Cloud-based HIS medical insurance data processing method and system

Abstract

The invention relates to the technical field of medical insurance management, and particularly discloses a cloud-based HIS medical insurance data processing method and system, and the method comprises the following steps: S1, determining child nodes, obtaining an undetermined time length, and generating a first sequence based on the undetermined time length of an auditor corresponding to the same child node; s2, calculating a deviation value, and determining an abnormal value in the deviation value; counting an abnormal proportion, and determining first-class personnel and second-class personnel based on the abnormal proportion; s3, obtaining a first-class sequence based on the first-class personnel, and determining a theoretical target number P1 of the first-class personnel in the first-class sequence; obtaining a second-class sequence based on the second-class personnel, and determining a theoretical target number P2 of the first second-class personnel in the second-class sequence; and calculating a judgment number based on the theoretical target numbers P1 and P2, and determining a target person based on the judgment number. According to the invention, congestion in the medical insurance reimbursement process can be reduced, and the overall auditing speed and the reimbursement efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical insurance management, and in particular to a cloud-based HIS medical insurance data processing method and system. Background Art

[0002] HIS refers to an information system used by hospitals to comprehensively manage various types of information such as patient diagnosis and treatment process, medical business management, administrative management, etc., and usually includes relevant functional modules such as medical insurance management.

[0003] Medical insurance, or health insurance, refers to a system or service that provides certain compensation or reimbursement of expenses when an individual is sick or needs medical services.

[0004] After patients submit their medical insurance reimbursement applications, they need to be reviewed by reviewers. Due to the differences in proficiency of different reviewers in process operation, policy understanding and practical experience, the processing speed of a certain review node is often relatively slow, which in turn leads to a large number of unreviewed applications at this node, resulting in obvious process congestion and affecting the efficiency of normal reimbursement. Summary of the invention

[0005] The purpose of the present invention is to provide a cloud-based HIS medical insurance data processing method and system to solve the following technical problems:

[0006] Due to the differences in proficiency of different reviewers in process operation, policy understanding and practical experience, the processing speed of a certain review node is often relatively slow, which in turn leads to a large backlog of applications that have not yet been reviewed at this node, resulting in obvious process congestion and affecting the efficiency of normal reimbursement.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A cloud-based HIS medical insurance data processing method comprises the following steps:

[0009] S1: Obtain the review process of medical insurance reimbursement, record the review nodes therein as sub-nodes, obtain the pending time from the medical insurance data stored in the cloud, the pending time indicates the time it takes for the reviewer corresponding to the sub-node to review a single medical insurance reimbursement application, sort the pending time in the order of the time axis, and generate a first sort;

[0010] S2: Calculate the deviation value ΔDi = Di - Di +1 , Di represents the waiting time of the i-th position in the first sorting, when the deviation value ΔDi≤μ and Di +1 >Dys, the deviation value is recorded as an abnormal value, μ is a preset value and μ<0, and Dys represents the average waiting time of the reviewers corresponding to a single sub-node;

[0011] Statistically calculate the proportion of the above-mentioned outliers in the deviation value, and take it as the outlier ratio BZC. When the outlier ratio BZC ≥ η1, mark the corresponding reviewer as a first-class reviewer; when the outlier ratio BZC ≤ η2, mark the corresponding reviewer as a second-class reviewer; η1 and η2 are preset first and second judgment coefficients, and η1 > η2.

[0012] S3: Sort the above-mentioned first-class reviewers in descending order according to the outlier ratio to obtain a first-class ranking. Determine the first-class reviewer CF1 at the top of the first-class ranking, and obtain the target quantity of the first-class reviewer CF1 within a preset sub-period. The target quantity is the number of medical insurance reimbursement applications with the review completed. Calculate the theoretical target quantity P1 of the first-class reviewer CF1 within m sub-periods as P1 = (1 - BZC CF1 ) * CF1, where m is the number of preset sub-periods, and BZC CF1 represents the outlier ratio of the second-class reviewer CF1.

[0013] Sort the above-mentioned second-class reviewers in descending order according to the outlier ratio to obtain a second-class ranking. Determine the second-class reviewer CS2 at the top of the second-class ranking, and obtain the target quantity of the second-class reviewer CS2 within the sub-period. Calculate the theoretical target quantity P2 of the second-class reviewer CS2 within m sub-periods as P2 = (1 - BZC CS2 ) * CS2, where BZC CS2 represents the outlier ratio of the second-class reviewer CS2.

[0014] When the judgment number K = P2 - P1 ≥ Kys, mark the first-class reviewer CF1 as the target reviewer, and prompt that there is an abnormality in the review efficiency of the target reviewer. Kys represents the judgment number threshold.

[0015] As a further solution of the present invention: in the step S2, the following steps are further included:

[0016] When the first ranking is monotonically increasing and the mean value of the corresponding pending time is greater than or equal to the average pending time Dys, take the corresponding reviewer as the target reviewer;

[0017] When the first ranking is monotonically decreasing and the mean value of the corresponding pending time is less than the average pending time Dys, do not take the corresponding reviewer as the target reviewer.

[0018] As a further solution of the present invention: in the step S3, when the mean value of the target quantity is greater than or equal to the preset mean value threshold, do not take the corresponding first-class reviewer as the target reviewer.

[0019] As a further solution of the present invention: in the step S3, when two or more abnormal proportions are the same and the largest, the corresponding two types of personnel are taken as average personnel, and the mean of the theoretical target number of the average personnel in the sub-period is calculated as the theoretical target number P2.

[0020] As a further solution of the present invention: in the step S3, when the total number of the one type of personnel is greater than or equal to a preset quantity threshold, an abnormal information is sent for reporting.

[0021] As a further solution of the present invention: the step S3 further includes the following steps:

[0022] Remove the first category of personnel CF1 from the first category of ranking to obtain a new category of ranking, repeat the above steps to determine whether the first target person in the new category of ranking is the target person;

[0023] Repeat the above steps to obtain all target persons.

[0024] As a further solution of the present invention: in the step S2, when calculating the average waiting time, when the difference between a certain waiting time and the average waiting time is greater than or equal to a preset difference threshold, the waiting time is removed and the average is calculated again.

[0025] A cloud-based HIS medical insurance data processing system, comprising:

[0026] Acquisition module: obtain the review process of medical insurance reimbursement, record the review nodes therein as sub-nodes, obtain the pending time from the medical insurance data stored in the cloud, the pending time indicates the time it takes for the reviewer corresponding to the sub-node to review a single medical insurance reimbursement application, sort the pending time in the order of the time axis, and generate a first sort;

[0027] Qualitative module: Calculate the deviation value ΔDi = Di-Di +1 , Di represents the waiting time of the i-th position in the first sorting, when the deviation value ΔDi≤μ and Di +1 >Dys, the deviation value is recorded as an abnormal value, μ is a preset value and μ<0, and Dys represents the average waiting time of the reviewers corresponding to a single sub-node;

[0028] The ratio of the abnormal value to the deviation value is counted and taken as the abnormal ratio BZC. When the abnormal ratio BZC≥η1, the corresponding auditor is marked as a first-class personnel; when the abnormal ratio BZC≤η2, the corresponding auditor is marked as a second-class personnel; η1 and η2 are the preset first judgment coefficient and second judgment coefficient, and η1>η2;

[0029] Result module: Sort the said first type of personnel in descending order according to the abnormal ratio to obtain a first type of sorting, determine the first type of personnel CF1 at the top of the said first type of sorting, obtain the target quantity of the said first type of personnel CF1 within a preset sub-cycle, where the said target quantity is the number of medical insurance reimbursement applications with the audit completed, and calculate the theoretical target quantity P1 of the said first type of personnel CF1 within m sub-cycles as P1=(1 - BZC CF1 )*CF1, where m is the number of preset sub-cycles, and BZC CF1 represents the abnormal ratio of the said second type of personnel CF1;

[0030] Sort the said second type of personnel in descending order according to the abnormal ratio to obtain a second type of sorting, determine the first second type of personnel CS2 at the top of the said second type of sorting, obtain the target quantity of the said second type of personnel CS2 within the said sub-cycle, and calculate the theoretical target quantity P2 of the said second type of personnel CS2 within m sub-cycles as P2=(1 - BZC CS2 )*CS2, where BZC CS2 represents the abnormal ratio of the said second type of personnel CS2;

[0031] When the judgment number K = P2 - P1 ≥ Kys, mark the said first type of personnel CF1 as the target personnel, and prompt that there is an abnormality in the audit efficiency of the said target personnel. Kys represents the judgment number threshold.

[0032] The beneficial effects of the present invention are as follows: first, a first ranking is generated according to the pending time of a single reviewer. By arranging the pending time in chronological order, the distribution of the time taken by each reviewer to process different applications can be clearly seen, thereby revealing their efficiency level on each application and providing data support for subsequent outlier analysis; then, the outliers are determined and the abnormal proportions are counted, and the first and second category personnel are distinguished according to the abnormal proportions; it is worth noting that the basis for determining the outliers is: the processing time of the next application (i+1) is longer and the processing time of the next application is greater than the average processing time; by counting the outliers and calculating the abnormal proportions (BZC), the overall performance of the reviewer over a period of time can be better understood. If the abnormal ratio of a certain auditor exceeds the set threshold, it can be classified as a Class I personnel, indicating that there may be abnormal problems in its audit efficiency, and if its abnormal ratio is low, it can be classified as a Class II personnel, and its audit efficiency is considered normal; after distinguishing between Class I and Class II personnel, the corresponding theoretical target number is calculated, and it is used to determine whether it is a target personnel; it can be understood that Class I personnel are personnel whose audit efficiency may be abnormal, and Class II personnel are personnel whose audit efficiency is not abnormal, which is used as a reference to determine whether Class I personnel have abnormalities; after the distinction between Class I and Class II personnel is completed, further calculation of the theoretical target numbers P1 and P2 can help determine whether the auditor has achieved the expected audit target and whether there is a significant efficiency difference. If the calculated judgment number K (i.e., P2-P1) is greater than the set threshold Kys, it can be clearly marked as a Class I personnel as a target personnel, and it is prompted that there may be abnormal efficiency, and the manager will take corresponding measures, including but not limited to training to improve proficiency, etc.; the present invention can reduce congestion in the medical insurance reimbursement process and improve the overall audit speed and reimbursement efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be further described below in conjunction with the accompanying drawings.

[0034] Figure 1 It is a flow chart of a cloud-based HIS medical insurance data processing method of the present invention. DETAILED DESCRIPTION

[0035] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.

[0036] See also Figure 1 As shown, the present invention is a cloud-based HIS medical insurance data processing method, comprising the following steps:

[0037] S1: Obtain the review process of medical insurance reimbursement, record the review nodes therein as sub-nodes, obtain the pending time from the medical insurance data stored in the cloud, the pending time indicates the time it takes for the reviewer corresponding to the sub-node to review a single medical insurance reimbursement application, sort the pending time in the order of the time axis, and generate a first sort;

[0038] S2: Calculate the deviation value ΔDi = Di - Di +1 , Di represents the waiting time of the i-th position in the first sorting, when the deviation value ΔDi≤μ and Di +1 >Dys, the deviation value is recorded as an abnormal value, μ is a preset value and μ<0, and Dys represents the average waiting time of the reviewers corresponding to a single sub-node;

[0039] The ratio of the abnormal value to the deviation value is counted and taken as the abnormal ratio BZC. When the abnormal ratio BZC≥η1, the corresponding auditor is marked as a first-class personnel; when the abnormal ratio BZC≤η2, the corresponding auditor is marked as a second-class personnel; η1 and η2 are the preset first judgment coefficient and second judgment coefficient, and η1>η2;

[0040] S3: Sort the first category of personnel in descending order according to the size of the abnormal proportion to obtain a first category ranking, determine the first category of personnel CF1 in the first category ranking, obtain the target number of the first category of personnel CF1 in a preset sub-period, the target number is the number of medical insurance reimbursement applications that have been reviewed, and calculate the theoretical target number P1 of the first category of personnel CF1 in m sub-periods = (1-BZC CF1 )*CF1, m is the preset number of sub-periods, BZC CF1 Indicates the abnormal proportion of CF1 of the second category personnel;

[0041] The second category personnel are sorted in descending order according to the size of the abnormal proportion to obtain a second category ranking, the first second category personnel CS2 in the second category ranking is determined, the target number of the second category personnel CS2 in the sub-period is obtained, and the theoretical target number P2 of the second category personnel CS2 in m sub-periods is calculated. CS2 )*CS2,BZC CS2 Indicates the abnormal proportion of the second category of personnel CS2;

[0042] When the judgment number K=P2-P1≥Kys, the type of personnel CF1 is marked as the target personnel, indicating that the audit efficiency of the target personnel is abnormal. Kys represents the judgment number threshold.

[0043] It should be noted that the first ranking is generated based on the pending time of a single reviewer. By arranging the pending time in chronological order, we can clearly see the distribution of the time each reviewer takes to process different applications, thereby revealing their efficiency level on each application and providing data support for the subsequent outlier analysis. After that, the outliers are determined and the abnormal proportions are counted, and the first and second category personnel are distinguished based on the abnormal proportions. It is worth noting that the basis for determining the outliers is that the processing time of the next application (i+1) is longer and the processing time of the next application is greater than the average processing time. By counting the outliers and calculating the abnormal proportions (BZC), we can better understand the overall performance of the reviewers over a period of time. If the abnormal ratio of a certain auditor exceeds the set threshold, it can be classified as a Class I personnel, indicating that there may be abnormal problems with its audit efficiency, while if its abnormal ratio is low, it can be classified as a Class II personnel, and its audit efficiency is considered normal; after distinguishing between Class I and Class II personnel, the corresponding theoretical target number is calculated, and it is used to determine whether it is a target personnel; it can be understood that Class I personnel are personnel whose audit efficiency may be abnormal, and Class II personnel are personnel whose audit efficiency is not abnormal, which is used as a reference to determine whether Class I personnel have abnormalities; after the distinction between Class I and Class II personnel is completed, further calculation of the theoretical target numbers P1 and P2 can help determine whether the auditor has achieved the expected audit target and whether there is a significant efficiency difference. If the calculated judgment number K (i.e. P2-P1) is greater than the set threshold Kys, it can be clearly marked as a Class I personnel as a target personnel, and it is prompted that there may be abnormal efficiency, and the manager will take corresponding measures, including but not limited to training to improve proficiency.

[0044] In another preferred embodiment of the present invention, the step S2 further includes the following steps:

[0045] When the first ranking is monotonically increasing and the corresponding mean of the pending time is greater than or equal to the average pending time Dys, the corresponding reviewer is taken as the target person;

[0046] When the first ranking is monotonically decreasing and the corresponding mean of the pending time is less than the average pending time Dys, the corresponding reviewer will not be regarded as the target person.

[0047] It is worth noting that the monotonically increasing first ranking means that the waiting time of the reviewer in processing the application is gradually increasing, indicating that the work efficiency of the person may gradually decrease during the review process. At this time, if the mean waiting time of the person is greater than or equal to the average waiting time Dys, it indicates that the overall review efficiency of the person is already low, so he can be regarded as a target person for further attention;

[0048] When the first ranking is monotonically decreasing, it means that the waiting time of the person is gradually decreasing, indicating that the person may gradually improve his efficiency or processing speed during the review process. If the average waiting time of the person is less than the average waiting time Dys, it means that the person is able to process the application efficiently as a whole, and therefore, he is not regarded as a target person.

[0049] In another preferred embodiment of the present invention, in step S3, when the mean of the target quantity is greater than or equal to a preset mean threshold, the corresponding category of personnel is not regarded as target personnel.

[0050] In another preferred embodiment of the present invention, in the step S3, when two or more abnormal proportions are the same and the largest, the corresponding two types of personnel are taken as average personnel, and the mean of the theoretical target number of the average personnel in the sub-period is calculated as the theoretical target number P2.

[0051] In another preferred embodiment of the present invention, in the step S3, when the total number of the one type of personnel is greater than or equal to a preset quantity threshold, an abnormal information is sent for reporting.

[0052] In another preferred embodiment of the present invention, the step S3 further includes the following steps:

[0053] Remove the first category of personnel CF1 from the first category of ranking to obtain a new category of ranking, repeat the above steps to determine whether the first target person in the new category of ranking is the target person;

[0054] Repeat the above steps to obtain all target persons.

[0055] In another preferred embodiment of the present invention, in the step S2, when calculating the average waiting time, when the difference between a certain waiting time and the average waiting time is greater than or equal to a preset difference threshold, the waiting time is removed and the average is calculated again.

[0056] A cloud-based HIS medical insurance data processing system, comprising:

[0057] Acquisition module: obtain the review process of medical insurance reimbursement, record the review nodes therein as sub-nodes, obtain the pending time from the medical insurance data stored in the cloud, the pending time indicates the time it takes for the reviewer corresponding to the sub-node to review a single medical insurance reimbursement application, sort the pending time in the order of the time axis, and generate a first sort;

[0058] Qualitative module: Calculate the deviation value ΔDi = Di-Di +1 , Di represents the waiting time of the i-th position in the first sorting, when the deviation value ΔDi≤μ and Di +1>Dys, the deviation value is recorded as an abnormal value, μ is a preset value and μ<0, and Dys represents the average waiting time of the reviewers corresponding to a single sub-node;

[0059] The ratio of the abnormal value to the deviation value is counted and taken as the abnormal ratio BZC. When the abnormal ratio BZC≥η1, the corresponding auditor is marked as a first-class personnel; when the abnormal ratio BZC≤η2, the corresponding auditor is marked as a second-class personnel; η1 and η2 are the preset first judgment coefficient and second judgment coefficient, and η1>η2;

[0060] Result module: Sort the first category of personnel in descending order according to the size of the abnormal proportion to obtain a first category ranking, determine the first category of personnel CF1 in the first category ranking, obtain the target number of the first category of personnel CF1 in the preset sub-period, the target number is the number of medical insurance reimbursement applications that have been reviewed, and calculate the theoretical target number P1 of the first category of personnel CF1 in m sub-periods = (1-BZC CF1 )*CF1, m is the preset number of sub-periods, BZC CF1 Indicates the abnormal proportion of CF1 of the second category personnel;

[0061] The second category personnel are sorted in descending order according to the size of the abnormal proportion to obtain a second category ranking, the first second category personnel CS2 in the second category ranking is determined, the target number of the second category personnel CS2 in the sub-period is obtained, and the theoretical target number P2 of the second category personnel CS2 in m sub-periods is calculated. CS2 )*CS2,BZC CS2 Indicates the abnormal proportion of the second category of personnel CS2;

[0062] When the judgment number K=P2-P1≥Kys, the type of personnel CF1 is marked as the target personnel, indicating that the audit efficiency of the target personnel is abnormal. Kys represents the judgment number threshold.

[0063] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A cloud-based HIS medical insurance data processing method, characterized in that: The following steps are involved: S1: Obtain the review process of medical insurance reimbursement, record the review nodes therein as sub-nodes, obtain the pending time from the medical insurance data stored in the cloud, the pending time indicates the time it takes for the reviewer corresponding to the sub-node to review a single medical insurance reimbursement application, sort the pending time in the order of the time axis, and generate a first sort; S2: Calculate the deviation value ΔDi = Di - Di +1 , Di represents the waiting time of the i-th position in the first sorting, when the deviation value ΔDi≤μ and Di +1 >Dys, the deviation value is recorded as an abnormal value, μ is a preset value and μ<0, and Dys represents the average waiting time of the reviewers corresponding to a single sub-node; The ratio of the abnormal value to the deviation value is counted and taken as the abnormal ratio BZC. When the abnormal ratio BZC≥η1, the corresponding auditor is marked as a first-class personnel; when the abnormal ratio BZC≤η2, the corresponding auditor is marked as a second-class personnel; η1 and η2 are the preset first judgment coefficient and second judgment coefficient, and η1>η2; S3: Sort the first category of personnel in descending order according to the size of the abnormal proportion to obtain a first category ranking, determine the first category of personnel CF1 in the first category ranking, obtain the target number of the first category of personnel CF1 in a preset sub-period, the target number is the number of medical insurance reimbursement applications that have been reviewed, and calculate the theoretical target number P1 of the first category of personnel CF1 in m sub-periods = (1-BZC CF1 )*CF1, m is the preset number of sub-periods, BZC CF1 Indicates the abnormal proportion of CF1 of the second category personnel; The second category personnel are sorted in descending order according to the size of the abnormal proportion to obtain a second category ranking, the first second category personnel CS2 in the second category ranking is determined, the target number of the second category personnel CS2 in the sub-period is obtained, and the theoretical target number P2 of the second category personnel CS2 in m sub-periods is calculated. CS2 )*CS2,BZC CS2 Indicates the abnormal proportion of the second category of personnel CS2; When the judgment number K=P2-P1≥Kys, the type of personnel CF1 is marked as the target personnel, indicating that the audit efficiency of the target personnel is abnormal. Kys represents the judgment number threshold.

2. A cloud-based HIS medical insurance data processing method according to claim 1, characterized in that: The step S2 further includes the following steps: When the first ranking is monotonically increasing and the corresponding mean of the pending time is greater than or equal to the average pending time Dys, the corresponding reviewer is taken as the target person; When the first ranking is monotonically decreasing and the corresponding mean of the pending time is less than the average pending time Dys, the corresponding reviewer will not be regarded as the target person.

3. The cloud-based HIS medical insurance data processing method according to claim 1, characterized in that: In the step S3, when the mean of the target quantity is greater than or equal to the preset mean threshold, the corresponding category of personnel is not regarded as the target personnel.

4. The cloud-based HIS medical insurance data processing method according to claim 1, characterized in that: In the step S3, when two or more abnormal proportions are the same and the largest, the corresponding two types of personnel are taken as average personnel, and the mean of the theoretical target number of the average personnel in the sub-period is calculated as the theoretical target number P2.

5. The cloud-based HIS medical insurance data processing method according to claim 1, characterized in that: In the step S3, when the total number of the one type of personnel is greater than or equal to a preset number threshold, an abnormal information is sent for reporting.

6. A cloud-based HIS medical insurance data processing method according to claim 1, characterized in that: The step S3 further includes the following steps: Remove the first category of personnel CF1 from the first category of ranking to obtain a new category of ranking, repeat the above steps to determine whether the first target person in the new category of ranking is the target person; Repeat the above steps to obtain all target persons.

7. The cloud-based HIS medical insurance data processing method according to claim 1, characterized in that: In the step S2, when calculating the average waiting time duration, when the difference between a certain waiting time duration and the average waiting time duration is greater than or equal to a preset difference threshold, the waiting time duration is removed and the average is calculated again.

8. A cloud-based HIS medical insurance data processing system, characterized in that: include: Acquisition module: obtain the review process of medical insurance reimbursement, record the review nodes therein as sub-nodes, obtain the pending time from the medical insurance data stored in the cloud, the pending time indicates the time it takes for the reviewer corresponding to the sub-node to review a single medical insurance reimbursement application, sort the pending time in the order of the time axis, and generate a first sort; Qualitative module: Calculate the deviation value ΔDi = Di-Di +1 , Di represents the waiting time of the i-th position in the first sorting, when the deviation value ΔDi≤μ and Di +1 >Dys, the deviation value is recorded as an abnormal value, μ is a preset value and μ<0, and Dys represents the average waiting time of the reviewers corresponding to a single sub-node; The ratio of the abnormal value to the deviation value is counted and taken as the abnormal ratio BZC. When the abnormal ratio BZC≥η1, the corresponding auditor is marked as a first-class personnel; when the abnormal ratio BZC≤η2, the corresponding auditor is marked as a second-class personnel; η1 and η2 are the preset first judgment coefficient and second judgment coefficient, and η1>η2; Result module: Sort the first category of personnel in descending order according to the size of the abnormal proportion to obtain a first category ranking, determine the first category of personnel CF1 in the first category ranking, obtain the target number of the first category of personnel CF1 in the preset sub-period, the target number is the number of medical insurance reimbursement applications that have been reviewed, and calculate the theoretical target number P1 of the first category of personnel CF1 in m sub-periods = (1-BZC CF1 )*CF1, m is the preset number of sub-periods, BZC CF1 Indicates the abnormal proportion of CF1 of the second category personnel; The second category personnel are sorted in descending order according to the size of the abnormal proportion to obtain a second category ranking, the first second category personnel CS2 in the second category ranking is determined, the target number of the second category personnel CS2 in the sub-period is obtained, and the theoretical target number P2 of the second category personnel CS2 in m sub-periods is calculated. CS2 )*CS2,BZC CS2 Indicates the abnormal proportion of the second category of personnel CS2; When the judgment number K=P2-P1≥Kys, the type of personnel CF1 is marked as the target personnel, indicating that the audit efficiency of the target personnel is abnormal. Kys represents the judgment number threshold.

Images