Network information decision management method based on user traffic statistical analysis

Abstract

The invention discloses a network information decision management method based on user traffic statistical analysis, which belongs to the technical field of network information management, and comprises the following steps: regularly reading the user traffic density of a target area, and if the user traffic density is greater than a preset density threshold, activating an edge processing unit of the target area; content discrete evaluation is carried out according to the user access feature set, if the content discrete degree is smaller than or equal to a preset scalar, concentration degree analysis is carried out on the user access feature set, and hot content is cached to an edge processing unit; if the content dispersion is greater than the predetermined scalar, carrying out concentration degree analysis on the associated access feature set, and caching the associated hotspot content to an edge processing unit; user access compensation of the target area is performed using the edge processing unit. The technical problems of response delay and low resource allocation efficiency caused by incapability of dynamically identifying hot contents according to user traffic density and flexibly caching the hot contents to edge nodes in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the field of network information management, and in particular to a network information decision-making management method based on user traffic statistical analysis. Background Art

[0002] With the rapid development of the Internet and the diversified growth of user needs, the distribution of data traffic in the network system has gradually shown regional and sudden characteristics, especially during major holidays and social hot events, the traffic concentration phenomenon is more obvious. In order to cope with this dynamically changing traffic distribution, the network system not only needs to have fast response and data processing capabilities, but also provide more flexible content distribution strategies to ensure that hot content can reach users in a timely manner.

[0003] At present, in the field of network information management, decision management based on traffic statistics analysis has gradually become a mainstream technology. However, the existing decision-making schemes still face some shortcomings, especially in the lack of flexibility in the refined management of content distribution. In addition, when faced with the situation of high content discreteness, the existing system lacks an effective method to dynamically identify and manage multi-region related hot content, thus failing to achieve the optimal allocation of network resources. Therefore, there is an urgent need for a network information decision management method based on user traffic statistics analysis to solve the technical problem that the existing technology cannot dynamically identify hot content according to user traffic density and flexibly cache it to edge nodes, resulting in response delays and inefficient resource allocation. Summary of the invention

[0004] This application provides a network information decision management method based on user traffic statistical analysis, aiming to solve the technical problem in the prior art that hot spot content cannot be dynamically identified according to user traffic density and flexibly cached to edge nodes, resulting in response delays and inefficient resource allocation.

[0005] In view of the above problems, the present application provides a network information decision management method based on user traffic statistical analysis, including regularly reading the user traffic density of a target area in a backbone network, and if the user traffic density is greater than a predetermined density threshold, activating an edge processing unit of the target area, wherein the edge processing unit is communicatively connected to the backbone network; reading a user access feature set of the target area, and performing content discrete evaluation based on the user access feature set, and if the content discreteness is less than or equal to a predetermined scalar, performing a concentration analysis on the user access feature set, determining hot content, and caching the hot content to the edge processing unit; if the content discreteness is greater than the predetermined scalar, reading an associated access feature set of an associated area, performing a concentration analysis on the associated access feature set, determining associated hot content, and caching the associated hot content to the edge processing unit; and using the edge processing unit to perform user access compensation for the target area.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The technical solution of regularly reading the user traffic density of the target area in the backbone network and dynamically activating the edge processing unit according to the traffic density solves the technical problem of the inability to dynamically identify hot content according to user traffic density and flexibly cache it to the edge node, resulting in response delay and inefficient resource allocation in the prior art, and achieves the technical effect of optimizing resource allocation and improving the response efficiency of hot spots. At the same time, by analyzing the content dispersion and concentration, identifying and caching hot content, the burden on the backbone network is further reduced, and the access speed of users in hot spots is significantly improved, thereby achieving efficient use of network resources and improving user experience.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A flow chart of a network information decision management method based on user traffic statistical analysis is provided for an embodiment of the present application.

[0009] Figure 2 A flowchart of determining associated areas in a network information decision management method based on user traffic statistical analysis is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0010] The overall idea of ​​the technical solution provided by this application is as follows: The embodiment of the present application provides a network information decision management method based on user traffic statistical analysis. By monitoring and analyzing the user traffic density of the target area in the network, the edge processing unit is dynamically activated to optimize the allocation of network resources. When the traffic density exceeds a predetermined threshold, the system will read the user access characteristics and evaluate the discreteness of the content, thereby determining the hot content and caching it.

[0011] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

[0012] Examples, such as Figure 1 As shown, the embodiment of the present application provides a network information decision management method based on user traffic statistical analysis, the method comprising: Step S100: In the backbone network, the user traffic density of the target area is read periodically. If the user traffic density is greater than a predetermined density threshold, the edge processing unit of the target area is activated, wherein the edge processing unit is communicatively connected to the backbone network.

[0013] Specifically, the backbone network refers to the core network that plays the main role of data transmission in the communication system and is responsible for a large amount of data transmission between different regions. It connects regional subnets and edge devices to ensure that data can be efficiently and quickly transmitted from one region to another. The target area refers to a specific geographical area or user-concentrated area in the network system where traffic monitoring and management is required. For example, a hotspot area can be a large shopping mall, a transportation hub, or a popular business district, where sudden increases in user traffic often occur. User traffic density is a measurement of the number of users or data access frequency in the target area per unit time, reflecting the user activity in the area. An increase in traffic density usually represents an increase in network demand in the area. For example, the number of users in a shopping mall increases significantly during peak shopping hours, resulting in an increase in traffic density in the area. The predetermined density threshold refers to the critical value of user traffic density preset by the system. This threshold is used as a trigger condition. When the traffic density in the target area exceeds this threshold, corresponding control measures will be taken, such as activating the edge processing unit to enhance data processing capabilities. The edge processing unit refers to a device or node located at the edge of the network, which is used to process data close to the user. It can share the burden of the backbone network and improve data processing speed. Common edge devices include smart routers, small data centers, etc.

[0014] In the backbone network, in order to respond to sudden increases in user traffic in the target area in a timely manner, the user traffic density in the area is regularly monitored and read. Usually, a traffic monitoring system or traffic detection equipment is used, such as using Deep Packet Inspection (DPI) tools to conduct in-depth analysis of traffic, or real-time monitoring of traffic data through the NetFlow protocol. The collected traffic density data is compared with the preset density threshold. If the traffic density exceeds the threshold, the system triggers the edge processing unit to provide more direct data processing services for the target area, avoiding the need for all data to be sent back to the backbone network center for processing. This edge processing unit is connected to the backbone network through a high-speed communication link to ensure that data can be quickly transmitted between the two.

[0015] This step dynamically activates the edge processing unit, so that the system can share the processing pressure of the backbone network during peak traffic periods, realize local data processing, greatly reduce the bandwidth requirements of the backbone network, shorten the delay when users access, and thus significantly improve the smoothness of user experience and the efficiency of network resource utilization. This method is particularly suitable for areas where user traffic is concentrated and varies greatly. It can flexibly respond to demand peaks and enhance the reliability and stability of network services.

[0016] Step S200: Read the user access feature set of the target area, perform content discrete evaluation based on the user access feature set, if the content discreteness is less than or equal to a predetermined scalar, perform concentration analysis on the user access feature set, determine hot content, and cache the hot content to the edge processing unit.

[0017] Specifically, the user access feature set refers to a data set formed by statistics and analysis of the access behavior of users in a specific area over a period of time, including access content, access frequency, access time, etc. These features can help identify user preferences and behavior patterns. Content dispersion evaluation is an evaluation method used to measure the distribution of access content in the user access feature set among different types or categories, usually by calculating the diversity and dispersion of access content. A lower dispersion means that users access content more concentratedly, while a higher dispersion indicates that users' interest in content is more dispersed. Concentration analysis refers to an in-depth analysis of the user access feature set to identify content with higher access frequency, thereby determining hot content. Common methods include frequency statistics, cluster analysis, etc. Hot content refers to content that is frequently accessed by users within a specific time, which usually reflects the user's current interests or needs. This type of content is of great significance for content distribution and resource management. Edge Processing Unit refers to a device located at the edge of the network, which is used to process and cache data to reduce the burden on the backbone network and improve response speed.

[0018] First, you need to read the user access feature set of the target area, usually with the help of data collection tools, such as WebAnalytics (website analysis) tools (such as Google Analytics, Adobe Analytics, etc.) to collect and organize data. It can record the content, frequency and time of user visits to form a detailed access feature set.

[0019] Next, the content is evaluated for discreteness based on the acquired user access feature set, and the discreteness of the accessed content is calculated. The discreteness calculation can be achieved by counting the diversity of user accessed content, for example, by calculating the standard deviation or variance of content access to evaluate the dispersion of the content. If the content discreteness is less than or equal to a predetermined scalar, it means that the user's interest in the content is relatively concentrated, and further concentration analysis will be performed. Specifically, the system will cluster or classify the content with higher access frequency, for example, through the K-Means clustering algorithm to determine the hot content.

[0020] Once hot content is identified, the system will cache it to the edge processing unit to quickly respond to user access requests. For example, on a news website, if a news article is accessed by a large number of users in a short period of time, the system determines that the article is hot content through discrete evaluation and concentration analysis, and caches it to the edge processing unit to ensure that other users can quickly obtain the content when accessing it without having to request it from the backbone network.

[0021] This step reads the user access feature set and performs discrete evaluation and concentration analysis. The system can quickly identify and cache hot content, thereby greatly improving the response speed and experience of user access. Especially in scenarios where user needs change rapidly, it can effectively reduce data transmission time and improve the smoothness of content access. In addition, this method can also reduce the burden on the backbone network and optimize resource allocation, so that the entire network system can still operate stably under high load conditions.

[0022] Step S300: If the content dispersion is greater than the predetermined scalar, read the associated access feature set of the associated area, perform concentration analysis on the associated access feature set, determine associated hotspot content, and cache the associated hotspot content to the edge processing unit.

[0023] Specifically, the predetermined scalar is the threshold of content dispersion, which is used to determine the concentration or dispersion of user access to content. If the content dispersion is greater than the predetermined scalar, it means that the access content in the area is relatively dispersed, and the system needs to introduce access data from associated areas for further analysis. An associated area refers to an area that has a certain correlation with the target area in terms of geographical location or content demand characteristics. The access feature set of the associated area can provide a reference for the target area and help identify hot content. Associated hot content is content that is frequently accessed and has concentrated user demand in the associated area, and can provide a reference for content recommendations in the target area.

[0024] First, the content dispersion of the target area is detected. If the content dispersion is greater than the predetermined scalar, it means that the content accessed by users in the target area is relatively scattered, and the system needs to introduce the access data of the associated area for supplementary analysis. At this time, the system reads the associated access feature set of the associated area, that is, the user access data of the area, including the content frequently accessed by users and their access frequency. The associated area can be determined based on geographical proximity or user demand similarity. Next, the system performs a concentration analysis on the associated access feature set. In this analysis, the system aggregates the content accessed by users in the associated area and calculates its access frequency, defining high-frequency content as hot content in the associated area. For example, in the access data of the associated area, if "local news" and "popular movies and TV series" are content with high access frequencies, then these contents will be identified as associated hot content.

[0025] The associated hotspot content is then cached to the edge processing unit in the target area to respond to the user's access needs more quickly. For example, when a shopping mall's Wi-Fi network detects that its user access needs are scattered, it can refer to the hotspot content of a neighboring shopping mall (such as promotional information or popular activities for a certain category of goods) for caching. In this way, even if user needs are not concentrated in the target area, the system can improve the user experience by providing hotspot content in the associated area.

[0026] In this step, by referencing the hot content in the associated area when the content is highly discrete, the system achieves dynamic expansion of demand data, and can provide content recommendations that meet user interests even in areas with dispersed demand. This method not only improves system resource utilization, but also improves user access experience, enabling edge processing units to refer to demand trends in neighboring areas for content distribution when user demand is unclear, thereby optimizing content response efficiency.

[0027] Step S400: Utilizing the edge processing unit, performing user access compensation in the target area.

[0028] Specifically, user access compensation means providing content through edge processing units to compensate for network deficiencies and ensure smooth user access experience when user demand peaks or network resources are insufficient in the target area. For example, when the network is congested, edge devices will take on some data requests.

[0029] First, obtain the user traffic density data of the target area through the backbone network or traffic monitoring tool to determine whether the area needs to enable the edge processing unit for access compensation. When user demand is frequent, the system will also dynamically update the hot content in the edge node based on the real-time user traffic density. Specifically, the system sets the update frequency according to the user traffic density. Assuming that the user traffic density is high, it means that the user requests are frequent, and the system will increase the update frequency. For example, refresh the hot content every 10 minutes to ensure that the content of the edge cache is the content that the user is most interested in recently. This can be achieved through automated scripts or algorithm-based update mechanisms. Specific methods such as cache replacement strategies (such as LFU or LRU) determine which content is updated and replaced first.

[0030] By performing user access compensation at the edge processing unit, the system significantly improves the efficiency of content distribution and the response speed of user access, especially during peak user traffic periods. The dynamic update of hot content ensures that users obtain the current high-demand content, effectively improving the user experience while reducing the bandwidth pressure of the backbone network.

[0031] Furthermore, obtaining a predetermined density threshold includes: obtaining a real-time load ratio of the backbone network; correcting an initial density threshold according to the real-time load ratio, and outputting the predetermined density threshold, wherein the predetermined density threshold is negatively correlated with the real-time load ratio.

[0032] Specifically, the real-time load ratio refers to the current utilization rate or load of the network backbone, usually expressed as a percentage of network bandwidth usage. For example, if the backbone network bandwidth is 100 Mbps and 60 Mbps is currently used, the real-time load ratio is 60%. This indicator is used to evaluate the operating status and available resources of the network. The initial density threshold refers to the baseline value of user traffic density set at the beginning of network operation or when there is no real-time monitoring data. This value is usually estimated based on historical data or expected traffic conditions as a starting reference for judging traffic conditions.

[0033] First, it is necessary to obtain the real-time load ratio of the backbone network. Through network monitoring tools, such as network performance monitoring software (such as SolarWinds, Nagios, etc.), the network bandwidth usage is tracked in real time and real-time load data is provided. Subsequently, the initial density threshold will be corrected based on the obtained real-time load ratio. Specifically, when the real-time load ratio is high, the system will dynamically increase the predetermined density threshold to avoid too many users from flooding into the target area under high load conditions, thereby causing network congestion; on the contrary, when the load ratio is low, the predetermined density threshold can be lowered to encourage more users to access. For example, if the initial density threshold is set to 500 users / square kilometer, but the real-time load ratio reaches 80%, the predetermined density threshold is adjusted to 400 users / square kilometer to reduce the load pressure. Conversely, when the network load is low (such as 40%), the predetermined density threshold will be increased to 600 users / square kilometer.

[0034] This step enables the network management system to more accurately adapt to different network load conditions and optimize the management of user traffic by real-time monitoring and dynamic adjustment of the predetermined density threshold. When the network load is high, the system can effectively prevent network overload and ensure the stability of user access; when the load is low, the system can maximize resource utilization and improve user experience. It can not only improve overall network performance, but also significantly reduce the risk of potential network bottlenecks, improve service quality and user satisfaction.

[0035] Furthermore, content discrete evaluation is performed based on the user access feature set, including: reading the user access feature set of the target area in the predetermined time zone, wherein the user access feature includes access content and access frequency; setting step weights based on the access frequency, frequency-marking the access content, and determining the user access content set; performing content discrete evaluation on the user access content set, and outputting the content discreteness.

[0036] Specifically, a scheduled time zone refers to a specific time period set by the system, during which the user access behavior in the target area will be collected and analyzed in a centralized manner. The scheduled time zone can be different time periods such as the morning peak, lunch time, and evening peak, which are used to capture the user behavior characteristics of a specific time period. User access characteristics refer to the description of the user's access behavior to content within a specific time period, including the access content (the specific information that the user browses or clicks) and the access frequency (the number of times or frequency that the user accesses the content). These characteristics reflect the user's interests or needs during a specific time period. Ladder weighting is a hierarchical weighting method that divides user access content into different weight levels according to the access frequency. For example, content with a higher access frequency is given a higher weight, and vice versa, a lower weight is given, so as to more clearly distinguish the popularity of the content. Frequency identification refers to the process of classifying and marking access content according to access frequency, which helps the system determine the hot spots of user access to content sets. Through this identification, the access content can be sorted according to popularity. Content dispersion is an indicator that measures the degree of dispersion of user access content in terms of categories, themes, or types, and is usually obtained by calculating the differences between content. A low dispersion indicates that users access concentrated content, whereas a high dispersion indicates that users have dispersed interests.

[0037] First, read the user access feature set of the target area within the predetermined time zone, and extract user access features such as access content and access frequency. This can be done through data collection tools or log analysis tools (such as Google Analytics or Splunk), which can collect and organize user access data in a specific period of time in the target area.

[0038] Next, the content is graded according to the frequency of user access and the ladder weight is set. For example, if a certain content is accessed frequently in a certain time zone, the system will give it a higher weight to highlight the popularity of the content. Conversely, content with a lower access frequency is given a lower weight. Through this classification, the main hot and unpopular content in the user access content set can be further determined.

[0039] Then, the system will mark the frequency of the content, marking the content with higher access frequency as "high-frequency content" and the content with lower access frequency as "low-frequency content", thus forming a clear set of user access content. In this way, the system can more intuitively understand the user's demand for content at different frequencies.

[0040] Finally, the content set accessed by the user is evaluated for content dispersion and the content dispersion is calculated. This evaluation is usually achieved through statistical dispersion indicators (such as variance). If the dispersion is low, it means that the user accesses more concentrated content, and the system can identify these high-frequency content as hot content.

[0041] For example, during the "late peak" period in the scheduled time zone of an e-commerce website, the system reads the user access feature set and finds that the access frequency of "winter coats" is significantly higher than that of other products, so it assigns a higher weight to this content. This high-frequency content is then marked and the content dispersion is calculated. If the dispersion is low, it means that users have a concentrated demand for "winter coats" during the late peak period, and the system will identify this content as hot content to facilitate subsequent caching and distribution optimization.

[0042] This frequency and discreteness-based evaluation method can accurately identify the content that users need in a specific period of time and determine whether the content is consistent. This method can not only effectively analyze user behavior characteristics, but also quickly locate hot content when the discreteness is low, so as to distribute content more accurately. Ultimately, this method improves the accuracy of content recommendations and the efficiency of system resource utilization, and optimizes the user access experience.

[0043] Furthermore, performing content discrete evaluation on the user access content set and outputting the content discreteness includes: distributing the user access content set in a predetermined three-dimensional space according to preset features, wherein the preset features include access frequency, content category, and timeliness; within the predetermined three-dimensional space, counting the amount of data in a unit grid area as a unit data density, and determining a plurality of unit data densities; performing variance calculation on the plurality of unit data densities, and setting the variance calculation result as the content discreteness.

[0044] Specifically, preset features are analysis dimensions selected by the system when evaluating user access to content, including access frequency (the number of times the content is accessed), content category (the subject or category to which the content belongs, such as technology, news, etc.) and timeliness (the time the content is published or the duration of relevance). These features are used to comprehensively characterize the user's access behavior. The preset three-dimensional space is a multidimensional data analysis model that distributes the user's access content set in a virtual three-dimensional coordinate system according to the three preset features (access frequency, content category and timeliness) to facilitate subsequent data distribution and density analysis. Unit data density indicates the degree of data concentration in each unit grid area in the preset three-dimensional space, that is, the number or frequency of user access to content in the area. It reflects the concentration of user interests in a specific area.

[0045] First, based on the preset features, the user access content set is distributed in a predetermined three-dimensional space. This space uses the three dimensions of access frequency, content category, and timeliness as coordinate axes. For example, the user access content of a news website can be mapped to a three-dimensional space, where the "access frequency" axis represents the number of visits to a certain content, the "content category" axis represents the theme of the content (such as sports, entertainment, technology, etc.), and the "timeliness" axis represents the release time of the content or the duration of relevance.

[0046] Next, divide the three-dimensional space into multiple unit grid areas, and count the amount of data in each unit grid, which is set as the unit data density. For example, a grid area mainly contains "high frequency of access, technology category, relatively recent" content, and the data density is relatively high; another grid contains "low frequency of access, entertainment category, expired content", and the data density is relatively low. This division can be achieved through a spatial grid division algorithm, such as a KD tree or other three-dimensional data segmentation algorithm, which is used to divide and calculate the density of each area in the three-dimensional space.

[0047] Subsequently, the variance of the unit data density of all unit grid areas is calculated to quantify the dispersion of the content. The larger the variance, the greater the difference in access frequency, category, and timeliness between different contents, that is, the content distribution is more dispersed; the smaller the variance, the more concentrated the content accessed by users is in terms of these characteristics.

[0048] For example, in the visit analysis of e-commerce platforms, the visit content of users in a predetermined time zone is distributed into three-dimensional space according to the three characteristics of "visit frequency, product category and launch time". Assuming that the high-frequency visits to the "smartphone" category are concentrated on several newly launched models, while the "home products" category is scattered across multiple subcategories and has a low visit frequency, then the unit data density of the smartphone grid is high, and the density of home products is low. Finally, through the discreteness obtained by variance calculation, the system recognizes that the user's focus during this period is concentrated on smartphones.

[0049] By evaluating the discreteness of the content set accessed by users, the system can determine whether the user's access needs are concentrated. Low discreteness means that the user's focus is concentrated, and the system can prioritize caching or recommending these high-demand contents to improve the user's access experience and network resource utilization; high discreteness indicates that the content demand is dispersed, which helps optimize accurate content recommendation. This method improves the pertinence and efficiency of content distribution in network information management, while reducing the burden on the backbone network.

[0050] Furthermore, a concentration analysis is performed on the user access feature set to determine hot content, including: clustering the user access content set into categories to determine multiple user access category sets; counting the total access frequency of the multiple user access category sets, and arranging them from large to small to obtain an access category sequence, selecting the first N categories of the access category sequence as hot categories, and determining N hot categories, where N is an integer greater than 1; selecting the first M access content with the largest access frequency among the N hot categories as access hotspots to obtain the hot content, where M is an integer greater than 1.

[0051] Specifically, category clustering is a data grouping technology that groups user-accessed content according to similar attributes, such as content themes or types. For example, on a news website, content can be divided into categories such as "sports", "entertainment", and "technology". Clustering helps the system extract the main access categories from the user access feature set, which facilitates subsequent hot content analysis. The access category sequence is the result of arranging the user access category set from high to low according to the access frequency. It can help identify the categories that are most concerned by users and determine the ranking. Hot categories are the top N categories with the highest access frequency, indicating the content categories with the highest user access volume. These categories reflect the main direction of user needs. Access hotspots are the top M specific content with the highest access frequency in the hot category. They can further help the system identify the specific content that users are most concerned about and provide support for content recommendations.

[0052] First, the user's access content set is clustered into categories, and similar content is grouped into one category. This can be achieved through clustering algorithms, such as K-means, DBSCAN, or hierarchical clustering methods. Next, the total access frequency of each user's access category set is counted to generate a sequence of access categories arranged from high to low in terms of access frequency. Through this sorting, the system can identify the categories that users pay the most attention to. The first N categories from this sequence are selected as hot categories (where N is greater than 1), and these categories reflect the main concerns of users.

[0053] After that, in each hot category, the system further analyzes the access frequency ratio and selects the top M most frequently accessed contents as access hotspots (where M is greater than 1). For example, in the "Technology" category, users access "Smartphone Releases" and "AI Technology Progress" the most frequently, so these two contents become the access hotspots of this category, reflecting users' demand for these specific contents.

[0054] This step can help the system accurately identify the user's hot spot needs. By determining the hot spot category and access hot spots, the system can optimize content distribution and caching strategies, improve the response speed of user requests and the accuracy of recommendations. Ultimately, this method optimizes resource utilization while improving user experience.

[0055] Further, such as Figure 2As shown, determining the associated area includes: obtaining multiple sub-areas in the target area, and obtaining the location coordinate sets, area type sets and multiple historical user access feature sets of the multiple sub-areas; randomly selecting a first sub-area from the multiple sub-areas, and obtaining the first location coordinates, the first area type and the first historical user access feature set of the first sub-area; performing type association analysis according to the first area type and other area types in the area type set, and determining multiple type associations; performing access feature association analysis according to the first historical user access feature set and multiple historical user access feature sets, and determining multiple access associations; screening the multiple type associations and multiple access associations, and determining multiple initial associated areas, wherein the sub-areas whose type associations meet the type association threshold and whose access associations meet the access association threshold are set as initial associated areas; screening the multiple initial associated areas according to the location coordinate set, and determining a preset number of first associated areas, wherein the closer to the first sub-area, the greater the selection priority; constructing an area mapping table according to the first sub-area and the first associated area, and inputting the target area into the area mapping table, and matching to determine the associated area.

[0056] Specifically, a sub-region refers to a smaller area divided within a target area. Each sub-region has different characteristics and user behaviors. For example, a city can be divided into multiple sub-regions, such as commercial, residential, and industrial areas. The location coordinate set refers to the geographic coordinate information associated with each sub-region. It is usually expressed in the form of longitude and latitude to describe the specific location of the sub-region. The area type set refers to the classification information of each sub-region, such as commercial, residential, and leisure areas. The area type can reflect the main functions and characteristics of the sub-region and affect the user's access behavior. The historical user access feature set refers to the user's access record to specific content within a certain period of time, including information such as access frequency, access time, and access content. These features can help the system understand the user's access habits. Type association analysis refers to the process of evaluating the relationship between different area types to determine their similarity or correlation. This analysis helps to identify other areas associated with the target area. Access association analysis is to analyze historical user access features to determine the similarities or differences in user access behaviors between different areas, so as to identify areas with strong correlation. The region mapping table is a structured data table used to record the relationship between the target region and its associated regions, which can help the system perform region matching and fast query.

[0057] First, multiple sub-regions within the target area are obtained, and the location coordinate set, region type set, and multiple historical user access feature sets of each sub-region are collected to provide a basis for subsequent analysis. For example, the sub-regions of city A include commercial areas, residential areas, etc., and each area has corresponding coordinates and access features.

[0058] Next, the system randomly selects a first sub-area from the multiple sub-areas and obtains the first location coordinates, the first area type, and the first historical user access feature set of the area. Assuming that the selected first sub-area is a commercial area, the system will extract its latitude and longitude information, type (such as "commercial area"), and user access data.

[0059] The system then performs a type relevance analysis, which compares the first area type with other area types to determine their similarities. For example, if the first sub-area is a "commercial area" and there are multiple "residential areas" and "leisure areas" around it, the system will evaluate the relevance of these areas to the commercial area and determine multiple type relevances that are similar to the target area type.

[0060] At the same time, the first historical user access feature set and all historical user access feature sets are analyzed for access correlation to evaluate the similarity of user access behaviors between them. For example, if users in the first sub-region frequently access certain electronic products, and the access behaviors of users in other regions are similar, then the two regions will be considered to have a high access correlation.

[0061] Next, the system will filter out multiple initial association areas, which are sub-areas whose type association and access association meet the set thresholds. For example, if the type association exceeds 0.8 and the access association exceeds 0.7, these areas will be considered qualified initial association areas.

[0062] Finally, the system screens multiple initial associated regions according to the location coordinate set, and determines that the regions closer to the first sub-region have higher priority, so as to select a preset number of first associated regions. By constructing a region mapping table, the system inputs the target region into the table, performs matching, and determines the final associated region.

[0063] This method effectively determines the associated areas, allowing the system to refer to the user's behavior patterns in similar areas when recommending content and processing data. By aggregating the access characteristics of different areas, the system can more accurately identify user needs and quickly provide content that matches user interests when the target area is short of content, thereby improving user experience and optimizing resource utilization efficiency.

[0064] Furthermore, multiple access correlations are determined, including: randomly selecting a second sub-area and obtaining a second historical user access feature set of the second sub-area; determining multiple historical time periods according to a predetermined time step, clustering the first historical user access feature set and the second historical user access feature set according to the multiple historical time periods, and performing content overlap analysis and traffic density overlap analysis on the multiple clustering results to determine multiple content overlaps and multiple density overlaps; counting the number of clustering results whose content overlap satisfies a predetermined overlap scalar and whose density overlap satisfies a predetermined density overlap scalar, setting it as a second access correlation, and adding it to the multiple access correlations.

[0065] Specifically, the second sub-region is a target region for comparative analysis with the first sub-region, so as to facilitate the calculation of the similarity of access features between different regions. The historical user access feature set refers to the access records of users in a specific time period, including information such as the type of content accessed, frequency and time, which is used to describe the user's access preference. The first historical user access feature set and the second historical user access feature set correspond to the access feature sets of the first sub-region and the second sub-region respectively. The predetermined time step is the time interval for dividing the historical access data into multiple time periods, which is convenient for fine-grained time period analysis. For example, a day is divided into several hours for data sampling to observe the access pattern of each time period. The content overlap indicates the degree of overlap of the content types accessed by users in the first sub-region and the second sub-region in cluster analysis. The traffic density overlap refers to the degree of overlap of the access density of users in different regions in a specific time period, so as to measure the similarity of user access behavior. The overlap scalar is a set value used to determine whether the overlap of content or traffic density meets the standard. Only when the overlap is higher than this scalar, it is considered to have a high correlation.

[0066] First, a second sub-area is randomly selected from multiple sub-areas, and a second historical user access feature set of the area is obtained. Assuming that the first sub-area is a commercial area and the second sub-area is an adjacent residential area, the access feature sets of these areas will contain information such as the specific content of the visit and the frequency of the visit.

[0067] Next, the system will divide the access feature set into multiple time periods according to a predetermined time step (such as every hour or every day) to facilitate more fine-grained comparative analysis. In each time period, the system will perform clustering operations on the first historical user access feature set and the second historical user access feature set. For example, the k-means clustering algorithm can be used to group similarly accessed content into a group, and data from different time periods can be divided into multiple clustering results.

[0068] After obtaining the clustering results, the system performs content overlap analysis and traffic density overlap analysis on each clustering result. In the content overlap analysis, the system compares the content accessed by users in the first and second sub-areas and counts the overlapping access types. For example, if users in both areas frequently access video streaming content, the overlap of the content is high; while in the traffic density overlap analysis, the system compares the density of access in different areas during the same time period. If the access density of the two areas is similar during peak hours, the traffic density overlap is high.

[0069] The system counts the clustering results that meet the conditions, that is, the number of clusters whose content overlap is higher than the predetermined overlap scalar and whose traffic density overlap is higher than the predetermined density overlap scalar, and defines it as the second access relevance. This relevance value is added to multiple access relevance sets to measure the similarity between two sub-areas.

[0070] This step analyzes the overlap of access characteristics and traffic density between different areas, and can identify the similarity of user access behavior between adjacent areas. This method ensures that the system can make decisions based on the similarity of associated areas when distributing content or configuring network resources. For example, when the demand for user access content in a commercial area increases, the content characteristics of the highly associated residential area can be referenced to pre-cache popular content and optimize resource allocation, thereby improving the accuracy of content recommendations, improving user experience, and increasing the overall resource utilization of the system.

[0071] Furthermore, user access compensation is performed in the target area, and then it also includes: setting a content update frequency according to the user traffic density, wherein the content update frequency is positively correlated with the user traffic density; and performing hotspot content update of the edge processing unit based on the content update frequency.

[0072] Specifically, content update frequency refers to the update frequency of cached content in the system. The higher the update frequency, the faster the content is refreshed to ensure that users get the latest hot content. In this solution, the content update frequency is positively correlated with user traffic density, that is, the denser the user traffic, the higher the frequency of content updates. The hot content update of the edge processing unit refers to the system dynamically updating the content with high user demand in the edge node (such as edge server or edge cache device) to ensure the provision of real-time and popular data services in high-traffic areas. For example, the most popular short video or news content is refreshed regularly so that users can get the latest content without accessing the central server.

[0073] According to the user traffic density in the target area, the frequency of content updates is dynamically set to optimize the update strategy of hot content in the edge processing unit. When the system detects an increase in user traffic density, the content update frequency increases accordingly to ensure that users get the latest hot content under high access frequency. Usually with the help of automated scripts and content update algorithms, the system sets different update frequencies based on traffic density thresholds. For example, using LFU (least used) or LFU-K (least used-K frequency) cache replacement algorithms, the content in the edge cache is evaluated and replaced, and content with high access frequency is retained first. These algorithms can effectively filter out content with high user access volume and update it first.

[0074] For example, in a subway station with dense traffic, users frequently access hot videos on a short video platform. When the system detects that the traffic density of the subway station is abnormally high, it will automatically adjust the update frequency of hot videos in the edge nodes to ensure that the latest and most popular video content is cached in this area. If user traffic continues to increase, the edge nodes will further increase the content update frequency to ensure that users can get the latest hot videos at any time.

[0075] By dynamically adjusting the content update frequency according to the user traffic density, the content service efficiency of edge nodes can be significantly improved, especially when users access intensively, users can instantly obtain the latest hot content. This dynamic update mechanism effectively reduces the load pressure of the backbone network and improves the user experience.

[0076] In summary, the network information decision management method based on user traffic statistical analysis provided by the embodiment of the present application has the following technical effects: 1. By regularly reading the user traffic density in the target area, when the density exceeds the predetermined threshold, the edge processing unit is activated to ensure that high-traffic areas are given priority for more efficient content processing support. This dynamic activation mechanism optimizes resource allocation and can cache hot content through edge nodes without increasing the burden on the backbone network, thereby improving user access speed and experience, while reducing the pressure on the central server and realizing distributed management of network information.

[0077] 2. The initial density threshold is adjusted by real-time load ratio, and the predetermined density threshold is dynamically generated, so that network resource allocation can flexibly respond to load changes. When the backbone network load increases, the threshold is automatically lowered, and the edge node is quickly activated to provide services, thereby improving the user response speed in high-density areas without increasing overall latency, and enhancing the system's load adaptability and efficiency.

[0078] 3. By evaluating the content discreteness in three-dimensional space and calculating the content discreteness, the system can intuitively analyze the diversity of user access behaviors. After determining the discreteness through variance calculation, hot content with high concentration can be identified and cached preferentially, improving the access response speed of hot content. At the same time, unnecessary content duplication storage is avoided, and the storage efficiency of the edge processing unit is optimized.

[0079] 4. By dynamically setting the content update frequency according to the user traffic density, the content update speed is increased during the peak user traffic period, so that the content accessed by users always remains real-time. This positive correlation ensures the timeliness of hot content and the availability of content in high-traffic areas, effectively improving the user experience, avoiding access delays caused by untimely content updates, and improving the overall response speed of the system and resource utilization efficiency.

[0080] Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

[0081] Furthermore, the first or second mentioned above not only represents an order relationship, but also represents a specific concept, and / or refers to the selection of multiple elements individually or in whole. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application intends to include these modifications and variations.

Claims

1. A network information decision management method based on user traffic statistical analysis, characterized in that: include: In the backbone network, the user traffic density of the target area is read periodically, and if the user traffic density is greater than a predetermined density threshold, the edge processing unit of the target area is activated, wherein the edge processing unit is communicatively connected to the backbone network; Reading a user access feature set of a target area, performing content discrete evaluation according to the user access feature set, and if the content discreteness is less than or equal to a predetermined scalar, performing a concentration analysis on the user access feature set to determine hot content, and caching the hot content to the edge processing unit; If the content dispersion is greater than the predetermined scalar, reading the associated access feature set of the associated area, performing concentration analysis on the associated access feature set, determining associated hotspot content, and caching the associated hotspot content to the edge processing unit; Using the edge processing unit, user access compensation for a target area is performed.

2. A network information decision management method based on user traffic statistics analysis according to claim 1, characterized in that: Get the predetermined density threshold, including: Get the real-time load ratio of the backbone network; The initial density threshold is corrected according to the real-time load ratio, and the predetermined density threshold is output, wherein the predetermined density threshold is negatively correlated with the real-time load ratio.

3. A network information decision management method based on user traffic statistics analysis according to claim 1, characterized in that: Performing discrete content evaluation according to the user access feature set includes: Reading a user access feature set of a target area within a predetermined time zone, wherein the user access feature includes access content and access frequency; Setting a step weight according to the access frequency, marking the access content by frequency, and determining a user access content set; A content dispersion evaluation is performed on the user access content set, and the content dispersion degree is output.

4. A network information decision management method based on user traffic statistical analysis according to claim 3, characterized in that: Performing content dispersion evaluation on the user access content set and outputting the content dispersion degree comprises: Distributing the user access content set in a predetermined three-dimensional space according to preset characteristics, wherein the preset characteristics include access frequency, content category, and timeliness; In the predetermined three-dimensional space, the amount of data in a unit grid area is counted as a unit data density, and a plurality of unit data densities are determined; Variance calculation is performed on the plurality of unit data densities, and a variance calculation result is set as the content dispersion.

5. A network information decision management method based on user traffic statistical analysis according to claim 3, characterized in that: Performing concentration analysis on the user access feature set to determine hot content includes: Clustering the user access content set into categories to determine a plurality of user access category sets; Counting the total access frequencies of the plurality of user access category sets and arranging them from large to small to obtain an access category sequence, selecting the first N categories of the access category sequence as hotspot categories, and determining N hotspot categories, where N is an integer greater than 1; The first M access contents with the largest access frequency in the N hotspot categories are selected as access hotspots to obtain the hotspot contents, where M is an integer greater than 1.

6. A network information decision management method based on user traffic statistics analysis according to claim 1, characterized in that: Identify areas of relevance, including: Acquire multiple sub-areas within the target area, and acquire location coordinate sets, area type sets, and multiple historical user access feature sets of the multiple sub-areas; Randomly select a first sub-region from the multiple sub-regions, and obtain a first position coordinate, a first region type, and a first historical user access feature set of the first sub-region; Performing type association analysis respectively according to the first area type and other area types in the area type set to determine a plurality of type associations; Performing access feature correlation analysis respectively according to the first historical user access feature set and multiple historical user access feature sets to determine multiple access correlations; Screening the multiple type associations and the multiple access associations to determine multiple initial association areas, wherein sub-areas whose type associations satisfy a type association threshold and whose access associations satisfy an access association threshold are set as initial association areas; Screening the plurality of initial associated regions according to the position coordinate set to determine a preset number of first associated regions, wherein the closer the region is to the first sub-region, the greater the selection priority; A region mapping table is constructed according to the first sub-region and the first associated region, and the target region is input into the region mapping table to match and determine the associated region.

7. A network information decision management method based on user traffic statistics analysis according to claim 6, characterized in that: Determine multiple access relevance, including: Randomly selecting a second sub-area, and obtaining a second historical user access feature set of the second sub-area; Determine multiple historical time periods according to a predetermined time step, cluster the first historical user access feature set and the second historical user access feature set according to the multiple historical time periods, and perform content overlap analysis and traffic density overlap analysis on the multiple clustering results to determine multiple content overlaps and multiple density overlaps; The number of clustering results whose content overlap degree satisfies a predetermined overlap scalar and whose density overlap degree satisfies a predetermined density overlap scalar is counted, set as a second access association degree, and added to the multiple access association degrees.

8. The network information decision management method based on user traffic statistics analysis according to claim 1 is characterized in that: Perform user access compensation in the target area, followed by: Setting a content update frequency according to the user traffic density, wherein the content update frequency is positively correlated with the user traffic density; The hot content update of the edge processing unit is performed based on the content update frequency.

Images