task matching degree; is the adjustment coefficient; is the load factor, reflecting the current task load ratio of financial personnel; the load factor is the ratio of the current task quantity of financial personnel to their maximum task capacity, and the adjustment coefficient is a parameter determined by historical data statistics and regression analysis based on the task completion efficiency of financial personnel at different load levels; According to the adjusted matching degree, perform a task reallocation operation to obtain a final task list.

2. The AI financial task allocation method based on an accounting knowledge graph according to claim 1, wherein The method of analyzing the task capabilities of financial personnel according to the task matching degree and performing an intelligent allocation operation to obtain a task allocation list includes: Traverse and analyze the list of financial personnel, and perform a descending order ranking according to the task matching degree to obtain a matching degree ranking list; Sort the tasks according to the difficulty level and assign them to the corresponding financial personnel in the matching degree ranking list to obtain a task allocation result; Record and file the task allocation result to obtain a task allocation list.

3. An AI financial task allocation method system based on an accounting knowledge graph, characterized in that, Used to implement the AI financial task allocation method based on an accounting knowledge graph as described in any one of claims 1 to 2, including: A data acquisition module, used to acquire the experience relevance data, skill relevance data, professional matching degree data of financial personnel, and the enterprise's financial business requirements; An accounting statement acquisition module, used to generate an accounting statement knowledge graph according to the enterprise's financial business requirements, and perform form integration and summary operations according to the accounting statement knowledge graph to obtain a complete set of accounting statements; An allocation degree calculation module, used to perform matching degree calculation based on the analytic hierarchy process, according to the complete set of accounting statements, the experience relevance data, the skill relevance data, and the professional matching degree data, to obtain a task matching degree; An allocation list generation module, used to analyze the task capabilities of financial personnel according to the task matching degree and perform an intelligent allocation operation to obtain a task allocation list; A task list generation module, used to perform data entry and node feedback operations according to the complete set of accounting statements and the task allocation list to obtain a real-time task list; A schedule generation module, used to perform progress analysis and task reallocation operations according to the real-time task list to obtain a final task schedule.

4. An electronic device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the AI financial task allocation method based on an accounting knowledge graph as described in any one of claims 1 to 2.

5. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the AI financial task allocation method based on an accounting knowledge graph as described in any one of claims 1 to 2.

Images