Network security analysis method based on artificial intelligence

Abstract

The invention relates to the technical field of network security analysis, and particularly discloses a network security analysis method based on artificial intelligence, and the method comprises the following steps: S1, obtaining a flow index of a website, and calculating an important coefficient of the website according to the obtained flow index; s2, obtaining the page jumping times when the visitor accesses the website and calculating the bandwidth occupied by accessing the website; s3, obtaining the number of visitors when the website crashes, and calculating the total bandwidth when the website crashes, the standard value of the number of the visitors and the average value of the total bandwidth; s4, the number of visitors of the website at the current moment is obtained, and first-class pressure and second-class pressure are calculated; and S5, calculating a website pressure comprehensive value, and analyzing the website security according to the website pressure comprehensive value. The network security analysis method is provided by analyzing the operation log of the target webpage and summarizing the rule, so that the collapse probability of the website is reduced, and the network security level is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of network security analysis, and in particular to a network security analysis method based on artificial intelligence. Background Art

[0002] Network security refers to the use of various technologies and management measures to protect the hardware, software and data of network systems from attacks, destruction, changes, leaks and improper use. Network security is a multi-dimensional concept. It not only involves the confidentiality, integrity and availability of information, but also covers the continuous and reliable operation of the system and the uninterrupted network services. This means that the scope of network security naturally includes the requirement to ensure the stability of the website. As an important form of network service, the stability of the website is crucial to user experience, business continuity and data security.

[0003] In the existing technology, when the number of visits to a website increases dramatically, the server may become overloaded because it cannot handle too many concurrent requests, resulting in a slow response speed or even crash, making it impossible to correctly analyze the current network security. Therefore, an efficient and accurate method is needed to analyze the current network security situation. Summary of the invention

[0004] The purpose of the present invention is to provide a network security analysis method based on artificial intelligence to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A network security analysis method based on artificial intelligence includes the following steps: S1: Obtain the traffic index of the website, which includes the total page views PV and the number of unique visitors UV, where the unique visitor refers to a visitor who does not visit the website repeatedly. The importance coefficient W of the website is calculated by a formula, which is specifically: ; Wherein, T is the website establishment time, γ1 is the preset first adjustment value, γ2 is the preset second adjustment value, and 0<γ1<γ2; S2: Get the number of page jumps when visitors visit the website, and calculate the bandwidth consumed by visitors to visit the website , where BW start Represents the bandwidth consumed by the website loading the login page, BW j Represents the bandwidth consumed by loading the page after the jth page jump; S3: Obtain the number of visitors N when the website crashes based on the operation log, and calculate the total bandwidth occupied when the website crashes , where BW total_nRepresents the bandwidth consumed by the nth visitor to the website; Calculate the average number of visitors and the average total bandwidth when the site crashed , where I represents the number of website crashes in the operation log, and N i represents the number of visitors when the website crashes for the i-th time, ZBW i Represents the total bandwidth occupied when the i-th website crashes; S4: Get the number of visitors N of the website at the current moment now , calculate the number ratio δ1=N now / N ave , calculate a type of pressure on the website at the current moment , where μ1 represents the preset first coefficient; Get the total bandwidth ZBW occupied by the website at the current moment now , calculate the bandwidth ratio δ2=ZBW now / ZBW ave , calculate the second type of pressure of the website at the current moment , where μ2 represents a preset second coefficient; S5: Calculate the comprehensive value of website pressure , and set the congestion interval of the website [YL sta -A, YL sta +A], where YL sta is the preset standard pressure value, A is the preset fluctuation range, and the safety status of the website is determined based on the comprehensive website pressure value.

[0006] As a further solution of the present invention: in the step S1, if the website establishment time T<T min , then stop the subsequent steps for the website, where T min Represents the preset minimum time for building a website.

[0007] As a further solution of the present invention: in the step S1, the browsing time t of the visitor browsing the website is obtained, and the browsing time t is less than t min The browsing records are not included in the total page views PV, where t min The preset minimum browsing time.

[0008] As a further solution of the present invention: in the step S2, the bandwidth BW consumed by loading the page after the j-th jump is calculated. j The methods specifically include: Get the number of web page files M after the jth jump page j , get the bandwidth bw consumed by loading the web page file, and calculate the bandwidth consumed by loading the page , where bw mRepresents the bandwidth consumed by loading the mth web page file after jumping to the page.

[0009] As a further solution of the present invention: in the step S3, the bandwidth consumed by page jump is eliminated and does not participate in the bandwidth consumption BW for accessing the website. total Calculation.

[0010] As a further solution of the present invention: in step S4, if the number of visitors to the website at the current moment is N now =0, then stop the subsequent steps and determine that the website is in a safe state at this time.

[0011] As a further solution of the present invention: in the step S5, the method for determining the security status of the website based on the comprehensive value of the website pressure specifically includes: If the website pressure comprehensive value YL<YL sta -A, indicating that the current website has a low probability of crashing and is in a safe state; If the website pressure comprehensive value YL is in the crowded interval [YL sta -A, YL sta +A], it means that the access speed of the current website has slowed down and the probability of crash is high, and the website is in an unstable state; If the website pressure comprehensive value YL>YL sta +A means the current website is in an unsafe state, reminding the website administrator to manage it.

[0012] As a further solution of the present invention: in the step S5, the comprehensive pressure value YL of the website when there is no visitor is obtained. null , set the standard pressure value YL sta When ensuring the comprehensive pressure value YL null <YL sta -A.

[0013] Beneficial effects of the present invention: In the present invention, the traffic indicators of the website are first obtained, wherein the traffic indicators of the website include the total page views PVhe the number of independent visitors UV. The importance of the website can be judged by the traffic indicators of the website. Therefore, in order to reduce the negative impact brought by the crash of the website, a higher judgment level will be set for the websites with higher importance during the security analysis, thereby reducing the probability of the website crashing.

[0014] The reason why the above two traffic indicators are selected is that they can describe the importance of the website in a balanced and objective manner. Among them, the total page views PV is one of the important indicators for measuring website traffic and visitor activities. The total page views PV refers to the page views or clicks within the website. It indicates the total number of visits to all pages on the website by visitors within a period of time. Whenever a visitor jumps to a page or refreshes the current page within the website, the total page views PV will increase by 1. By collecting and analyzing the total page views PV, you can understand the overall traffic situation of the website and evaluate the popularity of the page. High total page views usually mean that the website content is distinctive and attractive, and the visitor stays for a relatively long time. Therefore, the higher the total page views PV, the greater the pressure on the target. Therefore, when analyzing the security of the website, the higher the total page views PV, the lower the stability and security of the website. Since the rapid growth rate of the total page views PV will affect the final security analysis, it is necessary to slow down the growth rate of PV. According to the formula ln (lnPV), it can be seen that the growth rate of PV is much higher than the other two traffic indicators, so the growth rate of PV needs to be controlled within a reasonable range.

[0015] Unique Visitors UV refers to the number of unique visitors who visit a website within a specific period of time. It should be noted that, unlike PV, UV emphasizes the number of unique visitors. Therefore, when a person visits a website multiple times, the UV value is only counted as 1, while PV will be recorded multiple times. UV is crucial for website analysis. UV can reflect how many different visitors a website has attracted. A higher UV means that the website has a wide audience coverage. Therefore, the larger the audience coverage of a website, the more visitor information it retains and the higher the security requirements for its website. Once the website crashes, the probability of visitor information leakage will greatly increase.

[0016] Regarding the number of independent visitor IPs, the number of independent IPs is one of the important indicators to measure the popularity of a website. It reflects how many different devices have visited the website within a certain period of time. A higher number of independent IPs usually means that the website has attracted more independent visitors, indicating that the website content is attractive to users. Usually, IP is equal to UV, but there may be a virtual IP address. If the virtual IP address is also regarded as an independent visitor, the final analysis result will be inaccurate, so in the present invention, the independent visitor IP is discarded to achieve the effect of improving accuracy.

[0017] Then, the bandwidth of the visitor's access to the website is calculated by the number of page jumps J when the visitor visits the website. The reason why the bandwidth of the visitor's access to the website is calculated is that the content of different web pages is different. For example, some web pages contain video files, which are usually large, so more bandwidth and time are required to download. In addition, the storage and transmission of video files have high requirements for server resources. When a large number of visitors request video content at the same time in a short period of time, the website server may face overload, which may cause the response speed to slow down and reduce the visitor experience, or even interrupt the service and crash the website. Therefore, it is necessary to calculate the bandwidth of the visitor's access to the website based on the number of page jumps, so that the calculation result is more accurate.

[0018] The number of visitors N when the website crashes is obtained according to the operation log. When the website crashes, the number of visitors at this time cannot be obtained. Therefore, the number of visitors N when the website crashes is approximately regarded as the number when the website crashes. This method can roughly estimate the number of visitors when the website crashes, and thus calculate the total bandwidth ZBW when the website crashes, and obtain a relatively accurate result.

[0019] And according to the operation log of the website, we can calculate the standard value of the number of visitors and the average value of the total bandwidth when the website crashes. The above two parameters can roughly determine the maximum number of visitors and the maximum bandwidth value that the website can withstand. Then, we can judge the security of the website according to the calculated maximum number of visitors and the maximum bandwidth value that the website can withstand, so as to conduct more accurate analysis and improve the accuracy of network security analysis.

[0020] Get the number of visitors N of the website at the current moment now , the first type of pressure and second type of pressure of the website are calculated by formula. According to the formula, when the number ratio δ1 is greater than 1, the value of the first type of pressure will increase. As the number ratio δ1 increases, the first type of pressure on the website will increase exponentially. When the number ratio δ1 is less than 1, the first type of pressure will decrease. The total bandwidth of the website can be understood in the same way.

[0021] Calculate the comprehensive stress value of the website and analyze it based on the comprehensive stress value of the website. The formula shows that the higher the importance coefficient W of the website, the greater the comprehensive stress value YL of the website, because the higher the importance coefficient W of the website, the more important it is. Therefore, when analyzing, the comprehensive stress value YL of the website is increased by the importance coefficient W of the website, so that buffer space can be reserved for website crashes, reducing the possibility of website crashes.

[0022] In summary, the present invention provides an accurate network security analysis method by analyzing the target web page operation log and summarizing the rules, thereby reducing the probability of website crash and improving network security. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The present invention will be further described below in conjunction with the accompanying drawings.

[0024] Figure 1 It is a flow chart of a network security analysis method based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0025] 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.

[0026] See also Figure 1 As shown, the present invention is a network security analysis method based on artificial intelligence, comprising the following steps: S1: Obtain the traffic index of the website, which includes the total page views PV and the number of unique visitors UV, where the unique visitor refers to the visitor who does not visit the website repeatedly, and calculate the importance coefficient W of the website by the formula, which is specifically: ; Wherein, T is the website establishment time, γ1 is the preset first adjustment value, γ2 is the preset second adjustment value, and 0<γ1<γ2; S2: Get the number of page jumps when visitors visit the website, and calculate the bandwidth consumed by visitors to visit the website , where BW start Represents the bandwidth consumed by the website loading the login page, BW j Represents the bandwidth consumed by loading the page after the jth page jump; S3: Obtain the number of visitors N when the website crashes based on the operation log, and calculate the total bandwidth occupied when the website crashes , where BW total_n Represents the bandwidth consumed by the nth visitor to the website; Calculate the average number of visitors and the average total bandwidth when the site crashed , where I represents the number of website crashes in the operation log, and N i represents the number of visitors when the website crashes for the i-th time, ZBW i Represents the total bandwidth occupied when the i-th website crashes; S4: Get the number of visitors N of the website at the current moment now , calculate the number ratio δ1=N now / N ave , calculate a type of pressure on the website at the current moment , where μ1 represents the preset first coefficient; Get the total bandwidth ZBW occupied by the website at the current moment now , calculate the bandwidth ratio δ2=ZBW now / ZBW ave , calculate the second type of pressure of the website at the current moment , where μ2 represents a preset second coefficient; S5: Calculate the comprehensive value of website pressure , and set the congestion interval of the website [YL sta -A, YL sta +A], where YL sta is the preset standard pressure value, A is the preset fluctuation range, and the safety status of the website is determined based on the comprehensive website pressure value.

[0027] It should be noted that the website's traffic indicators are obtained first, where the website's traffic indicators include the total page views (PV) and the number of unique visitors (UV). The importance of the website can be judged through the website's traffic indicators. Therefore, in order to reduce the negative impact of website crashes, a higher judgment level will be set for websites with higher importance during security analysis, thereby reducing the probability of website crashes.

[0028] The reason why the above two traffic indicators are selected is that they can describe the importance of the website in a balanced and objective manner. Among them, the total page views PV is one of the important indicators for measuring website traffic and visitor activities. The total page views PV refers to the page views or clicks within the website. It indicates the total number of visits to all pages on the website by visitors within a period of time. Whenever a visitor jumps to a page or refreshes the current page within the website, the total page views PV will increase by 1. By collecting and analyzing the total page views PV, you can understand the overall traffic situation of the website and evaluate the popularity of the page. High total page views usually mean that the website content is distinctive and attractive, and the visitor stays for a relatively long time. Therefore, the higher the total page views PV, the greater the pressure on the target. Therefore, when analyzing the security of the website, the higher the total page views PV, the lower the stability and security of the website. Since the rapid growth rate of the total page views PV will affect the final security analysis, it is necessary to slow down the growth rate of PV. According to the formula ln (lnPV), it can be seen that the growth rate of PV is much higher than the other two traffic indicators, so the growth rate of PV needs to be controlled within a reasonable range.

[0029] Unique Visitors UV refers to the number of unique visitors who visit a website within a specific period of time. It should be noted that, unlike PV, UV emphasizes the number of unique visitors. Therefore, when a person visits a website multiple times, the UV value is only counted as 1, while PV will be recorded multiple times. UV is crucial for website analysis. UV can reflect how many different visitors a website has attracted. A higher UV means that the website has a wide audience coverage. Therefore, the larger the audience coverage of a website, the more visitor information it retains and the higher the security requirements for its website. Once the website crashes, the probability of visitor information leakage will greatly increase.

[0030] Regarding the number of independent visitor IPs, the number of independent IPs is one of the important indicators to measure the popularity of a website. It reflects how many different devices have visited the website within a certain period of time. A higher number of independent IPs usually means that the website has attracted more independent visitors, indicating that the website content is attractive to users. Usually, IP is equal to UV, but there may be a virtual IP address. If the virtual IP address is also regarded as an independent visitor, the final analysis result will be inaccurate, so in the present invention, the independent visitor IP is discarded to achieve the effect of improving accuracy.

[0031] Then, the bandwidth of the visitor's access to the website is calculated by the number of page jumps J when the visitor visits the website. The reason why the bandwidth of the visitor's access to the website is calculated is that the content of different web pages is different. For example, some web pages contain video files, which are usually large, so more bandwidth and time are required to download. In addition, the storage and transmission of video files have high requirements for server resources. When a large number of visitors request video content at the same time in a short period of time, the website server may face overload, which may cause the response speed to slow down and reduce the visitor experience, or even interrupt the service and crash the website. Therefore, it is necessary to calculate the bandwidth of the visitor's access to the website based on the number of page jumps, so that the calculation result is more accurate.

[0032] The number of visitors N when the website crashes is obtained according to the operation log. When the website crashes, the number of visitors at this time cannot be obtained. Therefore, the number of visitors N when the website crashes is approximately regarded as the number when the website crashes. This method can roughly estimate the number of visitors when the website crashes, and thus calculate the total bandwidth ZBW when the website crashes, and obtain a relatively accurate result.

[0033] And according to the operation log of the website, we can calculate the standard value of the number of visitors and the average value of the total bandwidth when the website crashes. The above two parameters can roughly determine the maximum number of visitors and the maximum bandwidth value that the website can withstand. Then, we can judge the security of the website according to the calculated maximum number of visitors and the maximum bandwidth value that the website can withstand, so as to conduct more accurate analysis and improve the accuracy of network security analysis.

[0034] Get the number of visitors N of the website at the current moment now , the first type of pressure and second type of pressure of the website are calculated by formula. According to the formula, when the number ratio δ1 is greater than 1, the value of the first type of pressure will increase. As the number ratio δ1 increases, the first type of pressure on the website will increase exponentially. When the number ratio δ1 is less than 1, the first type of pressure will decrease. The total bandwidth of the website can be understood in the same way.

[0035] Calculate the comprehensive stress value of the website and analyze it based on the comprehensive stress value of the website. The formula shows that the higher the importance coefficient W of the website, the greater the comprehensive stress value YL of the website, because the higher the importance coefficient W of the website, the more important it is. Therefore, when analyzing, the comprehensive stress value YL of the website is increased by the importance coefficient W of the website, so that buffer space can be reserved for website crashes, reducing the possibility of website crashes.

[0036] In another preferred embodiment of the present invention, if the website establishment time T<T min , then stop the subsequent steps for the website, where T min Represents the preset minimum time for building a website.

[0037] It is worth noting that when the website is built for a short time, the amount of valid data that the website can provide is small. Therefore, it is impossible to perform subsequent steps for the website with a short construction time. If the subsequent steps are forcibly executed, it will lead to large errors, thus causing wrong judgments.

[0038] And because the website was built a short time ago and its popularity is not high, it is difficult for the website to crash due to too many visitors. Therefore, this method cannot be used to analyze websites with a short building time.

[0039] In another preferred embodiment of the present invention, the browsing time t of the visitor browsing the website is obtained, and the browsing time t is less than t min The browsing records are not included in the total page views PV, where t min The preset minimum browsing time.

[0040] It is understandable that in actual situations, the website may be opened and closed repeatedly. If all such behaviors are recorded, the PV value will increase, even though mitigation measures have been taken for PV. In order to improve the accuracy of the final result analysis, the website access time t is set to be less than the preset minimum browsing time t min The browsing history does not participate in the calculation of the total page views PV.

[0041] In another preferred embodiment of the present invention, the bandwidth BW consumed by loading the page after the j-th jump is calculated. j The methods specifically include: Get the number of web page files M after the jth jump page j , get the bandwidth bw consumed by loading the web page file, and calculate the bandwidth consumed by loading the page , where bw m Represents the bandwidth consumed by loading the mth web page file after jumping to the page.

[0042] It should be noted that there are multiple web resources in a web page. For example, web resources may include web videos, web images, and web layouts. Different web resources occupy different sizes of bandwidth resources. Among them, web videos will occupy a large amount of bandwidth, which will cause a sudden increase in website pressure. Therefore, it is not advisable to simply regard different web resources as equal in size. It is necessary to calculate according to the actual situation of different pages so as to accurately analyze the security of the website.

[0043] In another preferred embodiment of the present invention, the bandwidth consumed by page jumps is eliminated and does not participate in the bandwidth consumption BW for accessing the website. total Calculation.

[0044] It should be noted that in the actual website operation process, jumping between web pages also consumes bandwidth, but generally speaking, the bandwidth consumed by web page jumping is almost negligible compared to the bandwidth consumed by web page loading. Therefore, the bandwidth consumed by page jumping is excluded and does not participate in the bandwidth consumption of visiting the website. total This not only reduces the amount of calculation but also simplifies the operation.

[0045] In another preferred embodiment of the present invention, if the number of visitors to the website at the current moment is N now =0, then stop the subsequent steps and determine that the website is in a safe state at this time.

[0046] It is understandable that when the number of visitors to the website is N now =0, under normal circumstances the website will not crash, so stop the subsequent steps on the website and analyze that the website is in a safe state at this time, thereby achieving the purpose of simplifying operations.

[0047] In another preferred embodiment of the present invention, the method for determining the security status of a website based on the website pressure comprehensive value specifically includes: If the website pressure comprehensive value YL<YL sta -A, indicating that the current website has a low probability of crashing and is in a safe state; If the website pressure comprehensive value YL is in the crowded interval [YL sta -A, YL sta+A], it means that the access speed of the current website has slowed down and the probability of crash is high, and the website is in an unstable state; If the website pressure comprehensive value YL>YL sta +A means the current website is in an unsafe state, reminding the website administrator to manage it.

[0048] It is worth noting that the method for determining the security status of a website based on the comprehensive website pressure value needs to be judged based on the comprehensive website pressure value YL. The higher the comprehensive website pressure value YL, the greater the probability of the website crashing. When the comprehensive website pressure value is small, visitors can access the website normally, and the probability of the website crashing is low. When the website pressure value is large, the number of visitors needs to be limited to prevent the website from crashing.

[0049] In another preferred embodiment of the present invention, the comprehensive pressure value YL of the website when there is no visitor is obtained. null , set the standard pressure value YL sta When ensuring the comprehensive pressure value YL null <YL sta -A.

[0050] It is worth noting that the standard pressure value YL is set sta When the pressure comprehensive value YL null <YL sta -A, the comprehensive value of the website's stress when there are no visitors YL null >YL sta -A means that the website cannot be accessed normally by visitors, which violates the objective facts. Therefore, based on objective facts, it is necessary to ensure that the pressure comprehensive value YL null <YL sta -A can ensure the normal operation of the website.

[0051] 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 network security analysis method based on artificial intelligence, characterized in that: The following steps are involved: S1: Obtain the traffic index of the website, which includes the total page views PV and the number of unique visitors UV, where the unique visitor refers to a visitor who does not visit the website repeatedly. The importance coefficient W of the website is calculated by a formula, which is specifically: ； Wherein, T is the website establishment time, γ1 is the preset first adjustment value, γ2 is the preset second adjustment value, and 0<γ1<γ2; S2: Get the number of page jumps when visitors visit the website, and calculate the bandwidth consumed by visitors to visit the website , where BW start Represents the bandwidth consumed by the website loading the login page, BW j Represents the bandwidth consumed by loading the page after the jth page jump; S3: Obtain the number of visitors N when the website crashes based on the operation log, and calculate the total bandwidth occupied when the website crashes , where BW total_n Represents the bandwidth consumed by the nth visitor to the website; Calculate the average number of visitors and the average total bandwidth when the site crashed , where I represents the number of website crashes in the operation log, and N i represents the number of visitors when the website crashes for the i-th time, ZBW i Represents the total bandwidth occupied when the i-th website crashes; S4: Get the number of visitors N of the website at the current moment now , calculate the number ratio δ1=N now / N ave , calculate a type of pressure on the website at the current moment , where μ1 represents the preset first coefficient; Get the total bandwidth ZBW occupied by the website at the current moment now , calculate the bandwidth ratio δ2=ZBW now / ZBW ave , calculate the second type of pressure of the website at the current moment , where μ2 represents a preset second coefficient; S5: Calculate the comprehensive value of website pressure , and set the congestion interval of the website [YL sta -A, YL sta +A], where YL sta is the preset standard pressure value, A is the preset fluctuation range, and the safety status of the website is determined based on the comprehensive website pressure value.

2. According to the artificial intelligence-based network security analysis method of claim 1, it is characterized in that: In step S1, if the website establishment time T<T min , then stop the subsequent steps for the website, where T min Represents the preset minimum time for building a website.

3. The network security analysis method based on artificial intelligence according to claim 1 is characterized in that: In step S1, the browsing time t of the visitor browsing the website is obtained, and the browsing time t is less than t min The browsing records are not included in the total page views PV, where t min The preset minimum browsing time.

4. The network security analysis method based on artificial intelligence according to claim 1 is characterized in that: In step S2, the bandwidth BW consumed by loading the page after the j-th jump is calculated. j The methods specifically include: Get the number of web page files M after the jth jump page j , get the bandwidth bw consumed by loading the web page file, and calculate the bandwidth consumed by loading the page , where bw m Represents the bandwidth consumed by loading the mth web page file after jumping to the page.

5. The network security analysis method based on artificial intelligence according to claim 1 is characterized in that: In step S3, the bandwidth consumed by page jumps is eliminated and does not participate in the bandwidth consumption BW of accessing the website. total Calculation.

6. The network security analysis method based on artificial intelligence according to claim 1 is characterized in that: In step S4, if the number of visitors to the website at the current moment is N now =0, then stop the subsequent steps and determine that the website is in a safe state at this time.

7. The network security analysis method based on artificial intelligence according to claim 1 is characterized in that: In the step S5, the method for determining the security status of the website based on the comprehensive value of website pressure specifically includes: If the website pressure comprehensive value YL<YL sta -A, indicating that the current website has a low probability of crashing and is in a safe state; If the website pressure comprehensive value YL is in the crowded interval [YL sta -A, YL sta +A], it means that the access speed of the current website has slowed down and the probability of crash is high, and the website is in an unstable state; If the website pressure comprehensive value YL>YL sta +A means the current website is in an unsafe state, reminding the website administrator to manage it.

8. The network security analysis method based on artificial intelligence according to claim 1 is characterized in that: In step S5, the comprehensive pressure value YL of the website when there is no visitor is obtained. null , set the standard pressure value YL sta When ensuring the comprehensive pressure value YL null <YL sta -A.

Images