Intelligent data management system and method based on artificial intelligence

Abstract

The invention discloses an intelligent data management system and method based on artificial intelligence. The system comprises a database, an extraction module I, a detection module, a risk calculation module II, a recording module, an extraction module II, a risk calculation module I, a judgment module, a calculation module I, a calculation module II, a supervision module and an updating module. The invention relates to the technical field of data management, in particular to an intelligent data management system and method based on artificial intelligence. The technical problem to be solved by the invention is to provide an intelligent data management system and method based on artificial intelligence, which can perform intelligent management on data input in an enterprise platform and improve the accuracy of enterprise data management.

Description

Technical Field

[0001] The present invention relates to the technical field of data management, and in particular to an intelligent data management system and method based on artificial intelligence. Background Art

[0002] The enterprise platform system contains a large amount of business data generated by the expansion of business information, such as order data, product data, etc. These data are usually manually entered into the system by responsible personnel. During the entry process, data anomalies will inevitably occur due to manual operations, thus affecting the quality and security of data stored in the enterprise database. In order for the enterprise platform system to operate accurately and effectively, it is often necessary to establish a data management system to control the entered data and the behavior of responsible personnel.

[0003] For data that may have quality problems, the usual way to deal with it is to extract the data in the database in batches, check and correct them one by one. For massive amounts of data, a lot of manpower and material resources are required to check, so as to eliminate abnormal data and reduce enterprise management risks.

[0004] At present, a Chinese invention patent with publication number CN117235063B proposes a data quality management method based on artificial intelligence technology. The method extracts part of the data and determines whether it is abnormal. After determining the responsibility number corresponding to the abnormal data, the method re-extracts part of the data and determines whether it is abnormal. The risk level of the responsibility number is determined based on the results of the two extractions. Based on the risk level, the responsibility number and the data entered are subject to behavioral control to reduce the quality problems of subsequent data entry into the system.

[0005] Regarding the above technical solution, this method only compares the partial data extracted twice to calculate the risk level of the responsible person. The accuracy and reliability of the calculation are low. At the same time, since the data entered by responsible persons with different working hours have different sensitivity levels, it is necessary to intelligently manage the data entered by the responsible persons according to the actual situation, thereby improving the accuracy of enterprise data management. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide an intelligent data management system and method based on artificial intelligence, which can intelligently manage the data entered into the enterprise platform and improve the accuracy and reliability of enterprise data management.

[0007] In the first aspect, the present invention provides an intelligent data management system based on artificial intelligence, which adopts the following technical solutions to achieve the purpose of the invention:

[0008] An intelligent data management system based on artificial intelligence, including the following modules:

[0009] Database: The output end is connected to the input end of Extraction Module I and is used to store the input data;

[0010] Extraction Module I: The input end is connected to the output end of the database, and the output end is connected to the input end of the Detection Module. It is used to extract part of the data in the database within the detection period T as the first extraction data;

[0011] Detection Module: The input end is connected to the output end of Extraction Module I, and the output end is connected to the input end of the Recording Module. It is used to detect whether there is any abnormality in the first extraction data. If there is no abnormality, it executes Extraction Step I. If there is an abnormality, it confirms the responsibility number corresponding to the abnormal data and extracts the data within the detection period T corresponding to this responsibility number as the second extraction data;

[0012] Recording Module: The input end is respectively connected to the output end of the Detection Module and the output end of the Update Module, and the output end is connected to the input end of the Judgment Module. It is used to record that the amount of abnormal data in the second extraction data is N, set the maximum allowable amount of abnormal data as A, set the warning error as B, and B satisfies

[0013] Judgment Module: The input end is connected to the output end of the Recording Module, and the output end is respectively connected to the input ends of Calculation Module I, Calculation Module II, and the Supervision Module. It is used to judge that if N≤B, the risk level of the responsibility number is level three and it executes Calculation Module I. If B<N≤A, the risk level of the responsibility number is level two and it executes Calculation Module II. If N>A, the risk level of the responsibility number is level one and it executes the Supervision Module;

[0014] Calculation Module I: The input end is connected to the output end of the Judgment Module, and the output end is connected to the input end of the Update Module. It is used to make Execute the Update Module;

[0015] Calculation Module II: The input end is connected to the output end of the Judgment Module, and the output end is connected to the input end of the Update Module. It is used to make Execute the Update Module;

[0016] Supervision Module: The input end is connected to the output end of the Judgment Module and is used to make the supervision method of the responsibility number be the interrupt limit input state and monitor the data input by the responsibility number within the detection period T;

[0017] Update Module: The input end is respectively connected to the output ends of Calculation Module I and Calculation Module II, and the output end is connected to the input end of the Recording Module. It is used to update the maximum allowable amount of abnormal data A in the Recording Module and execute Extraction Module I.

[0018] As a further limitation of this technical solution, in the Recording Module, the calculation model of the warning error B is: Where C is the sensitivity constant, and C∈(0,1).

[0019] The updating module is also used to update the warning error B in the recording module.

[0020] As a further limitation of the technical solution, the detection module is also used to extract the working time corresponding to the responsibility number as h;

[0021] The recording module is also used to determine the value of C according to the working time h corresponding to the responsibility number. The calculation model of the sensitivity constant C is: Among them, t1 is the set standard working time, and t2 is the maximum allowed working time.

[0022] As a further limitation of the technical solution, it also includes an extraction module II and a risk calculation module I;

[0023] Extraction module II: The input end is connected to the output end of the recording module, and the output end is connected to the input end of the risk calculation module I. It is used to extract the average abnormal data volume Q of all responsibility numbers with abnormalities in the database in the last detection cycle T, and the average working time is t avg ;

[0024] Risk calculation module I: The input end is connected to the output end of the extraction module II, and the output end is connected to the input end of the database, which is used to judge that if N≤Q, the risk value of the responsibility number is recorded as R=0, and if N>Q, the risk value of the responsibility number is recorded as

[0025] As a further limitation of the technical solution, it also includes a risk calculation module II;

[0026] Risk calculation module II: The input end is connected to the output end of the detection module, and the output end is connected to the input end of the database, which is used to set the cycle threshold T1. If the responsibility number without abnormal data has not extracted secondary sampling data within the cycle threshold T1, the risk value of the recorded responsibility number is R=0.

[0027] In a second aspect, the present invention provides an intelligent data management method based on artificial intelligence, which adopts the following technical solutions to achieve the purpose of the invention:

[0028] An intelligent data management method based on artificial intelligence comprises the following steps:

[0029] Extraction I: extract part of the data in the database within the detection period T as a sampling data;

[0030] Detection: Detect whether there is any abnormality in the once-drawn data. If there is no abnormality, execute the Extraction I step. If there is an abnormality, confirm the responsibility number corresponding to the abnormal data, and extract the data within the detection period T corresponding to this responsibility number as the second-drawn data;

[0031] Recording: Record that the amount of abnormal data in the second-drawn data is N, set the maximum allowable amount of abnormal data as A, set the warning error as B, and B satisfies

[0032] Judgment: If N ≤ B, the risk level of the responsibility number is level three, and execute the Calculation I step. If B < N ≤ A, the risk level of the responsibility number is level two, and execute the Calculation II step. If N > A, the risk level of the responsibility number is level one, and execute the Supervision step;

[0033] Calculation I: Let Execute the Update step;

[0034] Calculation II: Let Execute the Update step;

[0035] Supervision: Let the supervision method of the responsibility number be the interrupt-limited input state, and monitor the data input by the responsibility number within the detection period T;

[0036] Update: Update the maximum allowable amount of abnormal data A in the Recording step, and execute the Extraction I step.

[0037] As a further limitation of this technical solution, in the Recording step, the calculation model of the warning error B is: where C is the sensitivity constant, and C ∈ (0, 1).

[0038] In the Update step, the warning error B in the Recording step is also updated.

[0039] As a further limitation of this technical solution, in the Detection step, the working duration h corresponding to the responsibility number is also extracted;

[0040] In the Recording step, the value of C is also determined according to the working duration h corresponding to the responsibility number. The calculation model of the sensitivity constant C is: where t1 is the set standard working duration and t2 is the maximum allowable working duration.

[0041] As a further limitation of this technical solution, after the Update step, there are also an Extraction II step and a Risk Calculation I step;

[0042] Extraction II: Extract that the average amount of abnormal data of all responsibility numbers with abnormalities in the database in the previous detection period T is Q, and the average working duration is t avg ;

[0043] Risk calculation I: If N≤Q, the risk value of the responsibility number is recorded as R=0; if N>Q, the risk value of the responsibility number is recorded as

[0044] As a further limitation of the technical solution, the risk calculation step I is followed by a risk calculation step II;

[0045] Risk calculation II: Set the cycle threshold T1. If the responsibility number without abnormal data has no secondary sampling data extracted within the cycle threshold T1, the risk value of the recorded responsibility number is R=0.

[0046] Compared with the prior art, the advantages and positive effects of the present invention are:

[0047] 1. By comparing the relationship between the amount of abnormal data in the secondary sampled data of the responsibility number in a detection cycle and the warning error and the maximum allowable amount of abnormal data, the risk level of the responsibility number is divided into three, two and one levels. For the responsibility number with level one risk, behavior control and data monitoring are carried out. For the responsibility numbers with level three and level two risks, the maximum allowable amount of abnormal data of the responsibility number is reduced accordingly with the increase of the detection cycle, so as to timely discover the potential quality risks of the data entered by the responsibility number, so that the data entered by the responsibility number can be managed by artificial intelligence according to the actual work situation, so as to improve the accuracy of enterprise data management, make the enterprise platform run safely and stably, and facilitate the rapid and accurate judgment of the risk level of the responsibility number, and improve the efficiency and reliability of data management.

[0048] 2. By setting the sensitivity constant to calculate the warning error, it is convenient to perform intelligent regulation based on factors such as the actual working situation of the responsibility number and the sensitivity of the input data, and then judge the risk level of the responsibility number and regulate the maximum allowable amount of abnormal data, which is conducive to improving the intelligence level of data management and improving the accuracy and practicality of data management.

[0049] 3. The value of the sensitivity constant is determined by extracting the working hours of the responsibility number. When the working hours of the responsibility number are close to the standard hours, the value of the sensitivity constant approaches 1 to 0, so that the value of the warning error gradually decreases, and then the maximum allowable amount of abnormal data is gradually adjusted, so that the maximum allowable amount of abnormal data for the responsibility number that does not reach the standard working hours is gradually reduced. This setting makes it easy to regulate the risk level of the responsibility number according to the actual working situation of the responsibility number, which is conducive to improving the intelligence of data management and improving the accuracy and practicality of data management.

[0050] 4. By extracting the average abnormal data volume and average working time of all responsibility numbers with abnormalities in the previous detection cycle, the risk value of the responsibility number is calculated. If the abnormal data volume of the responsibility number is lower than the average abnormal data volume, the risk value is set to zero. Otherwise, the risk value is calculated. If the responsibility number without abnormal data does not extract the secondary draw data within the cycle threshold, the risk value of the responsibility number is recorded as 0. This setting makes it easy for management personnel to quickly and intuitively monitor the changes in the risk value of the responsibility number, which is conducive to improving the intelligence and practicality of data management. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0052] Figure 1 This is a system diagram of Embodiment 1 of the present invention;

[0053] Figure 2 This is a flow chart of Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the embodiments of the present invention Figure 1 and Figure 2 , the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] Embodiment 1: An intelligent data management system based on artificial intelligence, comprising the following modules:

[0056] Database: The input end is connected to the output end of the risk calculation module I and the risk calculation module II respectively, and the output end is connected to the input end of the extraction module I, and is used to store the input data;

[0057] Extraction module I: the input end is connected to the output end of the database, and the output end is connected to the input end of the detection module, and is used to extract part of the data in the database within the detection period T as a single sampling data;

[0058] Detection module: Its input end is connected to the output end of the extraction module I, and its output end is respectively connected to the input ends of the recording module and the risk calculation module II. It is used to detect whether there is any abnormality in the first extraction data. If there is no abnormality, the extraction I step is executed. If there is an abnormality, the responsibility number corresponding to the abnormal data is confirmed, and the data within the detection period T corresponding to the responsibility number is extracted as the second extraction data, and the working duration corresponding to the responsibility number is h;

[0059] Risk calculation module II: Its input end is connected to the output end of the detection module, and its output end is connected to the input end of the database. It is used to set the cycle threshold T1. If no second extraction data is extracted within the cycle threshold T1 for the responsibility numbers of the abnormal data, the risk value of the responsibility number is recorded as R = 0;

[0060] Recording module: Its input end is respectively connected to the output end of the detection module and the output end of the update module, and its output end is respectively connected to the input ends of the judgment module and the extraction module II. It is used to record that the amount of abnormal data in the second extraction data is N, set the maximum allowable amount of abnormal data as A, set the warning error as B, and the calculation model of the warning error B is: where C is the sensitivity constant, and the value of C is determined according to the working duration h corresponding to the responsibility number. The calculation model of the sensitivity constant C is: where t1 is the set standard working duration and t2 is the maximum allowable working duration;

[0061] Extraction module II: Its input end is connected to the output end of the recording module, and its output end is connected to the input end of the risk calculation module I. It is used to extract the average amount of abnormal data Q and the average working duration t of all responsibility numbers with abnormalities in the database in the previous detection period T avg ;

[0062] Risk calculation module I: Its input end is connected to the output end of the extraction module II, and its output end is connected to the input end of the database. It is used to judge that if N ≤ Q, the risk value of the responsibility number is recorded as R = 0. If N > Q, the risk value of the responsibility number is recorded as

[0063] Judgment module: Its input end is connected to the output end of the recording module, and its output end is respectively connected to the input ends of the calculation module I, the calculation module II and the supervision module. It is used to judge that if N ≤ B, the risk level of the responsibility number is level three, and the calculation module I is executed. If B < N ≤ A, the risk level of the responsibility number is level two, and the calculation module II is executed. If N > A, the risk level of the responsibility number is level one, and the supervision module is executed;

[0064] Calculation module I: Its input end is connected to the output end of the judgment module, and its output end is connected to the input end of the update module. It is used to make Execute the update module;

[0065] Calculation module II: The input end is connected to the output end of the judgment module, and the output end is connected to the input end of the update module, which is used to make Execute the update module;

[0066] Supervision module: The input end is connected to the output end of the judgment module, and is used to set the supervision mode of the responsibility number to the interruption restriction input state, and monitor the data input by the responsibility number within the detection period T;

[0067] Update module: The input end is connected to the output end of calculation module I and the output end of calculation module II respectively, and the output end is connected to the input end of the recording module, which is used to update the maximum allowable abnormal data amount A and warning error B in the recording module and execute the extraction module I.

[0068] The working principle of an intelligent data management system based on artificial intelligence in this embodiment is as follows:

[0069] The extraction module I extracts part of the data in the database within the detection period T as the primary sampling data, the detection module detects abnormal data and extracts the secondary sampling data of the responsibility number, and the recording module records the amount of abnormal data, the maximum allowable amount of abnormal data, the warning error, the sensitivity constant and other data in the secondary sampling data.

[0070] If the responsibility number without abnormal data has not extracted any secondary drawn-up data within the cycle threshold, the risk calculation module II records the risk value of the responsibility number as zero and passes the data to the database. The extraction module II extracts the average abnormal data volume and average working time of all responsibility numbers with abnormalities in the previous detection cycle T in the database. The risk calculation module I calculates and determines the risk value of the responsibility number and passes the data to the database. The management personnel analyze the risk changes of the responsibility number, so that the management personnel can quickly and intuitively monitor the risk value changes of the responsibility number through artificial intelligence calculation methods.

[0071] The judgment module judges the relationship between the amount of abnormal data in the secondary sampling data and the warning error and the maximum allowable amount of abnormal data, and divides the risk level of the responsibility number into three levels, two levels, and one level. The supervision module performs behavioral control and data monitoring on the responsibility number of the first-level risk. The calculation module I recalculates the maximum allowable amount of abnormal data for the responsibility number of the third-level risk and passes it to the update module. The calculation module II recalculates the maximum allowable amount of abnormal data for the responsibility number of the second-level risk and passes it to the update module. The update module updates the maximum allowable amount of abnormal data and the warning error in the recording module, and executes the extraction module I, so that as the detection cycle increases, the maximum allowable amount of abnormal data for the responsibility number decreases accordingly, so as to facilitate the timely discovery of potential quality risks in the data entered into the responsibility number.

[0072] With such settings, the warning error is calculated by setting the sensitivity constant, and the working hours of the responsible number are extracted to determine the value of the sensitivity constant. When the working hours of the responsible number are close to the standard hours, the value of the sensitivity constant approaches from 1 to 0, so that the value of the warning error gradually decreases, and then the maximum allowable abnormal data volume is gradually adjusted, so that the maximum allowable abnormal data volume of the responsible number that has not reached the standard working hours gradually decreases, which is convenient for intelligent control according to factors such as the actual working conditions of the responsible number and the sensitivity of the input data, conducive to improving the intelligence level of data management, improving the accuracy of enterprise data management, enabling the enterprise platform to operate safely and stably, and improving the efficiency and reliability of data management.

[0073] Embodiment 2: Based on the intelligent data management system based on artificial intelligence provided in the first embodiment, the second embodiment of the present application proposes an intelligent data management method based on artificial intelligence, including the following steps:

[0074] S1. Extraction I: Extract part of the data in the database within the detection period T as the first extraction data;

[0075] S2. Detection: Detect whether there is an abnormality in the first extraction data. If there is no abnormality, execute the extraction I step. If there is an abnormality, confirm the responsible number corresponding to the abnormal data, extract the data within the detection period T corresponding to the responsible number as the second extraction data, and extract the working hours corresponding to the responsible number as h;

[0076] S3. Recording: Record the abnormal data volume in the second extraction data as N, set the maximum allowable abnormal data volume as A, set the warning error as B, and the calculation model of the warning error B is: where C is the sensitivity constant, and the value of C is determined according to the working hours h corresponding to the responsible number. The calculation model of the sensitivity constant C is: where t1 is the set standard working hours and t2 is the maximum allowable working hours.

[0077] S4. Judgment: If N≤B, the risk level of the responsible number is level three, and execute the calculation I step. If B<N≤A, the risk level of the responsible number is level two, and execute the calculation II step. If N>A, the risk level of the responsible number is level one, and execute the supervision step;

[0078] S5. Calculation I: Let Execute the update step;

[0079] S6. Calculation II: Let Execute the update step;

[0080] S7. Supervision: Set the supervision mode of the responsibility number to the interruption restriction input state, and monitor the data input by the responsibility number within the detection period T.

[0081] S8. Update: Update the maximum allowable abnormal data volume A and the warning error B in the recording step, and execute the extraction I step.

[0082] S9, Extraction II: Extract the average abnormal data volume of all responsibility numbers with abnormalities in the last detection cycle T from the database, which is Q, and the average working time is t avg ;

[0083] S10, Risk Calculation I: If N≤Q, the risk value of the responsibility number is recorded as R=0; if N>Q, the risk value of the responsibility number is recorded as

[0084] S11. Risk calculation II: Set the cycle threshold T1. If the responsibility number without abnormal data has no secondary sampling data extracted within the cycle threshold T1, the risk value of the recorded responsibility number is R=0.

[0085] The working principle of an intelligent data management method based on artificial intelligence in this embodiment is as follows:

[0086] Extract and compare the relationship between the amount of abnormal data in the secondary sampling data of the responsibility number in a detection cycle, the warning error, and the maximum allowable amount of abnormal data, and divide the risk level of the responsibility number into three, two, and one levels. For the responsibility number with level one risk, behavioral control and data monitoring are carried out. For the responsibility numbers with level three and level two risks, the maximum allowable amount of abnormal data of the responsibility number is reduced accordingly with the increase of the detection cycle, so as to timely discover the potential quality risks of the data entered into the responsibility number.

[0087] The warning error is calculated by setting a sensitivity constant, and the working hours of the responsibility number are extracted to determine the value of the sensitivity constant. When the working hours of the responsibility number are close to the standard hours, the value of the sensitivity constant approaches from 1 to 0, so that the value of the warning error gradually decreases, and then the maximum allowable amount of abnormal data is gradually adjusted, so that the maximum allowable amount of abnormal data for the responsibility number that has not reached the standard working hours is gradually reduced, which is convenient for intelligent regulation based on factors such as the actual working situation of the responsibility number and the sensitivity of the input data.

[0088] Extract the average abnormal data volume and average working hours of all responsibility numbers with abnormalities in the previous detection cycle, and calculate the risk value of the responsibility number. If the abnormal data volume of the responsibility number is lower than the average abnormal data volume, the risk value is set to zero. Otherwise, the risk value is calculated and passed to the management personnel to determine the risk changes of the responsibility number. If the responsibility number without abnormal data has not extracted the secondary draw data within the cycle threshold, the risk value of the responsibility number is recorded as 0.

[0089] Such a setting makes it convenient to intelligently adjust the risk level of the responsibility number according to the actual working situation of the responsibility number, and to adjust the maximum allowable amount of abnormal data. It is also convenient for managers to quickly and intuitively monitor the changes in the risk value of the responsibility number through artificial intelligence calculations, which is conducive to improving the intelligence level of data management, improving the accuracy of enterprise data management, making the enterprise platform run safely and stably, and improving the efficiency and reliability of data management.

[0090] The above disclosure is only a specific embodiment of the present invention, but the present invention is not limited thereto, and any changes that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. An intelligent data management system based on artificial intelligence, characterized in that: Includes the following modules: Database: The output end is connected to the input end of the extraction module I and is used to store the entered data; Extraction module I: the input end is connected to the output end of the database, and the output end is connected to the input end of the detection module, and is used to extract part of the data in the database within the detection period T as a single sampling data; Detection module: The input end is connected to the output end of the extraction module I, and the output end is connected to the input end of the recording module, and is used to detect whether there is an abnormality in the primary extraction data. If there is no abnormality, the extraction step I is executed. If there is an abnormality, the responsibility number corresponding to the abnormal data is confirmed, and the data within the detection period T corresponding to the responsibility number is extracted as the secondary extraction data; Recording module: The input end is connected to the output end of the detection module and the output end of the update module respectively, and the output end is connected to the input end of the judgment module, which is used to record the amount of abnormal data in the secondary sampling data as N, set the maximum allowable amount of abnormal data as A, set the warning error as B, and B satisfies Judgment module: the input end is connected to the output end of the recording module, and the output end is connected to the input end of the calculation module I, the calculation module II and the supervision module respectively, and is used to judge that if N≤B, the risk level of the responsibility number is level three, and the calculation module I is executed; if B<N≤A, the risk level of the responsibility number is level two, and the calculation module II is executed; if N>AA, the risk level of the responsibility number is level one, and the supervision module is executed; Calculation module I: The input end is connected to the output end of the judgment module, and the output end is connected to the input end of the update module, which is used to make Execute the update module; Calculation module II: The input end is connected to the output end of the judgment module, and the output end is connected to the input end of the update module, which is used to make Execute the update module; Supervision module: The input end is connected to the output end of the judgment module, and is used to set the supervision mode of the responsibility number to the interruption restriction input state, and monitor the data input by the responsibility number within the detection period T; Update module: The input end is connected to the output end of calculation module I and the output end of calculation module II respectively, and the output end is connected to the input end of the recording module, which is used to update the maximum allowable abnormal data amount A in the recording module and execute the extraction module I.

2. The intelligent data management system based on artificial intelligence according to claim 1, characterized in that: In the recording module, the calculation model of the warning error B is: Where C is the sensitivity constant, and C∈(0,1); The updating module is also used to update the warning error B in the recording module.

3. The intelligent data management system based on artificial intelligence according to claim 2, characterized in that: The detection module is also used to extract the working time corresponding to the responsibility number as h; The recording module is also used to determine the value of C according to the working time h corresponding to the responsibility number. The calculation model of the sensitivity constant C is: Among them, t1 is the set standard working time, and t2 is the maximum allowed working time.

4. The intelligent data management system based on artificial intelligence according to claim 3 is characterized by: It also includes extraction module II and risk calculation module I; Extraction module II: The input end is connected to the output end of the recording module, and the output end is connected to the input end of the risk calculation module I. It is used to extract the average abnormal data volume Q of all responsibility numbers with abnormalities in the database in the last detection cycle T, and the average working time is t avg ; Risk calculation module I: The input end is connected to the output end of the extraction module II, and the output end is connected to the input end of the database, which is used to judge that if N≤Q, the risk value of the responsibility number is recorded as R=0, and if N>Q, the risk value of the responsibility number is recorded as 5. The intelligent data management system based on artificial intelligence according to claim 4, characterized in that: Also included is Risk Calculation Module II; Risk calculation module II: The input end is connected to the output end of the detection module, and the output end is connected to the input end of the database, which is used to set the cycle threshold T1. If the responsibility number without abnormal data has not extracted secondary sampling data within the cycle threshold T1, the risk value of the recorded responsibility number is R=0.

6. An intelligent data management method based on artificial intelligence, characterized in that: The following steps are involved: Extraction I: extract part of the data in the database within the detection period T as a sampling data; Detection: Detect whether there is any abnormality in the first-stage sampling data. If there is no abnormality, perform the extraction step I. If there is an abnormality, confirm the responsibility number corresponding to the abnormal data, and extract the data within the detection period T corresponding to the responsibility number as the second-stage sampling data; Record: Record the amount of abnormal data in the secondary sampling data as N, set the maximum allowable amount of abnormal data as A, set the warning error as B, and B satisfies Judgment: If N≤B, the risk level of the responsibility number is level 3, and the calculation step I is executed; if B<N≤A, the risk level of the responsibility number is level 2, and the calculation step II is executed; if N>A, the risk level of the responsibility number is level 1, and the supervision step is executed; Calculation I: Let Execute the update step; Computation II: Let Execute the update step; Supervision: Set the supervision mode of the responsible number to interrupt the restricted input state, and monitor the data input by the responsible number within the detection period T; Update: Update the maximum allowable amount of abnormal data A in the recording step and execute the extraction I step.

7. The intelligent data management method based on artificial intelligence according to claim 6, characterized in that: In the recording step, the calculation model of the warning error B is: Where C is the sensitivity constant, and c ∈ (0, 1); In the updating step, the warning error B in the recording step is also updated.

8. The intelligent data management method based on artificial intelligence according to claim 7, characterized in that: In the detection step, the working time corresponding to the responsibility number is also extracted as h; In the recording step, the value of C is also determined according to the working time h corresponding to the responsibility number, and the calculation model of the sensitivity constant C is: Among them, t1 is the set standard working time, and t2 is the maximum allowed working time.

9. The intelligent data management method based on artificial intelligence according to claim 8, characterized in that: After the updating step, an extraction II step and a risk calculation I step are also provided; Extraction II: Extract the average amount of abnormal data of all responsibility numbers with abnormalities in the last detection cycle T from the database, which is Q, and the average working time is t avg ; Risk calculation I: If N≤Q, the risk value of the responsibility number is recorded as R=0; if N>Q, the risk value of the responsibility number is recorded as 10. The intelligent data management method based on artificial intelligence according to claim 9, characterized in that: The risk calculation step I is followed by a risk calculation step II; Risk calculation II: Set the cycle threshold T1. If the responsibility number without abnormal data has no secondary sampling data extracted within the cycle threshold T1, the risk value of the recorded responsibility number is R=0.

Images