Government affair knowledge graph construction system based on artificial intelligence

Abstract

The application relates to the field of artificial intelligence, in particular to a government affair knowledge graph construction system based on artificial intelligence. The system comprises a sample collection module, an information receiving module, an information processing module and a verification module. Sample data information is collected by the sample collection module, and a knowledge graph database is constructed; actual data information uploaded by a user end is acquired by the information receiving module; the similarity of the sample data information and the actual data information is calculated by the information processing module; whether the knowledge graph database is abnormal is determined based on the similarity by the verification module; and whether the sample data information of the knowledge graph database is supplemented is determined. Through the cooperative work of the above modules, the application reduces the occurrence of the data integrity misjudgment of the knowledge graph caused by network lag, and can provide accurate government affair service recommendation, regulation interpretation and policy analysis, and improves the accuracy and convenience of government information query and intelligent question answering.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based system for constructing government knowledge graphs. Background Technology

[0002] In the development of e-government, knowledge graphs serve as a crucial tool for integrating and managing massive amounts of government data, supporting government decision-making and information retrieval. As the level of e-government informatization continues to increase, more efficient integration and utilization methods are needed. Knowledge graphs, by constructing relationships between entities, have become a key technological means to break down data silos and enhance data value. Currently, the application potential of knowledge graphs in scenarios such as government decision support, intelligent question answering, and process optimization is widely recognized.

[0003] However, faced with massive amounts of scattered government data, the construction and application of existing government knowledge graphs still rely on traditional data collection methods. A large amount of historical government data still requires manual or semi-manual processing, significantly impacting the comprehensiveness and timeliness of data collection. Existing knowledge graphs lack efficient real-time verification and automatic correction mechanisms when facing dynamically updated government needs, resulting in problems such as data lag and inaccurate relationships in some knowledge graphs.

[0004] Patent application publication number CN 114119317 A discloses a method for constructing a knowledge graph based on government service scenarios, belonging to the field of government service technology. This method employs a pre-built knowledge graph data model, performing multi-dimensional analysis of departmental business data and business entity objects in advance. It sorts out common object data and relationship information among various government departments and pre-builds them into the model layer and data layer of the knowledge graph, realizing the association between real-world business and the intelligent knowledge graph, making it ready to use for general scenarios. Simultaneously, it establishes personalized relationship graphs for different regional business and personalized needs, providing users with personalized services and decision support. This invention establishes personalized relationship graphs for different regional business and personalized needs, enabling the provision of personalized services and decision support to users.

[0005] Therefore, the invention has the following problems:

[0006] This invention does not take into account the possibility of network lag after the government knowledge graph is built, which may lead to misjudgment of the knowledge graph integrity verification. As a result, it has shortcomings in timeliness and data integrity when facing dynamically updated government needs. Summary of the Invention

[0007] To address this, the present invention provides an artificial intelligence-based government knowledge graph construction system to overcome the shortcomings of existing technologies that fail to consider the possibility of misjudgment of the completeness of the knowledge graph due to network lag during the verification of government knowledge graph construction, resulting in insufficient timeliness and data integrity when facing dynamically updated government needs.

[0008] To achieve the above objectives, the present invention provides an artificial intelligence-based government knowledge graph construction system, comprising:

[0009] The sample acquisition module is used to collect sample data information of sample images of different keywords within a historical period, and to construct a knowledge graph database based on the sample data information of sample images of different keywords within a historical period.

[0010] The information receiving module is used to obtain the actual data information of the actual image of the verification keyword uploaded by the user and the actual image of each document URL ID corresponding to the actual image of the keyword.

[0011] An information processing module, which is connected to the sample acquisition module and the information receiving module respectively, is used to obtain corresponding sample data information based on the actual image of the keyword uploaded by the user terminal, compare the sample data information with the actual data information, and calculate the similarity.

[0012] A verification module, connected to the information processing module, the information receiving module, and the sample acquisition module, is used to obtain the similarity and to determine whether there are anomalies in the knowledge graph database based on the difference between a preset similarity threshold and the similarity; including:

[0013] If the knowledge graph database is abnormal, the system network lag information is extracted, and the network status is determined to be abnormal based on the network lag information.

[0014] The sample data information includes: keyword sample images and mapping sample images formed by the document URL ID sample images corresponding to the keyword sample images.

[0015] Furthermore, the information processing module is used to obtain corresponding sample data information based on the actual image of the keyword uploaded by the user, compare the sample data information with the actual data information, and calculate the similarity, including:

[0016] Extract the keyword sample image and the corresponding document URL ID sample image to form the key point information of the mapped sample image, which serves as the contour marker of the sample image;

[0017] Extract the actual image of the keyword and the actual image of the document URL ID corresponding to the actual image of the keyword to form the key point information of the actual image, which serves as the outline marker of the actual image;

[0018] Based on the sample image contour markers and the actual image contour markers, an image overlap comparison is performed, and the ratio of the area of ​​the overlapping image contours to the total area of ​​the actual image contours is determined as the similarity.

[0019] Furthermore, the verification module is used to determine whether there are anomalies in the knowledge graph database based on the difference between a preset similarity threshold and the similarity, including:

[0020] The difference between the preset similarity threshold and the similarity is used to calculate and determine the similarity difference.

[0021] If the similarity difference is greater than the predetermined similarity difference threshold, it is determined that there is an anomaly.

[0022] If the similarity difference is less than or equal to the predetermined similarity difference threshold, it is determined that there is no anomaly.

[0023] Furthermore, the verification module is used to verify the operating status of the knowledge graph database based on the judgment result, including:

[0024] If the knowledge graph database is abnormal, the system network lag information is extracted, and the network status is determined to be abnormal based on the network lag information.

[0025] Furthermore, the verification module is used to extract system network congestion information when anomalies exist in the knowledge graph database, and to determine whether there are network status anomalies based on the network congestion information, including:

[0026] Used to extract the number of network lags within a historical period;

[0027] If the number of network lags exceeds the predetermined network lag threshold, it is determined that there is an abnormality in the network status;

[0028] If the number of network lags is less than or equal to the predetermined network lag threshold, the network status is determined to be normal.

[0029] Furthermore, the verification module is used to switch to a backup network channel when there is an anomaly in the network status.

[0030] Furthermore, the verification module is used to re-determine whether there are any anomalies in the knowledge graph database after switching to the backup channel.

[0031] Furthermore, the verification module is used to re-determine whether there are any abnormalities in the knowledge graph database. If there are still abnormalities, it is determined that the sample data is incomplete.

[0032] Furthermore, the verification module is used to supplement the knowledge graph database sample data information when it is determined that the sample data is incomplete.

[0033] Furthermore, the verification module is used to supplement the sample data information in the knowledge graph database, including:

[0034] Used to extract actual data information of the actual image of the verification keyword uploaded by the user and the actual image of each document URL ID corresponding to the actual image of the keyword;

[0035] Based on the actual data information of the verification keyword actual image and the document URL ID actual image of the corresponding keyword actual image uploaded by the user terminal, the data is uploaded to the sample collection module to cover the corresponding sample data information.

[0036] Compared with existing technologies, the beneficial effects of this invention are that it provides an artificial intelligence-based government knowledge graph construction system. Through a sample acquisition module, sample data information of different keyword sample images within a historical period is collected to construct a knowledge graph database. Through an information receiving module, actual data information of the actual images of verification keywords uploaded by the user terminal and the corresponding document URL ID sample images is obtained, thus providing a verification basis for the existing knowledge graph database. Through an information processing module, the corresponding actual data information of the keyword actual images uploaded by the user terminal is obtained and compared with the sample data information to calculate the similarity, quantifying the difference between the sample data information and the actual data information. Through a verification module, the similarity is obtained to verify whether there are any anomalies in the knowledge graph database. If anomalies are found, the influence of network conditions can be ruled out by determining whether the network status is abnormal. If network anomalies are found, after switching to a backup network channel, a second determination of whether there are any anomalies in the knowledge graph database can be made to clarify whether the sample data construction is complete, and thus determine whether the sample data information of the knowledge graph database needs to be supplemented. This invention, through the collaborative work between various modules, reduces the occurrence of misjudgments of data integrity due to network lag in knowledge graphs, thereby providing accurate government service recommendations, legal interpretations, and policy analysis, and improving the accuracy and convenience of government information queries and intelligent question answering.

[0037] In particular, this invention, through its information processing module, compares the sample data with the actual data to calculate similarity. This precise comparison of key points effectively improves the accuracy and efficiency of government data verification. By extracting image contour markers from the sample and actual images, the identification elements of government documents can be accurately reflected. The specific numerical value of similarity is determined by calculating the ratio of the area of ​​overlapping contours in the images to the total area of ​​the actual image contours, providing a quantitative basis for comparison.

[0038] In particular, the present invention uses a verification module to determine whether there are anomalies in the knowledge graph database by comparing the difference between a preset similarity threshold and the similarity. This can intuitively reflect the degree of fit between the database data and the actual situation, reduce the occurrence of invalid investigations caused by misjudgment, reduce the probability of data errors caused by missed judgment, and significantly improve the accuracy and response speed of anomaly detection in the knowledge graph database.

[0039] In particular, this invention uses a verification module to verify the operational status of the knowledge graph database based on the judgment results. If an anomaly is found in the knowledge graph database, system network congestion is extracted, and the network status is determined to be abnormal based on the network congestion information. By extracting the number of system network congestion instances and comparing it with a preset system network congestion threshold, the presence of network anomalies can be accurately determined. When the number of congestion instances exceeds the threshold, it is determined to be a network anomaly. By switching to a backup network channel, the stability of the data acquisition process is ensured, and a secondary anomaly detection of the knowledge graph database is initiated. When the number of congestion instances is within the normal range, network factors are quickly eliminated, allowing the focus to be placed on the completeness of the sample data construction and improving the efficiency of anomaly handling. Attached Figure Description

[0040] Figure 1 This is a structural block diagram of the government knowledge graph construction system based on artificial intelligence, according to an embodiment of the present invention.

[0041] Figure 2 This invention provides a logical judgment graph for calculating similarity by comparing the sample data information with the actual data information based on the actual image of the keyword uploaded by the user.

[0042] Figure 3 This invention provides a logical judgment graph for determining whether an anomaly exists in a knowledge graph database based on the difference between a preset similarity threshold and the similarity.

[0043] Figure 4 Based on the determination result, this embodiment of the invention verifies the logical determination graph of the knowledge graph database's operating status. Detailed Implementation

[0044] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0045] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0046] Please see Figure 1 The diagram shown is a structural block diagram of an artificial intelligence-based government knowledge graph construction system according to an embodiment of the present invention. The present invention provides an artificial intelligence-based government knowledge graph construction system, comprising:

[0047] The sample acquisition module is used to collect sample data information of sample images of different keywords within a historical period, and to construct a knowledge graph database based on the sample data information of sample images of different keywords within a historical period.

[0048] The information receiving module is used to obtain the actual data information of the actual image of the verification keyword uploaded by the user and the actual image of each document URL ID corresponding to the actual image of the keyword.

[0049] An information processing module, which is connected to the sample acquisition module and the information receiving module respectively, is used to obtain corresponding sample data information based on the actual image of the keyword uploaded by the user terminal, compare the sample data information with the actual data information, and calculate the similarity.

[0050] A verification module, connected to the information processing module, the information receiving module, and the sample acquisition module, is used to obtain the similarity and to determine whether there are anomalies in the knowledge graph database based on the difference between a preset similarity threshold and the similarity; including:

[0051] If the knowledge graph database is abnormal, the system network lag information is extracted, and the network status is determined to be abnormal based on the network lag information.

[0052] The sample data information includes: keyword sample images and mapping sample images formed by the document URL ID sample images corresponding to the keyword sample images.

[0053] During implementation, if the network status is abnormal, the system will switch to the backup network channel, and after the switch is completed, it will re-evaluate whether there is any abnormality in the knowledge graph database in order to determine whether to supplement the knowledge graph database sample data information.

[0054] This invention constructs a knowledge graph database by collecting sample data from different keyword sample images within a historical period through a sample acquisition module. An information receiving module acquires the actual data of the verification keyword images uploaded by the user and the corresponding document URL ID sample images, providing a verification basis for the existing knowledge graph database. An information processing module obtains the corresponding actual data from the keyword images uploaded by the user and compares it with the sample data to calculate similarity, quantifying the difference between the sample data and the actual data. A verification module uses the similarity score to verify whether the knowledge graph database contains anomalies. If anomalies are found, network status can be checked to rule out network-related issues. If network anomalies are present, a second check of the knowledge graph database after switching to a backup network channel clarifies the completeness of the sample data construction and determines whether additional sample data is needed. This invention, through the collaborative work between various modules, reduces the occurrence of misjudgments of data integrity caused by network lag in knowledge graphs, significantly improving the reliability and data integrity of knowledge graphs. This, in turn, can provide accurate government service recommendations, legal interpretations, and policy analysis, and enhance the accuracy and convenience of government information queries and intelligent question answering.

[0055] Please see Figure 2 As shown, this is a logical decision diagram illustrating how the present invention obtains corresponding sample data information based on actual images of keywords uploaded by users, compares the sample data information with the actual data information, and calculates the similarity. The information processing module of the present invention is used to obtain corresponding sample data information based on actual images of keywords uploaded by users, and the process of comparing the sample data information with the actual data information and calculating the similarity includes:

[0056] Extract the keyword sample image and the corresponding document URL ID sample image to form the key point information of the mapped sample image, which serves as the contour marker of the sample image;

[0057] Extract the actual image of the keyword and the actual image of the document URL ID corresponding to the actual image of the keyword to form the key point information of the actual image, which serves as the outline marker of the actual image;

[0058] Based on the sample image contour markers and the actual image contour markers, an image overlap comparison is performed, and the ratio of the area of ​​the overlapping image contours to the total area of ​​the actual image contours is determined as the similarity.

[0059] This invention, through an information processing module, compares the sample data with the actual data to calculate similarity. By precisely comparing key points, it effectively improves the accuracy and processing efficiency of government data verification. By extracting image contour markers from the sample and actual images, it accurately reflects the identification elements of government documents. By calculating the ratio of the area of ​​overlapping contours in the images to the total area of ​​the actual image contours, a specific numerical value of similarity is determined, providing a quantitative basis for comparison.

[0060] Please see Figure 3 As shown, this is a logical judgment diagram of an embodiment of the present invention for determining whether a knowledge graph database has an anomaly based on the difference between a preset similarity threshold and the similarity. The process by which the verification module of the present invention determines whether a knowledge graph database has an anomaly based on the difference between the preset similarity threshold and the similarity includes:

[0061] The difference between the preset similarity threshold and the similarity is used to calculate and determine the similarity difference.

[0062] If the similarity difference is greater than the predetermined similarity difference threshold, it is determined that there is an anomaly.

[0063] If the similarity difference is less than or equal to the predetermined similarity difference threshold, it is determined that there is no anomaly.

[0064] In practice, the similarity threshold is pre-determined and set between [97%, 100%].

[0065] In practice, the similarity difference threshold is obtained in advance. Under the condition that there are no abnormalities in the knowledge graph database within 3 months, the average of the similarity differences is used as the similarity difference threshold.

[0066] This invention uses a verification module to determine whether there are anomalies in a knowledge graph database by comparing the difference between a preset similarity threshold and the actual similarity. This can intuitively reflect the degree of consistency between the database data and the actual situation, reduce the occurrence of invalid investigations due to misjudgment, and lower the probability of data errors caused by missed judgments. It significantly improves the accuracy and response speed of anomaly detection in knowledge graph databases.

[0067] Please see Figure 4 As shown, this is a logical decision diagram for verifying the operating status of a knowledge graph database based on the determination result in an embodiment of the present invention. The verification module of the present invention verifies the operating status of the knowledge graph database based on the determination result, and the process includes:

[0068] If the knowledge graph database is abnormal, the system network lag information is extracted, and the network status is determined to be abnormal based on the network lag information.

[0069] This invention uses a verification module to verify the operational status of a knowledge graph database based on the judgment results. If an anomaly is found in the knowledge graph database, system network congestion is extracted, and the network status is determined to be abnormal based on the network congestion information. By extracting the number of system network congestion instances and comparing it with a preset system network congestion threshold, the existence of network anomalies can be accurately determined. When the number of congestion instances exceeds the threshold, it is determined to be a network anomaly. The system switches to a backup network channel to ensure the stability of the data acquisition process and initiates a secondary anomaly detection for the knowledge graph database. When the number of congestion instances is within the normal range, network factors are quickly ruled out, allowing the focus to be placed on the completeness of the sample data construction and improving the efficiency of anomaly handling.

[0070] Specifically, the verification module is used to extract system network congestion information when anomalies exist in the knowledge graph database, and to determine whether there are network anomalies based on the network congestion information, including:

[0071] Used to extract the number of network lags within a historical period;

[0072] If the number of network lags exceeds the predetermined network lag threshold, it is determined that there is an abnormality in the network status;

[0073] If the number of network lags is less than or equal to the predetermined network lag threshold, the network status is determined to be normal.

[0074] In practice, the threshold for the number of network lags is obtained in advance. The average number of network lags within one month, assuming no network abnormalities, is used as the threshold for the number of network lags.

[0075] Specifically, the verification module is used to switch to a backup network channel when there is an anomaly in the network status.

[0076] Specifically, the verification module is used to re-determine whether there are any anomalies in the knowledge graph database after switching to the backup channel.

[0077] Specifically, the verification module is used to re-determine whether there are any abnormalities in the knowledge graph database. If there are still abnormalities, it is determined that the sample data is incomplete.

[0078] Specifically, the verification module is used to supplement the knowledge graph database sample data information when it is determined that the sample data is incomplete.

[0079] Specifically, the verification module is used to supplement the sample data information in the knowledge graph database, including,

[0080] Used to extract actual data information of the actual image of the verification keyword uploaded by the user and the actual image of each document URL ID corresponding to the actual image of the keyword;

[0081] Based on the actual data information of the verification keyword actual image and the document URL ID actual image of the corresponding keyword actual image uploaded by the user terminal, the data is uploaded to the sample collection module to cover the corresponding sample data information.

[0082] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0083] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A government knowledge graph construction system based on artificial intelligence, characterized in that, include: The sample acquisition module is used to collect sample data information of sample images of different keywords within a historical period, and to construct a knowledge graph database based on the sample data information of sample images of different keywords within a historical period. The information receiving module is used to obtain the actual data information of the actual image of the verification keyword uploaded by the user and the actual image of each document URL ID corresponding to the actual image of the keyword. An information processing module, which is connected to the sample acquisition module and the information receiving module respectively, is used to obtain corresponding sample data information based on the actual image of the keyword uploaded by the user terminal, compare the sample data information with the actual data information, and calculate the similarity. A verification module, connected to the information processing module, the information receiving module, and the sample acquisition module, is used to obtain the similarity and to determine whether there are anomalies in the knowledge graph database based on the difference between a preset similarity threshold and the similarity; including: If the knowledge graph database is abnormal, the system network lag information is extracted, and the network status is determined to be abnormal based on the network lag information. The sample data information includes: keyword sample images and mapping sample images formed by the document URL ID sample images corresponding to the keyword sample images.

2. The government knowledge graph construction system based on artificial intelligence according to claim 1, characterized in that, The information processing module is used to obtain corresponding sample data information based on the actual image of the keyword uploaded by the user terminal, compare the sample data information with the actual data information, and calculate the similarity, including: Extract the keyword sample image and the corresponding document URL ID sample image to form the key point information of the mapped sample image, which serves as the contour marker of the sample image; Extract the actual image of the keyword and the actual image of the document URL ID corresponding to the actual image of the keyword to form the key point information of the actual image, which serves as the outline marker of the actual image; Based on the sample image contour markers and the actual image contour markers, an image overlap comparison is performed, and the ratio of the area of ​​the overlapping image contours to the total area of ​​the actual image contours is determined as the similarity.

3. The government knowledge graph construction system based on artificial intelligence according to claim 1, characterized in that, The verification module is used to determine whether there are anomalies in the knowledge graph database based on the difference between a preset similarity threshold and the similarity, including: The difference between the preset similarity threshold and the similarity is used to calculate and determine the similarity difference. If the similarity difference is greater than the predetermined similarity difference threshold, it is determined that there is an anomaly. If the similarity difference is less than or equal to the predetermined similarity difference threshold, it is determined that there is no anomaly.

4. The government knowledge graph construction system based on artificial intelligence according to claim 3, characterized in that, The verification module is used to verify the operating status of the knowledge graph database based on the determination result of whether there are any anomalies in the knowledge graph database, including: If the knowledge graph database is abnormal, the system network lag information is extracted, and the network status is determined to be abnormal based on the network lag information.

5. The government knowledge graph construction system based on artificial intelligence according to claim 4, characterized in that, The verification module is used to extract system network congestion information when anomalies exist in the knowledge graph database, and to determine whether there are network anomalies based on the network congestion information, including: Used to extract the number of network lags within a historical period; If the number of network lags exceeds the predetermined network lag threshold, it is determined that there is an abnormality in the network status; If the number of network lags is less than or equal to the predetermined network lag threshold, the network status is determined to be normal.

6. The government knowledge graph construction system based on artificial intelligence according to claim 5, characterized in that, The verification module is used to switch to a backup network channel when there is an anomaly in the network status.

7. The government knowledge graph construction system based on artificial intelligence according to claim 6, characterized in that, The verification module is used to re-determine whether there are any anomalies in the knowledge graph database after switching to the backup channel.

8. The government knowledge graph construction system based on artificial intelligence according to claim 7, characterized in that, The verification module is used to re-determine whether there are any abnormalities in the knowledge graph database. If there are still abnormalities, it is determined that the sample data is incomplete.

9. The government knowledge graph construction system based on artificial intelligence according to claim 8, characterized in that, The verification module is used to supplement the knowledge graph database sample data information when it is determined that the sample data is incomplete.

10. The government knowledge graph construction system based on artificial intelligence according to claim 9, characterized in that, The verification module is used to supplement the sample data information in the knowledge graph database, including: Used to extract actual data information of the actual image of the verification keyword uploaded by the user and the actual image of each document URL ID corresponding to the actual image of the keyword; Based on the actual data information of the verification keyword actual image and the document URL ID actual image of the corresponding keyword actual image uploaded by the user terminal, the data is uploaded to the sample collection module to cover the corresponding sample data information.

Images