Donated person accurate search and information verification system based on data analysis

Abstract

The invention discloses a donated person accurate search and information verification system based on data analysis, and belongs to the technical field of medical donation, and the system comprises a data integration platform which is used for obtaining disease associated data information and personnel associated data information related to donation from big data; the expected portrait generation module is used for generating a donated person portrait set according to the acquired disease associated data information; the calibration verification module is used for further analyzing and generating a donation coefficient according to the acquired associated data information; the determining module is used for determining the final donated personnel according to the sorting result of the donation coefficients; and the safety early warning module is used for performing safety early warning protection on data in the system. Through accurate measurement and calculation, according to basic information, disease information, doctor seeing and other multi-dimensional data of a helped person, people can be found through resources, the crowd needing to be rescued most can be actively recognized, manpower resource waste is reduced, and meanwhile rescue pertinence and timeliness are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical donation, and in particular relates to a system for accurately searching and verifying donated persons based on data analysis. Background Art

[0002] Charitable medical assistance refers to a system whereby society, through various charitable acts, provides assistance and financial support to impoverished individuals who are unable to afford treatment due to illness or accidental injury. Charitable medical assistance can help address the medical expenses of patients in need. As charitable resources continue to flow into the medical sector, it can continue to address the gap in medical insurance supply and help alleviate the burden of medical expenses for those in need.

[0003] Although the current selection of medical assistance recipients also tends to favor low-income groups and people in difficulty, the identification capabilities and information acquisition channels are insufficient. Most of them rely on manual teams to go to major hospitals to "dig out" patients in need. This not only consumes huge labor costs, but is also prone to errors and even fraud, resulting in errors in medical resource assistance. Summary of the Invention

[0004] The purpose of the present invention is to provide a system for accurately searching and verifying donated recipients based on data analysis, so as to solve the problems faced by the above-mentioned background technology.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A system for accurately searching and verifying donated recipients based on data analysis, comprising:

[0007] A data integration platform, wherein the data integration platform is used to obtain related data information related to donations from big data, wherein the related data information includes disease-related data information and personnel-related data information;

[0008] An expected portrait generation module, which is used to generate a portrait set of donated persons based on the acquired disease-related data information;

[0009] A calibration and verification module, which is used to further analyze the acquired disease-related data and person-related data to generate a donation coefficient and sort the persons in the portrait set of donated persons according to the size of the donation coefficient;

[0010] A determination module, which determines the final recipients of donations based on the ranking results;

[0011] A security warning module is used to provide security warning protection for data in the system.

[0012] Furthermore, the disease-related data information includes disease type data, regional poverty data, medical data, education data, donation data of social organizations, volunteer data, etc., and the personnel-related data information includes medical treatment data, medical expenses data, disease type data, family asset data, and family status data, etc.

[0013] Furthermore, the method for the expected portrait generation module to generate a portrait set of donated persons is:

[0014] According to the disease-related data information, multiple key features in the relevant data are extracted to form a key feature set;

[0015] Select the recipients of donations with relevant key features in the key feature set from the big data of the corresponding region to form an initial portrait set of recipients of donations;

[0016] According to the key characteristics of each recipient, through the formula Obtain the screening value CH of each recipient and compare the screening value CH with the preset screening threshold CH th Compare and set CH>CH th Mark the people who received donations, thus forming a portrait collection of the recipients;

[0017] Among them, n is the number of key features carried by the recipients, α i is the influence coefficient of the i-th key feature, and i∈[1,n].

[0018] Furthermore, the method for generating the donation coefficient by the calibration verification module is:

[0019] Obtain the personnel-related data information of each recipient in the recipient portrait set, and thus obtain the recipient's disease impact value R a And the capital impact value R b ;

[0020] Then, through the formula R=R a *w1+R b *w2 is used to calculate the donation coefficient R of the recipient;

[0021] Wherein, w1 and w2 are respective preset proportional coefficients.

[0022] Furthermore, the disease impact value R a The method to obtain it is:

[0023] Obtain the cumulative total number of visits m for each recipient of donations, as well as the medical expenses MY and hospitalization duration DY consumed during each visit, from the historical database of the corresponding region. This allows the development of a curve showing the change in medical expenses versus the number of visits x, MY(x), and a curve showing the change in hospitalization duration versus the number of visits x, DY(x).

[0024] By formula Obtain the disease impact value R a ;

[0025] Among them, DR is the severity of the disease, A is the current recovery level of the disease, MY0(x) is the curve of the standard medical expenses set by the system changing with the number of visits, and DY0(x) is the curve of the standard hospitalization time set by the system changing with the number of visits.

[0026] Furthermore, the capital impact value R b The acquisition method is:

[0027] Obtain the medical expenses MY consumed by the recipient each time he / she visits the hospital and the family asset value AS of the recipient;

[0028] By formula Get the capital impact value R b ;

[0029] Among them, MY j is the medical expenses consumed during the jth visit, and j∈[1,m], δ is the average personal expenditure in the area where the recipient is located, θ is the fixed reimbursement amount of medical insurance in the area where the recipient is located, ξ is the patient age coefficient, ω is the disease stage coefficient, γ is the medical insurance status coefficient of the recipient, and r a is the number of minors in the recipient's family, r b The number of elderly people in the families of the recipients of donations.

[0030] Furthermore, the method by which the determination module determines the final recipient of donations based on the ranking results is:

[0031] Obtain the donation coefficient R of the GG recipients in the portrait set of recipients, sort each person in descending order according to the donation coefficient R, select the top L recipients according to the predetermined number of recipients L and the donation amount, and give the top L recipients a corresponding proportion of donation funds according to the proportion of the donation funds.

[0032] Furthermore, the method by which the security warning module performs security warning protection on the data in the system is:

[0033] Get the information of the person who logged into the system and determine whether he is an authorized login person. If he is not authorized, an alarm will be issued. Otherwise, proceed to the next step.

[0034] Get a period ΔT n Curve Q of the number of views of logged-in users changing over time r (t) and the download volume change curve D r (t);

[0035] Thus, through the formula Get the risk value K r ;

[0036] when When , an alarm is issued;

[0037] in, is the preset risk threshold, maxQ r '(t) is Q r Maximum slope of (t), minQ r '(t) is Q r The minimum slope of (t), maxD r '(t) is D r The maximum slope of (t), minD r '(t) is D r Minimum slope of (t), ΔQ r and ΔD r are the preset slope comparison values, and t1 is the period ΔT n The start time of t2 is the period ΔT n The end time, This is the preset standard pageviews curve over time. This is a curve showing the change in download volume over time for a preset standard.

[0038] Beneficial effects of the present invention:

[0039] The present invention uses precise calculations to match resources to people based on multi-dimensional data such as the basic information, disease information, and medical treatment of the beneficiaries. It can proactively identify the groups most in need of assistance, reduce waste of human resources, and improve the targeted and timely nature of assistance. When searching for beneficiaries, the present invention can also combine data such as local payment standards, reimbursement ratios, medical insurance data, and the patient's personal circumstances to comprehensively assess the needs and feasibility of the beneficiaries, thereby optimizing resource allocation and avoiding waste of funds.

[0040] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. 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 paying any creative work.

[0042] Figure 1 This is a system block diagram of the present invention. DETAILED DESCRIPTION

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making any creative efforts shall fall within the scope of protection of the present invention.

[0044] In one embodiment, a system for accurately searching and verifying donated persons based on data analysis is disclosed. Figure 1 As shown, the system mainly includes the following modules:

[0045] Data integration platform, which is used to obtain related data information related to donations from big data. Related data information includes disease-related data information and personnel-related data information. Disease-related data information includes disease type data, regional poverty data, medical data, education data, donation data from social organizations, volunteer data, etc. Personnel-related data information includes medical treatment data, medical expense data, disease type data, family asset data, and family status data.

[0046] An expected portrait generation module is used to generate a portrait set of donated persons based on the acquired disease-related data information;

[0047] The calibration and verification module is used to further analyze the acquired disease-related data and personnel-related data to generate a donation coefficient and sort the persons in the portrait set of donated persons according to the size of the donation coefficient;

[0048] A determination module determines the final recipients of donations based on the ranking results;

[0049] Security warning module, the security warning module is used to provide security warning protection for data in the system.

[0050] Through the above technical solution, the present application obtains related data information related to donations from big data through the data integration platform, including disease-related data information and personnel-related data information. It directly finds people who need donations based on the relevant data in the big data, which can save a lot of manpower and material resources. Then, according to the expected portrait generation module, a portrait set of donated people is generated. In this way, the possible population in need of help can be directly screened from the big data, which can narrow the scope of donations and improve accuracy. Then, according to the calibration verification module, further analysis is performed to generate donation coefficients. According to the size of the donation coefficient, the people in the portrait set of donated people are sorted, and the donated people are determined according to the sorting order. In this way, the donated people can be accurately found and improved in accuracy. Finally, the security warning module is used to protect the data in the system through security warning, ensuring that the personal information of the donated people is not leaked or abused, and improving the security and integrity of the data. In this way, the present application realizes resource retrieval based on the basic information, disease information, medical treatment and other multi-dimensional data of the recipients through accurate measurement, which can actively identify the people who need help the most, reduce the waste of human resources and improve the targeted and timely nature of the assistance. When looking for recipients, we can combine local payment standards, reimbursement ratios, medical insurance data, and patient personal information to comprehensively assess the recipients' needs and feasibility, optimize resource allocation, and avoid wasting funds.

[0051] As an embodiment of the present invention, the method for generating a portrait set of donated persons by the expected portrait generation module is as follows:

[0052] According to the disease-related data information, multiple key features in the relevant data are extracted to form a key feature set;

[0053] Select the recipients of donations with relevant key features in the key feature set from the big data of the corresponding region to form an initial portrait set of recipients of donations;

[0054] According to the key characteristics of each recipient, through the formula Obtain the screening value CH of each recipient and compare the screening value CH with the preset screening threshold CH th Compare and set CH>CH th Mark the people who received donations, thus forming a portrait collection of the recipients;

[0055] Among them, n is the number of key features carried by the recipients, α i is the influence coefficient of the i-th key feature, and i∈[1,n].

[0056] Through the above technical solution, this embodiment provides a specific method for generating a portrait set of donated persons. First, based on the disease-related data information, multiple key features in the relevant data are extracted to form a key feature set. Then, the donated persons with relevant key features in the key feature set are screened from the big data of the corresponding region to form an initial portrait set of donated persons. This can narrow the scope of the screened persons for subsequent searches. For example, a key feature set is formed from a regional data set with key features such as major diseases, age, severity of the disease, financial consumption, and urgency of treatment. The donated persons with relevant key features in the key feature set are screened from the big data of the corresponding region to form an initial portrait set of donated persons. This can narrow the scope of the screened persons. Then, based on the key features carried by each donated person, the formula Obtain the screening value CH of each recipient, where n is the number of key features carried by the recipient, α i is the impact ratio coefficient of the i-th key feature. The impact ratio coefficient of each key feature can be determined based on relevant historical data and experience. For example, for the severity of the disease, the ratio can be determined according to the early, middle and late stages of the disease. The later the stage, the higher the ratio. Similarly, for capital consumption, the greater the proportion of the amount consumed, the higher the corresponding ratio. The screening value CH is compared with the preset screening threshold CH. th Compare and set CH>CH th The people are marked to form a portrait set of donated people. According to the size of the screening value of each donated person, it can be seen that the larger the screening value, the greater the possibility of needing donation and the more it meets the donation target. Therefore, the screening value CH is compared with the preset screening threshold CH th Compare and set CH>CH th The people who are donated can be marked to form a portrait set of the donated people. This can further narrow the scope of the people to be screened, save time in the subsequent search for the donated people, and provide convenience for the subsequent search.

[0057] As an embodiment of the present invention, the method for generating the contribution coefficient by the calibration verification module is:

[0058] Obtain the personnel-related data information of each recipient in the recipient portrait set, and thus obtain the recipient's disease impact value R a And the capital impact value R b ;

[0059] Then, through the formula R=R a *w1+R b *w2 is used to calculate the donation coefficient R of the recipient;

[0060] Among them, w1 and w2 are the preset proportional coefficients respectively, and the disease impact value R a The acquisition method is as follows: obtain the cumulative total number of visits m of each recipient of the donation, as well as the medical expenses MY and hospitalization duration DY consumed during each visit from the historical database of the corresponding region, and then formulate the curve of medical expenses changing with the number of visits x MY(x) and the curve of hospitalization duration changing with the number of visits x DY(x);

[0061] By formula Obtain the disease impact value R a ;

[0062] Among them, DR is the severity of the disease, A is the current recovery level of the disease, MY0(x) is the curve of the standard medical expenses set by the system changing with the number of visits, and DY0(x) is the curve of the standard hospitalization time set by the system changing with the number of visits;

[0063] Funding impact value R b The acquisition method is: obtain the medical expenses MY consumed by the recipient each time he / she visits the hospital and the family asset value AS of the recipient;

[0064] By formula Get the capital impact value R b ;

[0065] Among them, MY j is the medical expenses consumed during the jth visit, and j∈[1,m], δ is the average personal expenditure in the area where the recipient is located, θ is the fixed reimbursement amount of medical insurance in the area where the recipient is located, ξ is the patient age coefficient, ω is the disease stage coefficient, γ is the medical insurance status coefficient of the recipient, and r a is the number of minors in the recipient's family, r b The number of elderly people in the families of the recipients;

[0066] The method for the determination module to determine the final recipients of donations based on the sorting results is as follows: obtain the donation coefficient R of the GG recipients in the portrait set of recipients, sort each person in descending order according to the donation coefficient R, and select the top L recipients according to the predetermined number of recipients L and the donation amount as the recipients, and give the top L recipients a corresponding proportion of donation funds according to the proportion of the donation funds.

[0067] Through the above technical solution, the present invention provides a specific method for generating a donation coefficient and a specific method for determining a recipient based on the donation coefficient. First, the person-related data information of each recipient in the recipient portrait set is obtained, thereby obtaining the recipient's disease impact value R a And the capital impact value R b, where the disease impact value R a The acquisition method is: obtain the cumulative total number of visits m of each recipient from the historical database of the corresponding area, as well as the medical expenses MY and hospitalization time DY consumed during each visit, so as to formulate the curve of medical expenses changing with the number of visits x MY(x) and the curve of hospitalization time changing with the number of visits x DY(x), so as to use the formula Obtain the disease impact value R a Where It is expressed as the difference between the patient's medical expenses as the number of visits changes and the standard change, and the difference between the patient's hospitalization time as the number of visits changes and the standard change. Obviously, the larger the difference, the more serious the condition, and the greater the possibility of needing assistance. It indicates the severity of the disease and the degree of recovery of the patient. Obviously, the greater the severity of the disease, the worse the recovery, and the greater the possibility of needing help. The severity and recovery of the disease can be formulated according to the specific situation of the disease. For example, five (one to five) levels of severity can be formulated in advance according to the various types of diseases and the severity of the disease. The type of disease can be obtained to determine its severity (level), and the degree of recovery of the patient can be obtained through the medical records of the corresponding patients. For example, the postoperative recovery situation can be obtained based on the medical records of leukemia patients to determine their recovery level. Therefore, when the disease impact value R a The larger the value, the greater the possibility that the patient needs to be rescued; then obtain the medical expenses MY consumed by the recipient each time he visits the hospital and the family asset value AS of the recipient, through the formula Get the capital impact value R b ; From the formula It can be seen that when the total medical expenses consumed by the patient are higher and the family assets are

[0068] The smaller the case, or the more expenses that need to be pointed out within the family, the greater the possibility that the patient will need assistance when he or she becomes ill; and in the formula, δ is the average personal expenditure in the area where the recipient is located, which reflects the proportion of expenses that the individual still needs to bear under the major disease assistance policy, θ is the fixed reimbursement amount of medical insurance in the area where the recipient is located, which reflects the degree of reimbursement of medical insurance for the disease, ξ is the patient age coefficient, which reflects the treatment needs and cost expenditure of recipients of different age groups, ω is the disease stage coefficient, which is an adjustment coefficient determined according to the disease stage of the recipient, which takes into account that the treatment costs and resource requirements of different disease stages may be different, and γ is the medical insurance status coefficient of the recipient, which reflects the medical insurance treatment level of the recipient, that is, the type of medical insurance policy involved and its reimbursement scope. These values ​​can be formulated based on local policies and historical experience data. Therefore, a comprehensive analysis is conducted based on these factors to derive the funding impact value R. b , we can see that the larger the value, the greater the possibility that the patient needs to be rescued; therefore, the formula R=R a *w1+R b *w2 conducts a comprehensive analysis to determine the donation coefficient R of the recipients. Then, the individuals are ranked from largest to smallest based on the donation coefficient R. It is known that the higher the ranking, the greater the likelihood of needing assistance. Based on the predetermined number of recipients L and the amount of donations, the top L ranked individuals are selected as recipients, and the donation funds are allocated to the top L recipients in proportion to the amount of donations. This method allows for a comprehensive analysis of the impact of each patient's condition and the impact of the funds, resulting in a donation coefficient. This allows for precise identification of individuals in need of assistance based on the size of the donation coefficient, improving the efficiency and accuracy of charitable assistance.

[0069] It should be noted that the preset proportional coefficients w1 and w2, the system-set standard medical expenses curve MY0(x) that changes with the number of visits, and the system-set standard hospitalization time curve DY0(x) that changes with the number of visits can be formulated based on the sample data of the corresponding diseases in the big data, while the severity of the disease and the current degree of recovery of the disease are determined based on the actual situation of the patient's current condition. I will not elaborate on this again.

[0070] As an embodiment of the present invention, the method for the security warning module to perform security warning protection on the data in the system is:

[0071] Get the information of the person who logged into the system and determine whether he is an authorized login person. If he is not authorized, an alarm will be issued. Otherwise, proceed to the next step.

[0072] Get a period ΔT n Curve Q of the number of views of logged-in users changing over time r(t) and the download volume change curve D r (t);

[0073] Thus, through the formula Get the risk value K r ;

[0074] when When , an alarm is issued;

[0075] in, is the preset risk threshold, maxQ r '(t) is Q r Maximum slope of (t), minQ r '(t) is Q r The minimum slope of (t), maxD r '(t) is D r The maximum slope of (t), minD r '(t) is D r Minimum slope of (t), ΔQ r and ΔD r are the preset slope comparison values, and t1 is the period ΔT n The start time of t2 is the period ΔT n The end time, This is the preset standard pageviews curve over time. This is a curve showing the change in download volume over time for a preset standard.

[0076] Through the above technical solution, this embodiment provides a specific method for the security warning module to perform security warning protection on the data in the system. First, the information of the person logging into the system is obtained to determine whether the person is an authorized login person. If the person is judged to be an unauthorized login person, an alarm is issued. Otherwise, the next step is to set a monitoring period when the login person logs in and obtain the period ΔT. n Curve Q of the number of views of logged-in users changing over time r (t) and the download volume change curve D r (t), so through the formula

[0077] Get the risk value K r It can be seen that when the slope of the change in the number of views or the slope of the change in the number of downloads in a certain period of time suddenly changes greatly, or the number of downloads and views suddenly increases compared to normal times, it means that the possibility of being hacked or the account being abnormal is greater. Therefore, after obtaining the risk value, it is compared with the preset risk threshold. Compare, when When the number of logged-in accounts is abnormal, it indicates that there is a high possibility that the login account is abnormal, and an alarm will be issued to remind the administrator to confirm in order to ensure the security of data in the system.

[0078] It should be noted that the respective preset slope comparison values ​​ΔQ r and ΔD r , preset risk threshold Preset standard pageviews curve over time Preset standard download volume curve over time Both can be determined based on the historical operating data of the system combined with empirical data, and the period ΔT n The duration can be determined by humans according to actual conditions, so I will not elaborate on it here.

[0079] The above content is merely an example and explanation of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. A system for accurately searching and verifying donated recipients based on data analysis, characterized in that: The system comprises: A data integration platform, wherein the data integration platform is used to obtain related data information related to donations from big data, wherein the related data information includes disease-related data information and personnel-related data information; An expected portrait generation module, which is used to generate a portrait set of donated persons based on the acquired disease-related data information; A calibration and verification module, which is used to further analyze the acquired disease-related data and person-related data to generate a donation coefficient and sort the persons in the portrait set of donated persons according to the size of the donation coefficient; A determination module, which determines the final recipients of donations based on the ranking results; A security warning module is used to provide security warning protection for data in the system.

2. The system for accurately searching and verifying donated recipients based on data analysis according to claim 1 is characterized in that: The disease-related data information includes disease type data, regional poverty data, medical data, education data, donation data from social organizations, volunteer data, etc. The personnel-related data information includes medical treatment data, medical expense data, disease type data, family asset data and family status data.

3. The system for accurately searching and verifying donated recipients based on data analysis according to claim 2 is characterized in that: The method for generating a portrait set of donated persons by the expected portrait generation module is as follows: According to the disease-related data information, multiple key features in the relevant data are extracted to form a key feature set; Select the recipients of donations with relevant key features in the key feature set from the big data of the corresponding region to form an initial portrait set of recipients of donations; According to the key characteristics of each recipient, through the formula Obtain the screening value CH of each recipient and compare the screening value CH with the preset screening threshold CH th Compare and set CH>CH th Mark the people who received donations, thus forming a portrait collection of the recipients; Among them, n is the number of key features carried by the recipients, α i is the influence coefficient of the i-th key feature, and i∈[1,n].

4. The system for accurately searching and verifying donated recipients based on data analysis according to claim 1 is characterized in that: The method for generating the contribution coefficient by the calibration verification module is: Obtain the personnel-related data information of each recipient in the recipient portrait set, and thus obtain the recipient's disease impact value R a And the capital impact value R b ; Then, through the formula R=R a *w1+R b *w2 is used to calculate the donation coefficient R of the recipient; Wherein, w1 and w2 are respective preset proportional coefficients.

5. The system for accurately searching and verifying donated recipients based on data analysis according to claim 4 is characterized in that: The disease condition affects the value R a The method to obtain it is: The cumulative total number of visits m, the medical expenses MY and the length of hospital stay DY of each recipient are obtained from the historical database of the corresponding region, so as to formulate the curve of medical expenses changing with the number of visits x MY(x) and the curve of hospital stay changing with the number of visits x DY(x); By formula Obtain the disease impact value R a ; Among them, DR is the severity of the disease, A is the current recovery level of the disease, MY0(x) is the curve of the standard medical expenses set by the system changing with the number of visits, and DY0(x) is the curve of the standard hospitalization time set by the system changing with the number of visits.

6. The system for accurately searching and verifying donated recipients based on data analysis according to claim 4 is characterized in that: The capital impact value R b The acquisition method is: Obtain the medical expenses MY consumed by the recipient each time he / she visits the hospital and the family asset value AS of the recipient; By formula Get the capital impact value R b ; Among them, MY j is the medical expenses consumed during the jth visit, and j∈[1,m], δ is the average personal expenditure in the area where the recipient is located, θ is the fixed reimbursement amount of medical insurance in the area where the recipient is located, ξ is the patient age coefficient, ω is the disease stage coefficient, γ is the medical insurance status coefficient of the recipient, and r a is the number of minors in the recipient's family, r b The number of elderly people in the families of the recipients of donations.

7. The system for accurately searching and verifying donated recipients based on data analysis according to claim 4 is characterized in that: The method by which the determination module determines the final recipient of donations based on the ranking results is: Obtain the donation coefficient R of the GG recipients in the portrait set of recipients, sort each person in descending order according to the donation coefficient R, select the top L recipients according to the predetermined number of recipients L and the donation amount, and give the top L recipients a corresponding proportion of donation funds according to the proportion of the donation funds.

8. The system for accurately searching and verifying donated recipients based on data analysis according to claim 1 is characterized in that: The method for the security warning module to perform security warning protection on the data in the system is: Get the information of the person who logged into the system and determine whether he is an authorized login person. If he is not authorized, an alarm will be issued. Otherwise, proceed to the next step. Get a period ΔT n Curve Q of the number of views of logged-in users changing over time r (t) and the download volume change curve D r (t); Thus, through the formula Get the risk value K r ; when When , an alarm is issued; in, is the preset risk threshold, maxQ r '(t) is Q r Maximum slope of (t), minQ r '(t) is Q r The minimum slope of (t), maxD r '(t) is D r The maximum slope of (t), minD r '(t) is D r Minimum slope of (t), ΔQ r and ΔD r are the preset slope comparison values, and t1 is the period ΔT n The start time of t2 is the period ΔT n End time, Q r0 (t) is the curve of the preset standard page views changing over time, This is a curve showing the change in download volume over time for a preset standard.

Images