Dynamic routing control method and system based on user behavior analysis

Abstract

The invention relates to the technical field of information transmission control, and particularly discloses a dynamic routing control method based on user behavior analysis, which comprises the following steps: acquiring router log information, and acquiring user behavior information according to the router log information; obtaining user behavior characteristic values according to the user behavior information, wherein the user behavior characteristic values comprise a use time characteristic value, a website and application program use characteristic value and a traffic use characteristic value; dividing the users into a plurality of user categories according to the user behavior characteristic values; based on the user category and the user behavior feature value, obtaining a flow prediction value of the user in a specific time period in the future; and controlling the router to distribute the bandwidth according to the flow predicted value of the user in the specific time period in the future, so that the bandwidth can be distributed to all users according to the proportion, each user can obtain a certain bandwidth guarantee, and the situation that other users are suddenly robbed to break the network is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of information transmission control, and in particular to a dynamic routing control method and system based on user behavior analysis. Background Art

[0002] With the rapid development of Internet technology and the widespread popularity of smart devices, users have higher and higher requirements for network service quality. Traditional static routing control methods can no longer meet the dynamically changing traffic requirements in modern network environments. Static routing configuration is usually set at the beginning of network deployment. Once set, it is difficult to flexibly adjust according to changes in network traffic, which may lead to insufficient bandwidth and network congestion during peak network hours, and may waste precious network resources during off-peak hours.

[0003] In a traditional routing network, when a user performs a large-volume upload or download operation, such as downloading, movie buffering, and live streaming, network grabbing will occur, causing the network speed of other users in the routing network to be severely slowed down, affecting the normal use of a large number of other users. Summary of the invention

[0004] The purpose of the present invention is to provide a dynamic routing control method and system based on user behavior analysis to solve the technical problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A dynamic routing control method based on user behavior analysis, comprising:

[0007] Obtaining router log information, and obtaining user behavior information based on the router log information, wherein the user usage information includes connection device category, login time, connection duration, category of visited website addresses, category of application usage, and upload and download traffic values;

[0008] Acquire user behavior characteristic values ​​according to the user behavior information, wherein the user behavior characteristic values ​​include usage time characteristic values, website and application usage characteristic values, and traffic usage characteristic values;

[0009] Classifying users into a plurality of user categories according to the user behavior characteristic values;

[0010] Obtaining a traffic prediction value of the user in a specific future time period based on the user category and the user behavior feature value;

[0011] The router is controlled to allocate bandwidth according to the traffic prediction value of the user in a specific future time period.

[0012] Preferably, the step of obtaining the usage time characteristic value includes:

[0013] Divide the day into time periods;

[0014] Obtain the number of logins of the user in each of the time periods;

[0015] Obtain the total number of logins for all time periods within a preset number of days;

[0016] The ratio of the number of logins of the user in each of the time periods to the total number of logins in all the time periods of a preset number of days is calculated to obtain the usage time characteristic value.

[0017] Preferably, the website and application use the following steps to obtain the characteristic value:

[0018] Get the total number of websites and application categories run by users within a preset time period;

[0019] Get the number of times a user runs a website or application category within a preset time period;

[0020] Get the categories of websites and applications run by users during a preset time period;

[0021] Get the user's running time for each website and application category within a preset time period;

[0022] Get the user's habit change coefficient;

[0023] The website and application usage characteristic values ​​are calculated according to the website and application categories run by the user within a preset time period, the total number of website and application categories run by the user, the user's running time for each website and application category, and the habit change coefficient, and the calculation formula is:

[0024]

[0025] Among them, f jk represents the characteristic value of website and application usage, w(t i ) represents the habit change coefficient, j represents the user, t i represents the access time, k represents the category of websites and applications run by the user, and c jki represents user j’s access time t i The running time for website and application category k, m represents the total number of website and application categories run by the user, and n represents the number of times the user runs the website and application category.

[0026]

[0027] Among them, w(t i) represents the habit change coefficient, γ represents the change factor, T represents the current time, t i Indicates the access time.

[0028] Preferably, the step of acquiring the traffic usage characteristic value includes:

[0029] Divide the traffic consumption into several traffic intervals;

[0030] Divide a day into several time periods;

[0031] Obtaining the running traffic consumed by the user each time the user runs the website and the application in each time period, and calculating the number of times the running traffic is within each traffic interval;

[0032] Get the total number of times a user runs each website and application category within a preset time period;

[0033] The traffic usage characteristic value is calculated based on the ratio of the number of times the running traffic is within each of the traffic intervals to the total number of times the user uses each website and application category within a preset time period.

[0034]

[0035] Among them, L x represents the traffic usage characteristic value, a1 represents the a1th traffic interval, Y a1 represents the weight of the a1th flow interval, d a1 represents the number of times the operating flow is in the a1th flow interval, v k The total number of times users run website and application category k within the preset time period, a2 represents the a2th traffic interval, and Y a2 Indicates a2 The weight of the flow interval, d a2 Indicates the number of times the operating flow is in the a2th flow interval.

[0036] Preferably, the step of dividing users into several categories according to the user behavior feature values ​​comprises:

[0037] Obtain the user's usage time characteristic values, website and application usage characteristic values, and traffic usage characteristic values;

[0038] Randomly select a user as the initial user category;

[0039] Calculate the difference between the remaining users and the initial user category, and take the user with the largest difference between the initial user category and the second user category;

[0040] Calculate the difference between the remaining users and the initial user category and the second user category, and take the user with the largest difference between the initial user category and the second user category as the third user category;

[0041] Repeat the above steps to divide users into several categories.

[0042] Preferably, the step of obtaining the traffic prediction value of the user in a specific future time period based on the category to which the user belongs and the user behavior feature value includes:

[0043] Get the user behavior characteristic value of the current visiting user;

[0044] Compare the behavior characteristic value of the current access user with the user behavior characteristic value of the classified user, and select the user behavior characteristic value of the classified user that is closest to the current access user to classify the current access user;

[0045] Based on the user behavior feature values ​​of the classified users, the traffic forecast value of the user in a specific time period in the future is obtained.

[0046] Preferably, the step of controlling the router to allocate bandwidth according to the traffic prediction value of the user in a specific future time period includes:

[0047] Get the categories of websites and applications run by the user;

[0048] Get the weight of the categories of websites and applications you run;

[0049] Bandwidth is allocated based on the weight of said website and application categories.

[0050] The present invention also discloses a dynamic routing control system based on user behavior analysis, comprising:

[0051] The first module is used to obtain user behavior information based on the router log information;

[0052] In the second module, users calculate usage time characteristic values, website and application usage characteristic values, and traffic usage characteristic values;

[0053] The third module is used to divide users into several categories according to the usage time characteristic value, the website and application usage characteristic value and the traffic usage characteristic value, and calculate the traffic prediction value of the user in a specific time period in the future based on the user category and the user behavior characteristic value;

[0054] The fourth module is used to control the router to allocate bandwidth according to the user category.

[0055] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned dynamic routing control method based on user behavior analysis when executing the computer program.

[0056] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned dynamic routing control method based on user behavior analysis are implemented.

[0057] The beneficial effect of the present application is that bandwidth can be allocated to all users in proportion, so that each user can obtain a certain bandwidth guarantee without causing other users to be suddenly disconnected from the network due to network scramble. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 The figure is a schematic diagram of a method flow of an embodiment of the present application.

[0059] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0060] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0061] like Figure 1 As shown, the present application provides a dynamic routing control method based on user behavior analysis, comprising:

[0062] S1, obtaining router log information, and obtaining user behavior information according to the router log information, wherein the user usage information includes connection device category, login time, connection duration, category of visited website addresses, category of application usage, and upload and download traffic values;

[0063] S2, obtaining a user behavior characteristic value according to the user behavior information, wherein the user behavior characteristic value includes a usage time characteristic value, a website and application usage characteristic value, and a traffic usage characteristic value;

[0064] S3, dividing the user into several user categories according to the user behavior feature value;

[0065] S4, obtaining a traffic prediction value of the user in a specific future time period based on the user category and the user behavior feature value;

[0066] S5, controlling the router to allocate bandwidth according to the traffic prediction value of the user in a specific future time period.

[0067] The step of obtaining the usage time characteristic value comprises:

[0068] Divide the day into time periods;

[0069] Obtain the number of logins of the user in each of the time periods;

[0070] Obtain the total number of logins for all time periods within a preset number of days;

[0071] The ratio of the number of logins of the user in each of the time periods to the total number of logins in all the time periods of a preset number of days is calculated to obtain the usage time characteristic value.

[0072] For example, a day can be divided into 24 hours, with each hour as a time period; or in order to analyze user behavior more finely, a day can be divided into 48 time periods, that is, each half hour is a time period. Such a division helps to more accurately capture the user's login behavior patterns at different time points; suppose we use each hour as a time period, and count the number of times users log in to the router during the time period from 0:00 to 1:00 through router log records. For example, on a certain day, there are 5 login records from 0:00 to 1:00, 3 login records from 1:00 to 2:00, and so on, to obtain the exact number of login data in each time period; set a preset number of days, such as 7 days. In these 7 days, count the total number of logins in each time period. For example, for the time period from 0:00 to 1:00, add up the number of logins in this time period every day in these 7 days. Assuming that the number of logins from 0:00 to 1:00 on the first day is 5 times, 4 times on the second day, 6 times on the third day... 3 times on the seventh day, then the total number of logins in this time period within the preset number of days is 5+4+6+...+3=35 times (this is just an example, the actual calculation needs to be based on real data); taking the above 0:00-1:00 time period as an example, if the number of logins in this time period on a certain day is 4 times, and the total number of logins in this time period within the preset number of days (7 days) is 35 times, then the usage time characteristic value from 0:00 to 1:00 on that day is 4÷35≈0.114. Through such calculations, the usage time characteristic value of each time period in a day can be obtained, thereby reflecting the user's tendency to use the network in different time periods, and providing an important basis for subsequent dynamic routing control based on user behavior analysis. For example, if the usage time characteristic value of a certain time period is high, it means that the user uses the network more frequently in this time period, and when allocating bandwidth, it can be appropriately considered to reserve more resources for this time period.

[0073] Preferably, the website and application use the following steps to obtain the characteristic value:

[0074] Get the total number of websites and application categories run by users within a preset time period;

[0075] Get the number of times a user runs a website or application category within a preset time period;

[0076] Get the categories of websites and applications run by users during a preset time period;

[0077] Get the user's running time for each website and application category within a preset time period;

[0078] Get the user's habit change coefficient;

[0079] The website and application usage characteristic values ​​are calculated according to the website and application categories run by the user within a preset time period, the total number of website and application categories run by the user, the user's running time for each website and application category, and the habit change coefficient, and the calculation formula is:

[0080]

[0081] Among them, f jk represents the characteristic value of website and application usage, w(t i ) represents the habit change coefficient, j represents the user, t i represents the access time, k represents the category of websites and applications run by the user, and c jki represents user j’s access time t i The running time for website and application category k, m represents the total number of website and application categories run by the user, and n represents the number of times the user runs the website and application category.

[0082] Assume that user j has a running time c of website and application category k (such as live broadcast software) at access time t1 (for example, 10:00 on October 1, 2023). jk1 = 2 hours, the corresponding habit change coefficient w(t1) = 0.8 (assumed value); at the access time t1 (such as 15:00 on October 5, 2023) the running time c of the live broadcast software jk2 = 1.5 hours, the corresponding habit change coefficient w(t2) = 0.7 (assumed value), if n = 5 such access records are counted within the preset time period, then the numerator (calculate and add the products corresponding to other access records in sequence); the denominator needs to perform similar calculations and sums for all website and application categories m (such as 10) and the number of visits n under each category. Assume that for the office software category (category k), the user's running time c at different access times t3 is jk3 There are also corresponding records of the habit change coefficient w(t3), and its Then add up the sum of all categories (including live streaming software and office software, a total of 10 categories) to get the value of the denominator; substitute the values ​​of the numerator and denominator into the formula The user's usage characteristic values ​​for websites and application categories (such as live broadcast software and office software) can be calculated. Through such calculations, the user's preference for different website and application categories can be quantified. After considering the habit change coefficient, the user's behavior pattern can be more dynamically reflected, providing a basis for subsequent user classification and bandwidth allocation operations. For example, if a user has a high characteristic value for online video applications, corresponding bandwidth guarantee or optimization can be performed according to its importance when allocating network resources.

[0083] The calculation formula of habit variation coefficient is:

[0084]

[0085] Among them, w(t i ) represents the habit change coefficient, γ represents the change factor, T represents the current time, t i Indicates the access time.

[0086] The change factor is set according to the characteristics and business objectives of the network service. If the network service pays more attention to the impact of the user's recent behavior on the current decision and hopes to quickly respond to changes in the user's current behavior, a larger change factor (such as 0.9-0.99) can be set to make the weight of recent visits higher; if the network service needs to comprehensively consider the user's behavior over a longer period of time, such as long-term trend analysis of user behavior, and does not want the difference in weight between recent and long-term behavior to be too large, a smaller attenuation factor (such as 0.8-0.9) can be set; for example, for a rapidly changing usage environment, its main goal is to stay close to the user's recent habits in order to improve user experience and obtain stable traffic. In order to highlight the importance of recent behavior, the attenuation factor is set to 0.95, so that the weight of the user's recent behavior is 0.95, and the weight of earlier behavior will be reduced to, so that the prediction is more inclined to the user's recent behavior;

[0087] Preferably, the step of acquiring the traffic usage characteristic value includes:

[0088] Divide the traffic consumption into several traffic intervals;

[0089] Divide a day into several time periods;

[0090] Obtaining the running traffic consumed by the user each time the user runs the website and the application in each time period, and calculating the number of times the running traffic is within each traffic interval;

[0091] Get the total number of times a user runs each website and application category within a preset time period;

[0092] The traffic usage characteristic value is calculated based on the ratio of the number of times the running traffic is within each of the traffic intervals to the total number of times the user uses each website and application category within a preset time period.

[0093]

[0094] Among them, Lx represents the traffic usage characteristic value, a1 represents the a1th traffic interval, Ya1 represents the weight of the a1th traffic interval, and d a1 represents the number of times the operating flow is in the a1th flow interval, v k The total number of times users run website and application category k within the preset time period, a2 represents the a2th traffic interval, Ya2 represents the weight of the a2th traffic interval, d a2 Indicates the number of times the operating flow is in the a2th flow interval.

[0095] o For example, it can be divided into a low traffic interval (0-100MB), a medium traffic interval (101-500MB) and a high traffic interval (501MB and above). This division helps to analyze the user's behavior patterns at different traffic consumption levels in more detail, and divide a day into several time periods, such as 24 time periods, each of which is 1 hour. In this way, the user's traffic usage at different times of the day can be accurately recorded. In each divided time period, such as the time period from 9 am to 10 am, the running traffic consumed by the user each time the website and application are run is obtained. Suppose the user visited a news website, an online music platform, and a social media application during this time period. Visiting the news website consumed 30MB of traffic (in the low traffic interval), visiting the online music platform consumed 200MB of traffic (in the medium traffic interval), and visiting the social media application consumed 800MB of traffic (in the medium traffic interval). Then in the time period from 9 am to 10 am, the number of low traffic intervals d a1 (Assuming a1 represents the low flow interval) increases by 1, the number of times in the medium flow interval is d a2 (Assume a2 represents a medium traffic interval) and increase by 2 times. In this way, similar traffic statistics and interval times are recorded for each time period of the day, and a preset time period is set, such as a week. In this week, the total number of times users run each website and application category is counted. For example, for the social media category, users have run it a total of 20 times in this week; for the online video category, they have run it 15 times, and so on. Assume that for a website and application category k (such as the social media category), its total number of runs v k=20, set weights for each traffic interval, for example, low traffic interval weight Ya1 = 0.1, medium traffic interval weight Ya2 = 0.3, high traffic interval weight Ya3 = 0.6, (the sum of the weights here is 1, and the weight distribution can be adjusted according to actual conditions to reflect the importance of different traffic intervals). Taking the calculation of a user's traffic usage feature value as an example, assume that in the statistical process, for social media category (k), the number of low traffic intervals (a1) is d a1 =50 (the total number of visits to the low-flow interval in all time periods within the preset time period), the number of visits to the medium-flow interval (a2) d a2 = 30, total number of runs for social media category (b1) v b1 =80. According to the formula The traffic related to the social media category is calculated using the feature value as Through such calculations, we can obtain the characteristic values ​​of users' traffic usage for different websites and application categories. This value can reflect the user's traffic consumption tendency when using various applications, and provide an important basis for the subsequent bandwidth allocation in dynamic routing control based on user behavior analysis. For example, if the traffic usage characteristic value corresponding to a certain application category is high, it means that the user consumes a large amount of traffic when using applications of this category. When allocating bandwidth, we can consider giving corresponding resource guarantees to ensure user experience.

[0096] Preferably, the step of dividing users into several categories according to the user behavior feature values ​​comprises:

[0097] Obtain the user's usage time characteristic values, website and application usage characteristic values, and traffic usage characteristic values;

[0098] Randomly select a user as the initial user category;

[0099] Calculate the difference between the remaining users and the initial user category, and take the user with the largest difference between the initial user category and the second user category;

[0100] Calculate the difference between the remaining users and the initial user category and the second user category, and take the user with the largest difference between the initial user category and the second user category as the third user category;

[0101] Repeat the above steps to divide users into several categories.

[0102] The calculation formula is:

[0103]

[0104] Among them, x ij represents the i-th user behavior feature value of the remaining user j, u kjrepresents the i-th user behavior feature value of the initial user category k, m represents the user behavior feature value, d(x i ,u k ) represents the difference between the remaining users and the initial user category k; the above formula is used to repeat the calculation until the user category no longer changes significantly (which can be determined by setting a convergence threshold, such as when the change in the difference between the previous and subsequent values ​​is less than a certain value).

[0105] Preferably, the step of obtaining the traffic prediction value of the user in a specific future time period based on the category to which the user belongs and the user behavior feature value includes:

[0106] Get the user behavior characteristic value of the current visiting user;

[0107] Compare the behavior characteristic value of the current access user with the user behavior characteristic value of the classified user, and select the user behavior characteristic value of the classified user that is closest to the current access user to classify the current access user;

[0108] Based on the user behavior feature values ​​of the classified users, the traffic forecast value of the user in a specific time period in the future is obtained.

[0109] Users with higher user behavior feature values ​​can be allocated more traffic. For historical users, they can be directly matched through comparison. For new visiting users, they can be classified and matched as quickly as possible through short log records, thereby obtaining more accurate predictions.

[0110] Preferably, the step of controlling the router to allocate bandwidth according to the traffic prediction value of the user in a specific future time period includes:

[0111] Get the categories of websites and applications run by the user;

[0112] Get the weight of the categories of websites and applications you run;

[0113] Bandwidth is allocated based on the weight of said website and application categories.

[0114] According to the categories of websites and applications currently being run by the users, the traffic forecast values ​​of the users in a specific future time period can be obtained. The bandwidth can be allocated according to the use environment of the router and the importance of the websites and applications currently being run by each user. For example, for users with large traffic demands and users with small traffic demands, the bandwidth can be allocated in proportion according to the weights based on the forecast values. For example, the forecast value of the specific future time period for user j1 is 500 mb, the forecast value of the specific future time period for user j2 is 300 mb, and the forecast value of the specific future time period for user j3 is 100 mb. The weights of the categories of websites and applications respectively run by users j1, j2 and j3 are 1, 0.8 and 0.7 (for example, the weights of entertainment websites and applications in an office environment are relatively low). Then, the bandwidth values ​​allocated to users j1, j2 and j3 are 500 mb*1, 300 mb*0.8 and 100 mb*0.7 respectively. By allocating user bandwidth in this way, it is possible to prevent other users from being robbed of the network due to excessive traffic use by a certain user.

[0115] The present invention also discloses a dynamic routing control system based on user behavior analysis, comprising:

[0116] The first module is used to obtain user behavior information based on the router log information;

[0117] In the second module, users calculate usage time characteristic values, website and application usage characteristic values, and traffic usage characteristic values;

[0118] The third module is used to divide users into several categories according to the usage time characteristic value, the website and application usage characteristic value and the traffic usage characteristic value, and calculate the traffic prediction value of the user in a specific time period in the future based on the user category and the user behavior characteristic value;

[0119] The fourth module is used to control the router to allocate bandwidth according to the user category.

[0120] The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned dynamic routing control method based on user behavior analysis when executing the computer program.

[0121] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned dynamic routing control method based on user behavior analysis are implemented.

[0122] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it may include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, value library or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0123] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0124] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent results or equivalent process changes made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A dynamic routing control method based on user behavior analysis, characterized in that: include: Obtaining router log information, and obtaining user behavior information based on the router log information, wherein the user usage information includes connection device category, login time, connection duration, category of visited website addresses, category of application usage, and upload and download traffic values; Acquire user behavior characteristic values ​​according to the user behavior information, wherein the user behavior characteristic values ​​include usage time characteristic values, website and application usage characteristic values, and traffic usage characteristic values; Classifying users into a plurality of user categories according to the user behavior characteristic values; Obtaining a traffic prediction value of the user in a specific future time period based on the user category and the user behavior feature value; The router is controlled to allocate bandwidth according to the traffic prediction value of the user in a specific future time period.

2. A dynamic routing control method based on user behavior analysis according to claim 1, characterized in that: The step of obtaining the usage time characteristic value comprises: Divide the day into time periods; Obtain the number of logins of the user in each of the time periods; Obtain the total number of logins for all time periods within a preset number of days; The ratio of the number of logins of the user in each of the time periods to the total number of logins in all the time periods of a preset number of days is calculated to obtain the usage time characteristic value.

3. A dynamic routing control method based on user behavior analysis according to claim 2, characterized in that: The website and application use the following steps to obtain the characteristic values: Get the total number of websites and application categories run by users within a preset time period; Get the categories of websites and applications run by users during a preset time period; Get the number of times a user runs a website or application category within a preset time period; Get the user's running time for each website and application category within a preset time period; Get the user's habit change coefficient; The website and application usage characteristic values ​​are calculated according to the website and application categories run by the user within a preset time period, the total number of website and application categories run by the user, the user's running time for each website and application category, and the habit change coefficient, and the calculation formula is: Among them, f jk represents the characteristic value of website and application usage, w(t i ) represents the habit change coefficient, j represents the user, t i represents the access time, k represents the category of websites and applications run by the user, and c jki represents user j’s access time t i The running time for website and application category k, m represents the total number of website and application categories run by the user, and n represents the number of times the user runs the website and application category.

4. A dynamic routing control method based on user behavior analysis according to claim 3, characterized in that: The step of obtaining the traffic usage characteristic value comprises: Divide the traffic consumption into several traffic intervals; Divide a day into several time periods; Obtaining the running traffic consumed by the user each time the user runs the website and the application in each time period, and calculating the number of times the running traffic is within each traffic interval; Get the total number of times a user runs each website and application category within a preset time period; The traffic usage characteristic value is calculated based on the ratio of the number of times the running traffic is within each of the traffic intervals to the total number of times the user uses each website and application category within a preset time period. Among them, L x represents the traffic usage characteristic value, a1 represents the a1th traffic interval, Y a1 represents the weight of the a1th flow interval, d a1 represents the number of times the operating flow is in the a1th flow interval, v k The total number of times users run website and application category k within the preset time period, a2 represents the a2th traffic interval, and Y a2 Indicates a2 The weight of the flow interval, d a2 Indicates the number of times the operating flow is in the a2th flow interval.

5. A dynamic routing control method based on user behavior analysis according to claim 4, characterized in that: The step of dividing users into several categories according to the user behavior feature values ​​comprises: Obtain the user's usage time characteristic values, website and application usage characteristic values, and traffic usage characteristic values; Randomly select a user as the initial user category; Calculate the difference between the remaining users and the initial user category, and take the user with the largest difference between the initial user category and the second user category; Calculate the difference between the remaining users and the initial user category and the second user category, and take the user with the largest difference between the initial user category and the second user category as the third user category; Repeat the above steps to divide users into several categories.

6. A dynamic routing control method based on user behavior analysis according to claim 5, characterized in that: The step of obtaining the traffic prediction value of the user in a specific future time period based on the category to which the user belongs and the user behavior feature value comprises: Get the user behavior characteristic value of the current visiting user; Compare the behavior characteristic value of the current access user with the user behavior characteristic value of the classified user, and select the user behavior characteristic value of the classified user that is closest to the current access user to classify the current access user; Based on the user behavior feature values ​​of the classified users, the traffic forecast value of the user in a specific time period in the future is obtained.

7. A dynamic routing control method based on user behavior analysis according to claim 6, characterized in that: The step of controlling the router to allocate bandwidth according to the traffic prediction value of the user in a specific future time period includes: Get the categories of websites and applications run by the user; Get the weight of the categories of websites and applications you run; Bandwidth is allocated based on the weight of said website and application categories.

8. A dynamic routing control system based on user behavior analysis, characterized in that: include: The first module is used to obtain user behavior information based on the router log information; In the second module, users calculate usage time characteristic values, website and application usage characteristic values, and traffic usage characteristic values; The third module is used to divide users into several categories according to the usage time characteristic value, the website and application usage characteristic value and the traffic usage characteristic value, and calculate the traffic prediction value of the user in a specific time period in the future based on the user category and the user behavior characteristic value; The fourth module is used to control the router to allocate bandwidth according to the user category.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images