Government affair intelligent management system based on knowledge graph

Abstract

The invention discloses an intelligent government affair management system based on a knowledge graph. The intelligent government affair management system is used for improving the efficiency and accuracy of intelligent government affair management. According to the government affair request, a process entity node is determined in the primary layer knowledge graph; and constructing an ant colony flow path model according to the flow entity nodes. And determining a process entity node corresponding to each piece of image information in the government affair request in the ant colony process path model. And generating the execution progress of the corresponding process entity node through the image features. And calculating the historical use priority of each process entity node. And performing priority assignment on each process entity node according to the ant colony algorithm, the execution progress and the historical use priority, and determining a first target ant colony process path. And selecting a sub-layer knowledge graph according to the first target ant colony flow path and the government affair request. And generating a new ant colony flow path model according to the sub-layer knowledge graph and the government affair request, and generating a second target ant colony flow path. And processing the government affair request according to the second target ant colony flow path.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of government affairs processing, and in particular to an intelligent government affairs management system based on a knowledge graph. Background Art

[0002] Existing government services rely on government data. This data is large in volume, multi-system, and multi-layered, which poses challenges to government management systems. Knowledge graphs can now be used to assist management, effectively integrating government data.

[0003] The knowledge graph organizes knowledge in the government affairs field in a structured manner, covering various information such as policies and regulations, business processes, and departmental responsibilities. When a user asks for advice or handles business, the system can quickly locate relevant information in the knowledge graph to achieve accurate query and rapid response. In addition, a core of the government affairs service of the knowledge graph is to select the process entity node that is most suitable for the user to form a process service path. The knowledge graph can intelligently guide users to complete the business process based on user needs and input information. By analyzing the logical relationships and rules in the knowledge graph, the system can automatically determine the user's current stage of processing, and can also accurately analyze the operations and materials required for the next step.

[0004] However, the current knowledge graph-based government management system still has some problems. Government requests typically include many types of information, most commonly text and images. Existing knowledge graphs only analyze text information through semantic analysis models to determine process entity nodes in the knowledge graph, without utilizing image information. Secondly, in existing knowledge graphs, the same process entity node can lead to different process entity nodes. In other words, the same government request may have multiple process service paths, but the materials required for each process service path are different. This makes it impossible for the knowledge graph to determine the optimal process service path for users, reducing the efficiency and accuracy of intelligent government management. Secondly, the knowledge graph in the field of government affairs is composed of numerous government documents. These government documents are generally issued by superiors, but in subordinate regions, the superior government documents are usually refined to generate subordinate government documents with more specific steps. That is, each region can refine the policies of its region based on the policies of the upper level. This makes it impossible for a superior knowledge graph to obtain the subordinate government documents of the region in a timely manner, and it is impossible to update the knowledge graph based on the subordinate government documents of the region, which in turn leads to the inability of government requests to be processed according to the process entity nodes unique to the region, increasing processing time and difficulty, and further reducing the efficiency and accuracy of intelligent government management. Summary of the Invention

[0005] This application discloses a knowledge graph-based government affairs intelligent management system for improving the efficiency and accuracy of government affairs intelligent management.

[0006] In a first aspect, embodiments of the present application provide a knowledge graph-based intelligent government affairs management system, including: The government affairs intelligent management system determines several process entity nodes in the primary knowledge graph based on the text information in the government affairs request; The government affairs intelligent management system constructs an ant colony process path model according to the process sequence of several process entity nodes. There are several process paths composed of several process entity nodes in the ant colony process path model. The government affairs intelligent management system determines the process entity node corresponding to each image information in the government affairs request in the ant colony process path model; The government affairs intelligent management system extracts image features from image information and generates the execution progress of the corresponding process entity nodes based on the image features; The government affairs intelligent management system calculates the historical usage priority of each process entity node; The government affairs intelligent management system assigns priority to each process entity node based on the ant colony algorithm, execution progress, and historical usage priority, and determines the first target ant colony process path from the ant colony process path model. The sum of the priority assignments of all process entity nodes in the first target ant colony process path is higher than that of other process paths. The government affairs intelligent management system selects a sub-layer knowledge graph based on the first target ant colony process path and government affairs request; The government affairs intelligent management system generates a new ant colony process path model based on the sub-layer knowledge graph and government affairs requests, assigns priorities to the process entity nodes in the ant colony process path model, and generates a second target ant colony process path; The government affairs intelligent management system processes government affairs requests according to the second target ant colony process path.

[0007] Optionally, the government affairs intelligent management system determines the process entity node corresponding to each image information in the ant colony process path model, including: The government affairs intelligent management system determines the attachment type label of the image information; The government affairs intelligent management system determines the material requirement type label of each process entity node based on the primary knowledge graph. Each process entity node includes at least one material requirement type label. The government affairs intelligent management system matches the attachment type label and the material requirement type label so that each image information is matched to the corresponding process entity node in the ant colony process path model.

[0008] Optionally, the government affairs intelligent management system extracts image features from the image information and generates the execution progress of the corresponding process entity node based on the image features, including: The government affairs intelligent management system determines the verification feature recognition model and reference verification data based on the primary knowledge graph and material requirement type labels; The government affairs intelligent management system uses the verification feature recognition model to analyze the image information and the image features in the reference verification data, and generates verification results for all the image information in each process entity node; The government affairs intelligent management system generates node simulation route parameters for each process entity node based on all verification results in each process entity node; The intelligent government management system adjusts the execution progress of process entity nodes according to the node simulation route parameters.

[0009] Optionally, the government affairs intelligent management system calculates the historical usage priority of each process entity node, including: The government affairs intelligent management system determines the operation history information of each process entity node from the primary knowledge graph; The government affairs intelligent management system generates utilization rate parameters for each process entity node based on the operation history information of each process entity node; The intelligent government management system adjusts the historical usage priority of each process entity node based on the usage rate parameters.

[0010] Optionally, the government affairs intelligent management system selects a sub-layer knowledge graph according to the first target ant colony process path and the government affairs request, including: The government affairs intelligent management system determines the superior reference data for constructing the primary knowledge graph based on the first target ant colony process path; The government affairs intelligent management system determines the lower-level reference data in the primary knowledge graph based on the upper-level reference data and the government affairs area information of the government affairs request; The intelligent government management system selects the sub-layer knowledge graph with the highest correlation based on the lower-level reference data.

[0011] Optionally, the government affairs intelligent management system determines several process entity nodes in the primary knowledge graph based on the text information in the government affairs request, including: The government affairs intelligent management system performs semantic analysis on the text information in the government affairs request and determines several initial entity processing nodes from the primary knowledge graph; The government affairs intelligent management system performs peer node expansion processing based on the initial entity processing node; The government affairs intelligent management system performs a process entity node expansion based on the initial entity processing node, determines the next level process entity node, and repeatedly performs process entity node expansion on the process entity node to generate a process entity node set; The government affair intelligent management system combines a plurality of initial entity processing nodes and a plurality of process entity nodes to generate a plurality of process entity nodes.

[0012] In a second aspect, embodiments of the present application provide a computer readable storage medium, the computer readable storage medium having a program stored thereon, the program, when executed on a computer, performs the method of the first aspect and any optional implementation of the first aspect.

[0013] From the above technical solutions, the embodiments of the present application have the following advantages: The government affair intelligent management system first determines a plurality of process entity nodes in the initial layer knowledge graph according to the text information in the government affair request. Next, the government affair intelligent management system constructs an ant colony process path model according to the plurality of process entity nodes, and initializes the parameters of each process entity node in the ant colony process path model according to the ant colony algorithm. The government affair intelligent management system determines the corresponding process entity node of each image information in the ant colony process path model in the government affair request. The government affair intelligent management system extracts the image features in the image information, and adjusts the execution progress of the corresponding process entity node through the image features. The government affair intelligent management system adjusts the historical use priority of each process entity node. The government affair intelligent management system determines the first target ant colony process path through the ant colony algorithm and the ant colony process path model. The government affair intelligent management system selects the sub-layer knowledge graph according to the first target ant colony process path and the government affair request. The government affair intelligent management system generates the sub-layer knowledge graph for the government affair request and processes according to the second target ant colony process path.

[0014] Through the image feature analysis method and the ant colony algorithm, the process entity nodes in the ant colony process path model are screened to determine the optimal first target ant colony process path. Then, the corresponding superior file of the first target ant colony process path is determined, and the corresponding regional subordinate file of the superior file is determined according to the government affair request, so as to find the corresponding sub-layer knowledge graph. Finally, the government affair intelligent management system generates the sub-layer knowledge graph for the government affair request and processes according to the second target ant colony process path. The image features are reasonably used, and the first target ant colony process path determined according to the initial layer knowledge graph is used to find a more suitable sub-layer knowledge graph. Finally, the second target ant colony process path corresponding to the region is determined for the government affair request through the sub-layer knowledge graph, and the efficiency and accuracy of the government affair intelligent management are improved. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0016] Figure 1 This is a schematic diagram of an embodiment of the knowledge graph-based government affairs intelligent management system of this application; Figure 2 A schematic diagram of an embodiment of a method for determining process entity nodes in the government affairs intelligent management system of this application; Figure 3 A schematic diagram of an embodiment of a method for adjusting the execution progress of the government affairs intelligent management system of this application; Figure 4 A schematic diagram of an embodiment of a method for adjusting historical usage priority in the government affairs intelligent management system of this application; Figure 5 A schematic diagram of an embodiment of a method for selecting a sub-layer knowledge graph for the government affairs intelligent management system of this application; Figure 6 A schematic diagram of another embodiment of the method for determining process entity nodes in the government affairs intelligent management system of this application. DETAILED DESCRIPTION

[0017] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0018] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0019] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0020] As used in this specification and the appended claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0021] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0022] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0023] In the existing technology, there are still some problems with the government management system based on knowledge graphs. Government requests usually include many types of information, most commonly text and images. Existing knowledge graphs only analyze text information through semantic analysis models to determine process entity nodes in the knowledge graph, and do not utilize image information. Secondly, in existing knowledge graphs, the same process entity node can lead to different process entity nodes. In other words, the same government request may have many process service paths, but the materials required for each process service path are different. This makes it impossible for the knowledge graph to determine the optimal process service path for users, reducing the efficiency and accuracy of intelligent government management. Secondly, the knowledge graph in the field of government affairs is composed of numerous government documents. These government documents are generally issued by superiors, but in subordinate regions, the superior government documents are usually refined to generate subordinate government documents with more specific steps. That is, each region can refine the policies of its region based on the policies of the upper level. This makes it impossible for a superior knowledge graph to obtain the subordinate government documents of the region in a timely manner, and it is impossible to update the knowledge graph based on the subordinate government documents of the region, which in turn leads to the inability of government requests to be processed according to the process entity nodes unique to the region, increasing processing time and difficulty, and further reducing the efficiency and accuracy of intelligent government management.

[0024] Based on this, this application discloses a government affairs intelligent management system based on knowledge graph, which is used to improve the efficiency and accuracy of government affairs intelligent management.

[0025] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0026] The method of the present application can be applied to a server, device, terminal or other device with logic processing capability, and the present application does not limit this. For the convenience of description, the following description is based on the example of the execution subject being a terminal.

[0027] See also Figure 1 This application provides an embodiment of a government affairs intelligent management system based on a knowledge graph, including: 101. The government affairs intelligent management system determines several process entity nodes in the primary knowledge graph based on the text information in the government affairs request; In this embodiment, after receiving a government affairs request, the government affairs intelligent management system first obtains semantic information through semantic analysis, and then uses the semantic information to determine several process entity nodes from the primary knowledge graph. The specific content will be described in detail later.

[0028] 102. The government affairs intelligent management system constructs an ant colony process path model according to the process sequence of a number of process entity nodes. There are a number of process paths composed of a number of process entity nodes in the ant colony process path model. In this embodiment, the government affairs intelligent management system associates several acquired process entity nodes, so that they form at least one process path. This creates an ant colony process path model, fitting each process entity node into a path. Subsequently, the parameters of each process entity node are initialized based on the ant colony algorithm and the parameters of the process entity node, ensuring that they meet the requirements of the ant colony algorithm. Specifically, the execution progress (heuristic information) and historical usage priority (initial information concentration) of each process entity node are calculated. Then, information such as the pheromone importance factor, heuristic information importance factor, ant colony size, pheromone volatility coefficient, and pheromone intensity are set as appropriate to aid the ant colony algorithm's calculations.

[0029] 103. The government affairs intelligent management system determines the process entity node corresponding to each image information in the government affairs request in the ant colony process path model; After the terminal initializes the ant colony process path model, it is necessary to determine the process entity node corresponding to each image information in the government affairs request in the ant colony process path model, and to match it according to the label corresponding to the image information, because each process entity node will require at least one type of proof material, and the proof material uploaded by the requester is often an image, and will be annotated with the corresponding label. At this time, it is necessary to use the model to determine whether the image is accurate. Before determining whether it is accurate, it is necessary to determine which nodes these images correspond to, and then use the recognition model corresponding to this node for identification. For example, a process entity node requires the requester to upload a request. After receiving the file with the note "request", this file is matched with the process entity node. The specific steps are described in the subsequent embodiments.

[0030] 104. The government affairs intelligent management system extracts image features from the image information and generates the execution progress of the corresponding process entity node based on the image features; The government affairs intelligent management system extracts and analyzes image features from image information, and generates the execution progress of process entity nodes based on the analysis results of the image features. This method can determine the completeness of the materials required for each process entity node. That is, when the requester uses a government affairs request to reach the process entity node, whether the image data in the government affairs request can quickly complete this node, allowing the government affairs request to smoothly proceed to the next process entity node. The execution progress represents the completion degree of the process entity node, thereby accurately updating the path corresponding to each node in the ant colony process path model, making the ant colony process path model more consistent with the difficulty of all process entity nodes in real processing scenarios. This parameter is also an important parameter for subsequent ant colony calculation. The specific steps are described in detail in the subsequent embodiments.

[0031] 105. The government affairs intelligent management system calculates the historical usage priority of each process entity node; In order to better make the ant colony process path model conform to the real scenario, it is necessary to calculate the historical usage priority of each process entity node based on the government processing data (historical data) in the real scenario. The government processing data here can also be obtained based on the primary knowledge graph. The historical usage priority indicates how frequently this process entity node is used by the requester. The specific processing steps will be explained in detail later.

[0032] 106. The government affairs intelligent management system assigns a priority value to each process entity node based on the ant colony algorithm, execution progress, and historical usage priority, and determines a first target ant colony process path from the ant colony process path model. The sum of the priority values ​​assigned to all process entity nodes in the first target ant colony process path is higher than that of other process paths. In this embodiment, when the ant colony process path model completes the execution progress and historical usage priority After the update, it can be iterated based on the algorithm to assign priority to each process entity node. After the assignment is completed, the path of the government process can be selected, and finally the first target ant colony process path can be obtained. There are many process entity nodes in the first target ant colony process path.

[0033] Specifically, first according to the execution progress of each node and historical usage priority Calculate the probability that the requester will choose this node:

[0034] Among them, the number of process entity nodes belonging to the same level is determined to be M, and is the execution progress and historical usage priority of the k-th process entity node, is the execution progress weight (the importance of pheromone), The weight of the historical usage priority (importance of heuristic information) is used to calculate the selection probability of the kth process entity node, combined with the execution progress history usage priority of the process entity nodes at the same level. There are many other methods for calculating the above-mentioned probability, which are not limited here.

[0035] After calculating each process entity node, the ant colony algorithm and initialized model parameters (such as the ant colony size, pheromone volatility coefficient, and pheromone intensity) are used to calculate the number of times each process entity node is selected. For example, if the number of ants is set to 50,000, starting from the initial process entity node, each ant passes through a process entity node, leaving behind a historical priority assignment. This increases the historical priority, increasing the probability P of this process entity node being selected, influencing the selection of subsequent ants. Furthermore, the pheromone volatility can be configured according to the ant colony algorithm (after each ant passes, the historical priority of all process entity nodes is uniformly reduced by a value that is less than the value the ants assign to the historical priority increase) to accelerate iteration. Once all ants have completed assigning values ​​to the process entity nodes, the current historical priority of each process entity node is determined as the priority assignment. The unique sum of priority assignments is calculated for each process path, and the process path with the largest sum is selected as the first target ant colony process path.

[0036] 107. The government affairs intelligent management system selects a sub-layer knowledge graph based on the first target ant colony process path and the government affairs request; In the prior art, when a government request is received, while the approximate major project to be addressed can be determined, the regional policy of the request cannot be determined, making it impossible to assign it to an appropriate knowledge graph. This is because the text in government requests is often highly regionalized, and the wording used in superior and subordinate government documents can differ when compared. Subordinate government documents refine the policies of superior government documents. After policy refinement, the subordinate document may include new vocabulary or convert vocabulary from the superior document into new vocabulary (different vocabulary but the same meaning), which can lead to lexical regionalization. For example, different nouns may be used in superior and subordinate government documents to represent the same meaning. Furthermore, the solutions used in the documents after the refinement differ from those in the superior document. This leads to differences in the processes of the superior and subordinate knowledge graphs. Furthermore, due to the regionalized vocabulary in the government request, the superior knowledge graph (the primary knowledge graph) may not provide the most appropriate service solution for the government request.

[0037] To address the above issues, in this embodiment, after using the primary knowledge graph in conjunction with the ant algorithm to obtain the first target ant colony process path, the first target ant colony process path may not be the most suitable for the government request, but it has a relatively high priority in the primary knowledge graph constructed by the superior policy document. In this case, it is necessary to select a sub-layer knowledge graph that corresponds to the corresponding area and can match the government document based on the combination of the first target ant colony process path and the government request. The specific selection of the sub-layer knowledge graph will be explained in the subsequent embodiments.

[0038] 108. The government affairs intelligent management system generates a new ant colony process path model based on the sub-layer knowledge graph and the government affairs request, assigns priorities to the process entity nodes in the ant colony process path model, and generates a second target ant colony process path; 109. The government affairs intelligent management system processes the government affairs request according to the second target ant colony process path.

[0039] In this embodiment, the government affairs intelligent management system generates a new ant colony process path model based on the sub-layer knowledge graph and the government affairs request. Specifically, it generates a new ant colony process path model based on the sub-layer knowledge graph that better matches the government affairs request. This step is similar to step 101 above and is not further described here. The government affairs intelligent management system then determines several process entity nodes in the primary knowledge graph based on the text information in the government affairs request. It also determines the process entity node corresponding to each image information in the government affairs request in the ant colony process path model. Next, it extracts image features from the image information and generates the execution progress of the corresponding process entity node based on the image features. It calculates the historical usage priority of each process entity node. It then assigns a priority to each process entity node based on the ant colony algorithm, the execution progress, and the historical usage priority. A second target ant colony process path is determined from the ant colony process path model. This method is similar to steps 102 to 106 above and is similar to the detailed description of these steps in subsequent embodiments. Please refer to the corresponding analysis of these steps and do not elaborate on them here. Finally, it is determined that the second target ant colony process path is a process path that better matches the government affairs request. At this point, the requester can be served according to each process node of the second target ant colony process path.

[0040] In this embodiment, the government affairs intelligent management system first identifies several process entity nodes in the primary knowledge graph for the text information in the government affairs request. Next, the government affairs intelligent management system constructs an ant colony process path model based on the several process entity nodes and initializes the parameters of each process entity node in the ant colony process path model using the ant colony algorithm. The government affairs intelligent management system determines the process entity node corresponding to each image information in the government affairs request in the ant colony process path model. The government affairs intelligent management system extracts image features from the image information and adjusts the execution progress of the corresponding process entity node based on the image features. The government affairs intelligent management system adjusts the historical usage priority of each process entity node. The government affairs intelligent management system determines a first target ant colony process path using the ant colony algorithm and the ant colony process path model. The government affairs intelligent management system selects a sub-layer knowledge graph based on the first target ant colony process path and the government affairs request. The government affairs intelligent management system generates and processes the government affairs request based on the second target ant colony process path based on the sub-layer knowledge graph.

[0041] By using image feature analysis methods and ant colony algorithms to screen the process entity nodes in the ant colony process path model, the optimal first target ant colony process path is determined. Then, based on the superior file corresponding to the first target ant colony process path, the regional subordinate file corresponding to the superior file is determined based on the government request, thereby finding the corresponding sub-layer knowledge graph. Finally, the government affairs intelligent management system generates and processes the government affairs request based on the sub-layer knowledge graph according to the second target ant colony process path. While making reasonable use of image features, a more suitable sub-layer knowledge graph is found based on the first target ant colony process path determined by the primary knowledge graph. Finally, the sub-layer knowledge graph is used to determine the second target ant colony process path that matches the corresponding region for the government affairs request, thereby improving the efficiency and accuracy of government affairs intelligent management.

[0042] See also Figure 2 The present application provides an embodiment of a method for determining a process entity node, including: 201. The intelligent government management system determines the attachment type label of the image information; This example utilizes the tag information of the image uploaded in the government affairs request. For example, if image A is uploaded, the user tags the attachment type as "government affairs contract." Another example is if image B is uploaded, the tag is "environmental pollution evidence." Tags help bind images to corresponding process entity nodes.

[0043] 202. The government affairs intelligent management system determines the material requirement type label of each process entity node based on the primary knowledge graph. Each process entity node includes at least one material requirement type label. In this embodiment, after the government affairs intelligent management system determines the attachment type tag of the image information, it also needs to determine the corresponding files to be received at each process entity node. The government affairs intelligent management system can use the primary knowledge graph to create a node file list. For example, the knowledge graph can determine that a process entity node requires the upload of image information such as government contracts and delivery receipts.

[0044] The government affairs intelligent management system determines the material requirement type label of each process entity node based on the primary knowledge graph. Each process entity node includes at least one material requirement type label.

[0045] 203. The government affairs intelligent management system matches the attachment type label and the material requirement type label so that each image information is matched to the corresponding process entity node in the ant colony process path model.

[0046] After the government affairs intelligent management system determines the attachment type label and the material requirement type label of each process entity node, the government affairs intelligent management system matches the attachment type label and the material requirement type label so that each image information is matched to the corresponding process entity node in the ant colony process path model.

[0047] See also Figure 3 , the present application provides an embodiment of a method for adjusting execution progress, comprising: 301. The government affairs intelligent management system determines the verification feature recognition model and reference verification data based on the primary knowledge graph and material requirement type labels; 302. The government affairs intelligent management system analyzes the image information and the image features in the reference verification data through the verification feature recognition model, and generates verification results for all image information in each process entity node; In this embodiment, the government affairs intelligent management system first determines the verification feature recognition model and reference verification data based on the primary knowledge graph and the material requirement type tag. Specifically, the system uses the material requirement type tag as an entity to locate the corresponding verification feature recognition model and reference verification data within the primary knowledge graph. For example, if the material requirement type tag is a government affairs sales contract, the location of the verification feature recognition model (government affairs sales contract recognition model) and the location of the reference verification data (historical government affairs sales contracts) are determined from the primary knowledge graph using this tag. The government affairs intelligent management system then invokes this model and retrieves historical contract data from the database.

[0048] Next, the government affairs intelligent management system uses a verification feature recognition model to analyze the image information and the image features in the reference verification data to determine whether the image information and the reference verification data have all the necessary features. If they do, the image information is confirmed to be a qualified image file. Let's further illustrate this using the example above: confirming that the image data is indeed a government affairs contract.

[0049] The government affairs intelligent management system needs to generate verification results for all image information in each process entity node. In other words, each image information on this process entity node needs to be confirmed to be compliant. If it does not comply, the model and reference verification data need to be replaced and re-analyzed to determine whether each image information in each process entity node has the file features corresponding to the process entity node. If any corresponding reference file in the process entity node is not available, a verification result indicating that the image information does not comply with the process entity node needs to be given. If it complies, a confidence value for the image information needs to be generated based on the number of compliant feature points and placed in the verification result. The confidence value is a parameter greater than 0 and less than 1.

[0050] 303. The government affairs intelligent management system generates node simulation route parameters for each process entity node based on all verification results in each process entity node; The intelligent government management system generates node simulation path parameters for each process entity node based on all verification results within that node. This involves integrating and analyzing the confidence data for all image data within each node to determine whether the required data for each node is complete. This in turn determines the difficulty of completing that process entity node and generates node simulation path parameters based on the difficulty. The greater the difficulty, the larger the node simulation path parameters. For example, consider a process entity node containing items such as a sales contract, a commission contract, and a stamp from a relevant entity. The image data contains a contract, a stamp, and a blank image (the blank image represents an unsubmitted item). The system verifies the contract and stamp to determine whether it is a sales contract or a commission contract, and whether the stamp is from the relevant entity. The resulting verification results are: the probability (confidence parameter) that the contract is a sales contract or a commission contract, and the probability that the stamp is from the relevant entity.

[0051]

[0052] in, Simulate distance parameters for nodes, is the number of images with verification results, is the confidence parameter of the i-th image with verification results, is the number of images without verification results, is the parameter of the jth image that does not have a verification result, which is usually set to 1 directly. , but it is generally set to a fixed value higher than 0.95.

[0053] 304. The intelligent government affairs management system adjusts the execution progress of the process entity nodes according to the node simulation route parameters.

[0054] The government affairs intelligent management system adjusts the execution progress of the process entity nodes according to the node simulation path parameters. The formula is as follows:

[0055] in, is the execution progress, S is the standard value, and all process entity nodes are the same. By calculating the execution progress for each process entity node using image features, we can better generate an execution progress that is more consistent with the current scenario for each process entity node. This allows the subsequent ant algorithm to highlight the completion difficulty of each process entity node and identify the process entity node that best meets the government request.

[0056] See also Figure 4, the present application provides an embodiment of a method for adjusting historical usage priority, comprising: 401. The intelligent government management system determines the operation history information of each process entity node from the primary knowledge graph; 402. The government affairs intelligent management system generates a utilization rate parameter for each process entity node based on the operation history information of each process entity node; 403. The intelligent government management system adjusts the historical usage priority of each process entity node according to the usage rate parameters.

[0057] In this embodiment, the government affairs intelligent management system determines the operation history information of each process entity node from the primary knowledge graph. The operation history information of each process entity node can be found through the primary knowledge graph. Then, the application frequency of each process entity node is determined based on the operation history information. The higher the application frequency, the larger the utilization parameter of the process entity node. At this time, it is only necessary to adjust the historical usage priority of the process entity node based on the size of the utilization parameter. The government affairs intelligent management system can adjust the historical usage priority of each process entity node based on the utilization parameter. The formula is as follows:

[0058] in, Priority for historical use, is the usage parameter, which is a value greater than 0 and less than 1. The initial usage priority before adjustment is usually set to 1. The historical usage priority adjustment method of this embodiment can be combined and calculated based on the applicable data of the process entity node in the node's knowledge graph. The ant colony process path model will be more consistent with the actual scenario and can determine the process entity node that is more consistent with the government request.

[0059] See also Figure 5 , this application provides an embodiment of a method for selecting a sub-layer knowledge graph, including: 501. The government affairs intelligent management system determines the superior reference data for constructing the primary knowledge graph according to the first target ant colony process path; 502. The government affairs intelligent management system determines the lower-level reference data in the primary knowledge graph based on the upper-level reference data and the government affairs regional information of the government affairs request; 503. The intelligent government management system selects the sub-layer knowledge graph with the highest correlation based on the lower-level reference data.

[0060] In this embodiment, the government affairs intelligent management system determines the superior reference data for constructing the primary knowledge graph based on the first target ant colony process path, that is, it searches for relevant policy documents and other data for constructing this process path from the primary knowledge graph based on the first target ant colony process path. The government affairs intelligent management system then determines the government affairs area information based on the government affairs request, that is, it determines the service area of ​​the government affairs request, and then finds the corresponding subordinate reference data in the reference files published by the region itself based on the obtained superior reference data. The government affairs intelligent management system selects the sub-layer knowledge graph with the greatest correlation based on the subordinate reference data, that is, it determines the sub-layer knowledge graph constructed using these files, and selects the sub-layer knowledge graph with the highest correlation. If among the many sub-layer knowledge graphs in the region, there is a knowledge graph that uses part or all of the subordinate reference data in the construction process, then one or two sub-layer knowledge graphs with the greatest correlation are selected to improve the accuracy of subsequent services.

[0061] See also Figure 6 The present application provides another embodiment of a method for determining a process entity node, including: 601. The government affairs intelligent management system performs semantic analysis on the text information in the government affairs request and determines several initial entity processing nodes from the primary knowledge graph; 602. The intelligent government management system performs peer node expansion processing based on the initial entity processing node; 603. The government affairs intelligent management system performs a process entity node expansion based on the initial entity processing node, determines the next level process entity node, and repeatedly performs process entity node expansion on the process entity node to generate a process entity node set; 604. The intelligent government affairs management system combines a number of initial entity processing nodes and process entity nodes to generate a number of process entity nodes.

[0062] In this embodiment, the government affairs intelligent management system performs semantic analysis on the text information in the government affairs request, and then determines several initial entity processing nodes from the primary knowledge graph based on the keywords obtained from the semantic analysis results. The government affairs intelligent management system then performs same-level node expansion processing based on the initial entity processing nodes, that is, analyzes the initial node to determine whether there are nodes with the same level and similar government affairs types. The government affairs intelligent management system then performs a process entity node expansion based on the initial entity processing node, determines the next-level process entity node, and repeats the process node expansion on the process entity node to generate a process entity node set, that is, constructs a process network for the initial entity processing node, determines a government affairs process network composed of multiple nodes, integrates all nodes in the government affairs process network, and merges several initial entity processing nodes and process entity node sets to generate several process entity nodes.

[0063] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0064] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0065] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0066] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0067] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk, and other media that can store program code.

Claims

1. The government affairs intelligent management system based on knowledge graph is characterized by: include: The government affairs intelligent management system determines several process entity nodes in the primary knowledge graph based on the text information in the government affairs request; The government affairs intelligent management system constructs an ant colony process path model according to the process sequence of a plurality of process entity nodes, wherein the ant colony process path model contains a plurality of process paths composed of a plurality of process entity nodes; The government affairs intelligent management system determines the process entity node corresponding to each piece of image information in the government affairs request in the ant colony process path model; The government affairs intelligent management system extracts image features from the image information and generates the execution progress of the corresponding process entity node based on the image features; The government affairs intelligent management system calculates the historical usage priority of each process entity node; The government affairs intelligent management system assigns a priority value to each process entity node based on the ant colony algorithm, execution progress, and historical usage priority, and determines a first target ant colony process path from the ant colony process path model, wherein the sum of the priority values ​​assigned to all process entity nodes in the first target ant colony process path is higher than that of other process paths; The government affairs intelligent management system selects a sub-layer knowledge graph according to the first target ant colony process path and the government affairs request; The government affairs intelligent management system generates a new ant colony process path model according to the sub-layer knowledge graph and the government affairs request, and assigns priorities to the process entity nodes in the ant colony process path model to generate a second target ant colony process path; The government affairs intelligent management system processes the government affairs request according to the second target ant colony process path.

2. The intelligent government affairs management system according to claim 1, characterized in that: The government affairs intelligent management system determines the process entity node corresponding to each piece of image information in the ant colony process path model, including: The intelligent government affairs management system determines an attachment type tag of the image information; The government affairs intelligent management system determines a material requirement type label for each process entity node based on the primary knowledge graph, and each process entity node includes at least one material requirement type label; The government affairs intelligent management system matches the attachment type label with the material requirement type label so that each piece of image information is matched to a corresponding process entity node in the ant colony process path model.

3. The intelligent government affairs management system according to claim 2, characterized in that: The intelligent government affairs management system extracts image features from the image information and generates the execution progress of the corresponding process entity node based on the image features, including: The government affairs intelligent management system determines a verification feature recognition model and reference verification data based on the primary knowledge graph and the material requirement type label; The intelligent government affairs management system analyzes the image information and the image features in the reference verification data through the verification feature recognition model, and generates a verification result for all the image information in each process entity node; The government affairs intelligent management system generates node simulation path parameters for each process entity node according to all verification results in each process entity node; The government affairs intelligent management system adjusts the execution progress of the process entity node according to the node simulation path parameters.

4. The intelligent government affairs management system according to claim 1, characterized in that: The government affairs intelligent management system calculates the historical usage priority of each process entity node, including: The government affairs intelligent management system determines the operation history information of each process entity node from the primary knowledge graph; The government affairs intelligent management system generates a utilization rate parameter for each process entity node based on the operation history information of each process entity node; The government affairs intelligent management system adjusts the historical usage priority of each process entity node according to the usage rate parameter.

5. The intelligent government affairs management system according to any one of claims 1 to 4, characterized in that: The government affairs intelligent management system selects a sub-layer knowledge graph according to the first target ant colony process path and the government affairs request, including: The government affairs intelligent management system determines the upper-level reference data for constructing the primary knowledge graph according to the first target ant colony process path; The government affairs intelligent management system determines the lower-level reference data in the primary knowledge graph based on the upper-level reference data and the government affairs area information of the government affairs request; The government affairs intelligent management system selects the sub-layer knowledge graph with the greatest correlation based on the subordinate reference data.

6. The intelligent government affairs management system according to any one of claims 1 to 4, characterized in that: The government affairs intelligent management system determines several process entity nodes in the primary knowledge graph based on the text information in the government affairs request, including: The government affairs intelligent management system performs semantic analysis on the text information in the government affairs request and determines several initial entity processing nodes from the primary knowledge graph; The government affairs intelligent management system performs peer node expansion processing based on the initial entity processing node; The government affairs intelligent management system performs a process entity node expansion based on the initial entity processing node, determines the next level process entity node, and repeatedly performs process entity node expansion on the process entity node to generate a process entity node set; The government affairs intelligent management system combines a number of initial entity processing nodes and process entity node sets to generate a number of process entity nodes.

Images