Building-based user group relationship portrait analysis method and device, and electronic equipment

Abstract

The application relates to the technical field of wireless communication services, and provides a building-based user group relationship portrait analysis method and device and electronic equipment. The method comprises the following steps: acquiring user group relationship data; performing building coverage feature identification based on the user group relationship data to obtain building resident users; performing information vectorization processing on the user group relationship data based on the building resident users and performing correlation analysis processing based on a space matrix to obtain a user group relationship portrait; the user group relationship portrait comprises a building network correlation relationship of a user; and the user group relationship portrait is analyzed to obtain a building attribute-based user group analysis result. The wired and wireless combined user group relationship portrait technology is simple in data acquisition, can more accurately and stereoscopically position building users, and can comprehensively and accurately identify user group relationships.

Description

Technical Field

[0001] This application relates to the field of wireless communication service technology, specifically to a method, apparatus, and electronic device for analyzing user group relationship profiles based on buildings. Background Technology

[0002] With the widespread application of familiar-based marketing models in social networks, recommendation systems, and advertising, user group relationship identification is becoming increasingly important. User group relationship identification technology can help businesses better understand the relationships between users, understand user needs, analyze market trends, formulate marketing strategies, and provide personalized recommendation services to users.

[0003] Existing user group relationship identification technologies have the following three problems:

[0004] 1. Determining permanent residents in a building based on two-dimensional planar positioning is not accurate enough.

[0005] Existing methods project all user information from the same building onto a two-dimensional plane, and then identify and mine potential relationships between users by analyzing their interactions, behaviors, and attributes. However, this approach fails to accurately pinpoint the floor where a person is located. Since most office buildings are sublet to multiple companies as workspaces, corresponding to different group relationships, two-dimensional location methods struggle to accurately identify user group relationships.

[0006] 2. The data needs to be sourced from a wide range of sources, and some of it contains sensitive personal information, making it difficult to obtain.

[0007] Existing methods rely on cameras for data collection, requiring consent from traffic management departments, shopping malls, office buildings, residential property management companies, and private business owners. Furthermore, due to the large volume of data, the retention period is generally short, making long-term data collection difficult, time-consuming, and labor-intensive. Additionally, the raw mobile CDR (Call Detail Records) data used in existing methods is sensitive personal information and cannot be obtained or used without consent.

[0008] 3. Relying on user business behavior makes it difficult to obtain relationships with the entire user base.

[0009] Existing methods require users to have similar activity patterns and telephone business behavior before they can be analyzed. With the development of instant messaging networks, people are increasingly using instant messaging software to communicate, and the frequency of telephone business behavior has decreased significantly. Therefore, the user group relationship profiles obtained by existing methods are not comprehensive enough. Summary of the Invention

[0010] This application provides a method, apparatus, and electronic device for analyzing user group relationships based on buildings, in order to solve the technical problems of inaccurate positioning, difficulty in data acquisition, and low recognition accuracy in the prior art.

[0011] In a first aspect, embodiments of this application provide a method for analyzing user group relationships based on buildings, comprising: acquiring user group relationship data; identifying building coverage features based on the user group relationship data to obtain resident users of the building; performing information vectorization processing and spatial matrix-based association analysis processing on the user group relationship data based on the resident users of the building to obtain a user group relationship profile; wherein the user group relationship profile includes the user's building network association relationship; and analyzing the user group relationship profile to obtain user group analysis results based on building attributes.

[0012] In one embodiment, user group relationship data includes XDR data and electronic map data; building coverage feature identification is performed based on user group relationship data to obtain building resident users, including: determining the initial location of resident users in a building using the intersection method based on the latitude and longitude information of the building layer of XDR data and electronic map data; filtering the initially located resident users according to building attributes to obtain a partial resident user; extracting building coverage features of resident users based on the partial resident user; and identifying and extracting the initially located resident users based on the building coverage features to obtain the building resident users.

[0013] In one embodiment, the building attributes include office buildings and residential buildings. The preliminary target users are filtered according to the building attributes to obtain a partial number of target users. This includes: when the building attribute is an office building, the preliminary target users are filtered according to the user patterns of office buildings to obtain a partial number of target users that meet the characteristics of office buildings; when the building attribute is a residential building, the preliminary target users are filtered according to the user patterns of residential buildings to obtain a partial number of target users that meet the characteristics of residential buildings.

[0014] In one embodiment, user group relationship data includes XDR data, home broadband WIFI soft probe data, and comprehensive data. Based on the building's resident users, the user group relationship data undergoes information vectorization processing and spatial matrix-based correlation analysis to obtain a user group relationship profile. This includes: obtaining first vector data based on building resident users and XDR data; the first vector data is the WLAN access vector of the XDR user; obtaining second vector data based on comprehensive data and home broadband WIFI soft probe data; the second vector data is the WLAN access vector of the home broadband WIFI soft probe user; and calculating a correlation coefficient matrix based on spatial vectors using the first and second vector data to match the unique MAC address corresponding to each building resident user, thus obtaining the user group relationship profile.

[0015] In one embodiment, the building network association includes the association between each user, building, IMIS number, MAC address, and access WLAN name.

[0016] In one embodiment, the process of calculating a correlation coefficient matrix based on spatial vectors using first vector data and second vector data to match the unique MAC address of each resident user in a building includes: adding the first vector data to the second vector data to form a new spatial matrix; calculating the matrix correlation coefficient of the new spatial matrix and outputting the correlation coefficient matrix; and based on the correlation coefficient matrix, selecting the WLAN access vector of the home broadband WIFI soft probe user that is closest to the WLAN access vector of the XDR user, and matching the IMSI of the selected XDR user with the MAC address of the home broadband WIFI soft probe user.

[0017] In one embodiment, the user group relationship profile is analyzed to obtain user group analysis results based on building attributes, including: performing group relationship mining analysis based on the user group relationship profile to obtain interpersonal relationship analysis results of the user group in the building; and performing communication dimension analysis based on the user group relationship profile to obtain communication status analysis results of the building.

[0018] Secondly, embodiments of this application provide a user group relationship profiling and analysis device based on buildings, comprising: a data acquisition module for acquiring user group relationship data; a feature recognition module for identifying building coverage features based on the user group relationship data to obtain resident users of the building; a relationship association module for performing information vectorization processing and spatial matrix-based association analysis processing on the user group relationship data based on resident users of the building to obtain a user group relationship profile; wherein the user group relationship profile includes the user's building network association relationship; and a profile analysis module for analyzing the user group relationship profile to obtain user group analysis results based on building attributes.

[0019] Thirdly, embodiments of this application provide an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the program to implement the building-based user group relationship profile analysis method of the first aspect.

[0020] Fourthly, embodiments of this application provide a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the building-based user group relationship profiling analysis method of the first aspect.

[0021] The user group relationship profiling method, apparatus, and electronic device provided in this application's embodiments process user group relationship data; identify building coverage features based on the user group relationship data to obtain resident users of the building; perform information vectorization processing and spatial matrix-based correlation analysis on the user group relationship data based on the resident users of the building to obtain a user group relationship profile including the user's building network association relationship; and finally, obtain user group analysis results based on building attributes based on the analysis of the user group relationship profile. Through the above methods, this application's embodiments innovatively propose a wired and wireless joint user group relationship profiling technology, which can locate building users in three dimensions, resulting in more accurate positioning; the user group relationship data used does not involve user privacy and can be obtained in batches, making the data relatively easy to acquire; through joint analysis of wireless network and wired broadband data, it basically covers most of the population, resulting in a more comprehensive and accurate user group relationship profile. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is one of the flowcharts illustrating the user group relationship profiling analysis method based on buildings provided in this application embodiment;

[0024] Figure 2 This is the second flowchart of the user group relationship profiling analysis method based on buildings provided in the embodiments of this application;

[0025] Figure 3 This is a flowchart illustrating the method for identifying resident users based on building coverage features provided in this application embodiment;

[0026] Figure 4 This is a flowchart illustrating the wired-wireless correlation analysis method based on WLAN handover time vector provided in an embodiment of this application.

[0027] Figure 5 This is a schematic diagram of the WLAN access duration calculation method within a time window provided in an embodiment of this application;

[0028] Figure 6 This is a schematic diagram of the vectorization of user WLAN access information provided by the home broadband WIFI soft probe in this application embodiment;

[0029] Figure 7This is a schematic diagram of the WLAN access vectors for XDR users and home broadband WIFI soft probe users provided in the embodiments of this application;

[0030] Figure 8 This is a schematic diagram of the structure of the user group relationship profiling and analysis device based on buildings provided in the embodiments of this application;

[0031] Figure 9 This is a schematic diagram of the physical structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0033] This application innovatively proposes a method, device, electronic device, and storage medium for analyzing user group relationships based on buildings, which can accurately and comprehensively identify different user group relationships and provide precise user care.

[0034] Please see Figure 1 , Figure 1 This is one of the flowcharts illustrating the building-based user group relationship profiling analysis method provided in this application. In this embodiment, the building-based user group relationship profiling analysis method includes steps S110 to S140, each step of which is detailed below:

[0035] S110: Obtain user group relationship data.

[0036] The user group relationship data includes XDR (Extended Detection and Response) data, home broadband Wi-Fi probe data, comprehensive data, and building electronic map data. The purpose of this step is primarily to collect data based on software acquisition and home broadband Wi-Fi probe data. Specifically:

[0037] 1) XDR data

[0038] The XDR data can be software-acquired XDR data. Software-acquired XDR data is data output by the SCA equipment provided by the equipment manufacturer in the IF1 standard format, which mainly includes fields such as time, cell identifier, user identifier, WLAN identifier, latitude and longitude.

[0039] Among them, the MDT sampling data collected by the wireless network base station side includes WLAN information detected by the user terminal.

[0040] Table 1 below shows a sample of soft-sampling XDR data:

[0041]

[0042] Table 1

[0043] 2) Home Wi-Fi soft probe data

[0044] Home broadband WIFI soft probe data refers to the data collected by the home broadband WIFI soft probe, also known as home broadband WIFI soft probe data. Specifically, it includes fields such as time, mobile phone user MAC (Media Access Control) address, and WLAN identifier accessed by the mobile phone.

[0045] Table 2 below shows a sample of home broadband Wi-Fi soft probe data:

[0046]

[0047] Table 2

[0048] 3) Comprehensive data

[0049] Comprehensive data can include a list of relationships between buildings and WLAN coverage. A single building can contain multiple WLAN coverages.

[0050] 4) Building electronic map data

[0051] Electronic map data mainly includes building names and building floor information (building boundary latitude and longitude).

[0052] S120: Based on user group relationship data, identify building coverage features to obtain the building's resident users.

[0053] In one embodiment, the step of identifying building coverage features based on user group relationship data to obtain the resident users of a building specifically includes:

[0054] Based on the latitude and longitude information of building layers in XDR data and electronic map data, the intersection method is used to determine the initial location of users in the building. The initial location of users is then filtered according to building attributes to obtain a partial list of users. Building coverage features of permanent users are extracted based on the partial list of users. Based on the building coverage features, the initial location of users is identified and extracted to obtain the permanent users of the building.

[0055] In one embodiment, building attributes include office buildings and residential buildings. The step of filtering initially identified resident users based on building attributes to obtain a partial list of resident users specifically includes:

[0056] When the building is an office building, the initial target users are filtered according to the user patterns of office buildings to obtain a portion of the target users that match the characteristics of office buildings; when the building is a residential building, the initial target users are filtered according to the user patterns of residential buildings to obtain a portion of the target users that match the characteristics of residential buildings.

[0057] The purpose of this step is to identify resident users based on building coverage characteristics. It mainly uses corresponding rules to filter users based on building attributes and users' general daily routines. First, some user information is collected using XDR data (with latitude and longitude). Then, the intersection method is used to locate the first batch of building users (i.e., initially locate resident users) by combining the building layer information of the electronic map. Through this first batch of building users, information such as the residential area and neighboring areas of the building users are locked. Then, a similarity algorithm is used to lock more resident users in the buildings.

[0058] S130: Based on the data of resident users in the building, information vectorization and spatial matrix-based correlation analysis are performed on the user group relationship data to obtain a user group relationship profile.

[0059] In one embodiment, the steps of performing information vectorization processing and spatial matrix-based correlation analysis on user group relationship data based on building resident users to obtain a user group relationship profile specifically include:

[0060] The first vector data is obtained based on the building's resident users and XDR data; the first vector data is the WLAN access vector of the XDR user; the second vector data is obtained based on the comprehensive data and home broadband WIFI soft probe data; the second vector data is the WLAN access vector of the home broadband WIFI soft probe user; based on the first vector data and the second vector data, a correlation coefficient matrix is ​​calculated based on spatial vectors to match the unique MAC address corresponding to each building's resident user, thus obtaining a user group relationship profile.

[0061] User group relationship profiles include users' building network associations. Optionally, the building network associations can include the association between each user's building, IMIS number, MAC address, and access WLAN name, i.e., "building + IMIS number + MAC address + access WLAN name".

[0062] In one embodiment, the step of calculating the correlation coefficient matrix based on spatial vectors using the first vector data and the second vector data, and matching the unique MAC address corresponding to each resident user in a building, specifically includes:

[0063] Add the first vector data to the second vector data to form a new spatial matrix; calculate the matrix correlation coefficient of the new spatial matrix and output the correlation coefficient matrix; based on the correlation coefficient matrix, select the WLAN access vector of the home broadband WIFI soft probe user that is most similar to the WLAN access vector of the XDR user, and match the IMSI (International Mobile Subscriber Identity) of the selected XDR user with the MAC address of the home broadband WIFI soft probe user.

[0064] The purpose of this step is to perform wired-to-wireless correlation analysis based on WLAN handover time vectors, to filter out the WLAN access vectors (i.e., second vector data) of home broadband Wi-Fi soft probe users that are most similar to the WLAN access vector of the XDR user (i.e., the first vector data), to match the IMSI of the XDR user with the MAC address of the home broadband Wi-Fi soft probe user, and then obtain the corresponding WLAN information of each resident user in the building. Finally, the relationship is determined based on the characteristics of the WLAN access of each resident user in the building, thus completing the correlation analysis of wired and wireless users.

[0065] Specifically, after matching the unique MAC address of each resident user in a building, the data from the WIFI soft probe (a list of relationships between WLAN names and MAC addresses) can be used to obtain the WLAN information accessed by each terminal user (the name of the WLAN accessed, the access time, and the duration of use). Finally, based on the characteristics of each resident user's WLAN access, a group relationship profile can be obtained.

[0066] S140: Analyze the user group relationship profile to obtain user group analysis results based on building attributes.

[0067] In one embodiment, the step of analyzing user group relationship profiles to obtain user group analysis results based on building attributes specifically includes:

[0068] Based on user group relationship profiles, we conduct group relationship mining and analysis to obtain the interpersonal relationship analysis results of the user groups in the building; based on user group relationship profiles, we conduct communication dimension analysis to obtain the communication status analysis results of the building.

[0069] Based on the aforementioned user group relationship profile, the correspondence between each mobile phone user and the name of the Wi-Fi they access can be determined. This information, combined with building attributes, allows for the following analysis: Interpersonal relationship analysis can identify users who have been using the same Wi-Fi network for multiple days, classifying them as colleagues, relatives, or friends, supporting targeted marketing based on personal connections. Communication analysis can statistically determine the ratio of users on the local network to competitors within the WLAN network, enabling the development of targeted marketing and counter-strategies for different buildings. Specifically:

[0070] 1) Group Relationship Mining and Analysis

[0071] Based on the building's characteristics, for office buildings, the statistics are collected during weekdays and normal commuting hours. Users who consistently stay on the same WLAN for multiple consecutive days are identified as colleagues. For residential buildings, the statistics are primarily collected at night. Users who consistently stay on the same WLAN for multiple consecutive days are identified as friends or family. After defining the user relationship profile, targeted marketing can be conducted among acquaintances, such as by offering free data or providing personalized support to encourage customers to promote the service, thereby further expanding the market.

[0072] 2) Market share analysis

[0073] Based on the aforementioned user group relationship profiles, the market share of different operators can be calculated. This allows us to determine which operator a mobile phone user with a matching MAC address and IMSI number within the same WLAN belongs to, thus deriving the market share (numerator being the target operator's users, denominator being all users). For buildings with low market share, targeted marketing can be implemented to retain existing customers while maximizing the number of new customers.

[0074] In summary, this embodiment provides a method for user group relationship profiling analysis based on buildings. It processes user group relationship data, including XDR data, home broadband Wi-Fi probe data, comprehensive data, and building electronic map data. Based on this data, it identifies building coverage features to obtain resident users. Then, based on these resident users, it performs information vectorization and spatial matrix-based correlation analysis on the user group relationship data to obtain a user group relationship profile that includes the user's building network relationships. Finally, it analyzes the user group relationship profile to obtain user group analysis results based on building attributes. Through this approach, this embodiment innovatively proposes a wired and wireless joint user group relationship profiling technology, which can locate building users in three dimensions for greater accuracy. The user group relationship data used does not infringe on user privacy and can be obtained in batches, making the data relatively easy to acquire. By jointly analyzing wireless network and wired broadband data, it basically covers most of the population, resulting in a more comprehensive and accurate user group relationship profile.

[0075] This method employs XDR data (user latitude and longitude information) and building layer information from electronic maps (building boundary latitude and longitude information). First, based on the resident user identification technology using building coverage characteristics, users are mapped to two-dimensional building plane locations. Then, based on WLAN handover time vector, wired and wireless correlation analysis technology is used to determine the WLAN information accessed by each wireless user. Finally, by combining multiple dimensions such as "building location + building attributes + user access WLAN name and time", the "three-dimensional" user group relationship and the operator market share of the corresponding WLAN user group are mined, supporting market-side familiar marketing and precise user care.

[0076] Meanwhile, this method, based on WLAN handover time vector analysis, can associate the MAC address of wireless user equipment with the IMSI of XDR users, thereby analyzing and obtaining the market share of users in the network among buildings or different user groups, and accurately supporting the targeted counter-marketing of buildings with low market share.

[0077] Please see Figure 2 , Figure 2 This is the second flowchart of the user group relationship profiling analysis method based on buildings provided in the embodiments of this application.

[0078] The user group relationship profiling method based on buildings in this embodiment can be roughly divided into four parts:

[0079] 1) Data acquisition based on software acquisition and home broadband WIFI probe;

[0080] 2) Identification of resident users based on building coverage features;

[0081] 3) Wired-to-wireless correlation analysis based on WLAN handover time vector;

[0082] 4) User group relationship profiling and operation analysis.

[0083] The "data collection based on software acquisition and home broadband WIFI probe" and "user group relationship profiling and operation analysis" which have been described in detail in the above embodiments will not be repeated here. The following focuses on the two parts: "identification of resident users based on building coverage characteristics" and "wired and wireless correlation analysis based on WLAN handover time vector".

[0084] Identification of resident users based on building coverage features:

[0085] Please see Figure 3 , Figure 3 This is a flowchart illustrating the resident user identification method based on building coverage features provided in this application embodiment. The implementation process is as follows:

[0086] 1. Building resident resident statistics based on the intersection method

[0087] Based on user-level XDR data and combined with the latitude and longitude information of the building layer on the electronic map, the intersection method is used to initially locate some of the resident users in the building.

[0088] The intersection method is as follows: draw a ray from point 1 through the irregular polygon, and eliminate vertices, boundary points, etc. If the number of intersection points is odd, the polygon is inside; if it is even, the polygon is outside.

[0089] 2. Rooting resident users based on user patterns

[0090] Based on building attributes, buildings are initially categorized into residential buildings and office buildings. Optionally, in other embodiments, building attributes can be expanded to include other dimensions as appropriate, depending on the actual situation.

[0091] Further screening of users residing in buildings to determine whether the building is a residential building or an office building:

[0092] If it is a residential building, the hourly granularity is locked from 0:00 to 5:00 AM. If a user spends 3 hours in the building, it is counted as 1 day of residence. If a user spends more than 15 days in the building in a month, the user is defined as a permanent resident user of the building.

[0093] If it is an office building, then it is divided into weekdays and non-weekdays. If a user spends more than 3 hours in the building between 8-12 pm and 2-6 pm on weekdays, and this happens more than 12 days a month, then the user is defined as a resident user of the building.

[0094] 3. Extract building coverage features of resident users.

[0095] After identifying a subset of resident users in a building using the method described above, information such as the resident community and neighboring areas of these users (including information such as electrical signal strength) can be used to locate them.

[0096] 4. Expanding the sample of resident users based on building coverage characteristics

[0097] After determining the resident community and neighboring community information of the building user, the user is selected if the main service community is consistent and the top 5 neighboring communities have a similarity coefficient of more than 0.8 based on Jaccard (and the power level fluctuation is within a certain range). Then, the user is determined to be a resident user of the building. The method in step 2 above is used to locate more resident users of the building.

[0098] Wired-Wireless Correlation Analysis Based on WLAN Handover Time Vector:

[0099] Data collected by home broadband WIFI soft probes, based on the user level, can only collect the device MAC address of mobile phone users, and cannot collect user identification information such as IMEI and IMSI. This hinders the correlation analysis at the wireless user level. In order to solve the correlation analysis problem between wired and wireless users, this embodiment creatively proposes a method for correlation analysis based on periodic time windows and WLAN information.

[0100] This method first uses the building's resident user information and XDR data (which belongs to wireless data) to locate the strongest WLAN information, and then aggregates it into a structured one-dimensional vector with a five-minute time window granularity.

[0101] Among them, a 288-dimensional vector for 24 hours a day is used. If the WLAN appears in TOP1 in a certain time window, the WLAN access ratio within the time window is calculated by combining the timestamps of the first access and the last disconnection. Otherwise, it is set to 0.

[0102] Then, the correlation between buildings and WLAN coverage is obtained through comprehensive data. Combined with home broadband WIFI soft probe data (which belongs to wired data), the MAC address of each user under WLAN is also aggregated into a structured one-dimensional vector according to a five-minute time window. Finally, the correlation coefficient matrix is ​​calculated based on the spatial vector to match the unique MAC address of each resident user in the building, realizing the association of wired and wireless user-level data collection.

[0103] Optionally, during the matching process, data from multiple consecutive days can be combined, while also considering the principle of the most recent timestamp, to ensure that each user is uniquely matched.

[0104] Please see Figure 4 , Figure 4 This is a flowchart illustrating the wired-to-wireless correlation analysis method based on WLAN handover time vector provided in this application embodiment. The implementation process is as follows:

[0105] 1. Vectorization of XDR user WLAN access information

[0106] After identifying the resident users in the building, based on the consistency of users' WLAN access and disconnection, XDR data is processed into a structured one-dimensional vector according to the time window. This can distinguish the WLAN switching characteristics of each user as much as possible (such as going to the bathroom, picking up a package, going to the tea room, etc., which are basically different for different users), ensuring the uniqueness of subsequent association and matching with home broadband WIFI soft probe users.

[0107] i) Combine XDR data to identify the WLAN information that appears most frequently in the TOP1 frequency for each building's resident users, and use this WLAN information as the user's access WLAN.

[0108] ii) Based on user WLAN access information, data processing is performed at a five-minute time window granularity. If the WLAN appears in the top TOP1 position within a certain time window, the WLAN access ratio within the time window is calculated by combining the timestamps of the first access and the last disconnection. The calculation method is as follows: Figure 5 As shown, Figure 5 This is a schematic diagram of the WLAN access duration calculation method within a time window provided in the embodiments of this application.

[0109] Figure 5 In the above, assuming the time window is denoted as T (five minutes), the WLAN access ratios of user A in time windows T1 and T5 are (T1-t1) / T and (t2-T4) / T, respectively.

[0110] The WLAN access ratios for user B in time windows T1 and T5 are (t4-t3) / T and (t6-t5) / T, respectively.

[0111] iii) Within each five-minute time window, if WLAN access is continuous, the value is set to 1; if access occurs during a partial time period, the WLAN access ratio is calculated (if there are multiple time periods within a time window, these are summed); if there is no WLAN access within the time window, the value is set to 0. The data is aggregated into a structured one-dimensional vector at the granularity of the five-minute time window, such as... Figure 6 As shown, Figure 6 This is a schematic diagram of the vectorization of user WLAN access information provided by the home broadband WIFI soft probe in the embodiments of this application.

[0112] 2. Vectorization of WLAN access information for home broadband Wi-Fi soft probe users

[0113] By obtaining the correlation between building and WLAN coverage through comprehensive data, and combining the home broadband WIFI soft probe data, the MAC address of each user under WLAN is also aggregated into a structured one-dimensional vector according to a five-minute time window.

[0114] 3. Wired-Wireless Correlation Analysis Based on Spatial Matrix

[0115] Correlation analysis is performed on users who are connected to the same WLAN as XDR users and home broadband WIFI soft probe users. First, two spatial matrices are selected to form XDR users and home broadband WIFI soft probe users under the same WLAN. The correlation between the two spatial matrices is calculated and the correlation coefficient matrix is ​​output. Finally, the correlation analysis of wired and wireless user data is realized, and the association relationship of "building + IMIS number + MAC address + accessed WLAN name" for each user is obtained.

[0116] The specific calculation process is as follows:

[0117] a. Use XDR to collect data on resident users in buildings and match them with home broadband WIFI soft probe users. For each XDR user, add its structured vector to the spatial matrix of the home broadband WIFI soft probe to form a new spatial matrix.

[0118] Please see Figure 7 , Figure 7 This is a schematic diagram of the WLAN access vectors for XDR users and home broadband WIFI soft probe users provided in the embodiments of this application.

[0119] b. Based on the newly formed spatial matrix, calculate the matrix correlation coefficient and output the correlation coefficient matrix. The calculation process includes: ① calculating the mean of the spatial matrix, ② centering the spatial matrix, ③ standardizing the spatial matrix, and ④ calculating the correlation coefficient of the spatial matrix. Specifically:

[0120] ①The formula for finding the mean of a spatial matrix is:

[0121] ②The formula for centering a spatial matrix is:

[0122] in

[0123] ③ The formula for standardizing the spatial matrix is:

[0124]

[0125] ④ The formula for calculating the correlation coefficient of a spatial matrix is:

[0126] Where M represents the mean vector formed by calculating the mean of each row of the matrix; N represents the number of columns in the matrix; X1, X2, ... X N B represents the vector corresponding to the columns of the matrix; B represents the centered (normalized) matrix. X represents the vector obtained by subtracting the mean vector M from the column vectors corresponding to the matrix; k The vector representing the Kth column of the matrix; The vector is the result of subtracting the mean vector M from the vector in the Kth column of the matrix; D represents the variance, which is the variance of a column vector; S represents the correlation coefficient matrix.

[0127] c. Based on the above process, output the correlation coefficient matrix, and select the WLAN access vector of the home broadband WIFI soft probe user that is most similar to the WLAN access vector of the XDR user. Match the IMSI of the XDR user with the MAC address of the home broadband WIFI soft probe user, thus completing the correlation analysis of wired and wireless users. The correlation coefficient matrix shown in Table 3 can be obtained as follows:

[0128]

[0129]

[0130] Table 3

[0131] As shown in Table 3, the values ​​in the table represent the similarity between two access vectors; the larger the absolute value, the higher the relative value. Here, A represents the WLAN access vector of the XDR user, and B to I are the WLAN access vectors of the home broadband Wi-Fi soft probe users. The vector with the highest similarity to A is the WLAN access vector I of the home broadband Wi-Fi soft probe user. Associating the IMSI of XDR user A with the MAC address of home broadband Wi-Fi soft probe user I, the output of the wired-wireless association analysis results is shown in Table 4 as an example.

[0132]

[0133]

[0134] Table 4

[0135] In summary, this application innovatively proposes a wired and wireless joint user group relationship profiling technology by introducing data collected from wired home broadband WIFI soft probes. This technology can more comprehensively and accurately identify user group relationships, supporting the market in formulating marketing strategies. Its biggest advantage compared to existing technologies is:

[0136] 1) The newly proposed method can locate users in three dimensions, making it more accurate.

[0137] Compared to existing methods that project all users in the same building onto a two-dimensional plane but fail to accurately locate the floor where a person is located, this application introduces data collected by a wired home broadband WIFI soft probe. Then, it uses the intersection method to locate resident users within the building. Finally, based on the WLAN handover time vector, it performs wired-wireless correlation analysis to determine the WLAN accessed by the wireless user. Each WLAN corresponds to a different floor in the same office building, thus extending the user location from the two-dimensional plane to three-dimensional positioning.

[0138] 2) The newly proposed method involves data that is relatively easy to obtain.

[0139] Compared to existing methods that rely on cameras for data collection and require consent from traffic management departments, shopping malls, office buildings, residential property management companies, and private business owners, which is difficult to achieve, the XDR data, home broadband WIFI soft probe data, comprehensive data, and electronic map data used in this application do not involve user privacy and can be obtained in batches.

[0140] 3) The newly proposed method identifies user relationships more comprehensively and accurately.

[0141] Compared to existing methods for identifying user group relationships, which rely heavily on telephone communication between users (and which have significantly decreased as people increasingly depend on instant messaging software), this application utilizes a combined analysis of ubiquitous wireless and wired broadband data. Excluding a small number of elderly people and young children who do not use mobile devices, it essentially covers the majority of the population, resulting in a more comprehensive and accurate understanding of user group relationships.

[0142] The following describes the user group relationship profiling and analysis device based on buildings provided in the embodiments of this application. The user group relationship profiling and analysis device based on buildings described below can be referred to in correspondence with the user group relationship profiling and analysis method based on buildings described above.

[0143] Please see Figure 8 , Figure 8 This is a schematic diagram of the structure of the building-based user group relationship profiling and analysis device provided in this application embodiment. In this embodiment, the building-based user group relationship profiling and analysis device includes:

[0144] The data acquisition module 810 is used to acquire user group relationship data, which includes XDR data, home broadband WIFI soft probe data, comprehensive data and building electronic map data.

[0145] The feature recognition module 820 is used to identify building coverage features based on user group relationship data to obtain the building's resident users.

[0146] The relationship association module 830 is used to perform information vectorization processing and spatial matrix-based association analysis processing on user group relationship data based on building resident users to obtain user group relationship profiles; the user group relationship profiles include the users' building network association relationships.

[0147] The profile analysis module 840 is used to analyze user group relationship profiles and obtain user group analysis results based on building attributes.

[0148] In one embodiment, the feature recognition module 820 is specifically used to: determine the initial location of resident users of a building based on the latitude and longitude information of the building layer of XDR data and electronic map data using the intersection method; filter the initially located resident users according to the building attributes to obtain a partial resident user; extract the building coverage features of permanent resident users based on the partial resident users; and identify and extract the initially located resident users based on the building coverage features to obtain the permanent resident users of the building.

[0149] In one embodiment, the building attributes include office buildings and residential buildings. The feature recognition module 820 is specifically used to: when the building attribute is an office building, filter the initially located resident users according to the user patterns of office buildings to obtain a portion of resident users that meet the characteristics of office buildings; when the building attribute is a residential building, filter the initially located resident users according to the user patterns of residential buildings to obtain a portion of resident users that meet the characteristics of residential buildings.

[0150] In one embodiment, the relationship association module 830 is specifically used to: obtain first vector data based on building resident users and XDR data; the first vector data is the WLAN access vector of the XDR user; obtain second vector data based on comprehensive data and home broadband WIFI soft probe data; the second vector data is the WLAN access vector of the home broadband WIFI soft probe user; calculate the correlation coefficient matrix based on spatial vectors according to the first vector data and the second vector data, match the unique MAC address corresponding to each building resident user, and obtain a user group relationship profile.

[0151] In one embodiment, the building network association includes the association between each user's building, IMIS number, MAC address, and access WLAN name.

[0152] In one embodiment, the relationship association module 830 is specifically used to: add the first vector data to the second vector data to form a new spatial matrix; calculate the matrix correlation coefficient of the new spatial matrix and output the correlation coefficient matrix; based on the correlation coefficient matrix, select the home broadband WIFI soft probe user WLAN access vector that is most similar to the XDR user WLAN access vector, and match the selected XDR user's IMSI with the home broadband WIFI soft probe user's MAC address.

[0153] In one embodiment, the profile analysis module 840 is specifically used to: perform group relationship mining analysis based on the user group relationship profile to obtain the interpersonal relationship analysis results of the user group in the building; and perform communication dimension analysis based on the user group relationship profile to obtain the communication status analysis results of the building.

[0154] On the other hand, embodiments of this application also provide an electronic device. Figure 9 This is a schematic diagram of the physical structure of the electronic device provided in the embodiments of this application, such as... Figure 9 As shown, the electronic device may include: a memory 920, a processor 910, and a computer program stored in the memory 920 and executable on the processor 910. When the processor 910 executes the program, it implements the building-based user group relationship profiling analysis method provided by the methods described above.

[0155] Optionally, the electronic device may further include a communication bus 930 and a communication interface 940, wherein the processor 910, the communication interface 940, and the memory 920 communicate with each other via the communication bus 930. The processor 910 can call a computer program in the memory 920 to execute a building-based user group relationship profiling analysis method, which includes:

[0156] The process involves acquiring user group relationship data, including XDR data, home broadband Wi-Fi soft probe data, comprehensive data, and building electronic map data; identifying building coverage features based on the user group relationship data to obtain resident users in each building; performing information vectorization processing and spatial matrix-based correlation analysis on the user group relationship data based on resident users to obtain user group relationship profiles; these profiles include users' building network relationships; and analyzing these profiles yields user group analysis results based on building attributes.

[0157] Furthermore, the logical instructions in the aforementioned memory 920 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0158] On the other hand, this application also provides a non-transitory computer-readable storage medium storing a computer program thereon. When the computer program is executed by a processor, it is implemented to perform the building-based user group relationship profiling analysis method provided by the above methods. The steps and principles of the method have been described in detail in the above methods and will not be repeated here.

[0159] Non-transitory computer-readable storage media can be any available medium or data storage device that can be accessed by a processor, including but not limited to magnetic storage (e.g., floppy disks, hard disks, magnetic tapes, magneto-optical disks (MOs), etc.), optical storage (e.g., CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (e.g., ROMs, EPROMs, EEPROMs, non-volatile memory (NAND flash), solid-state drives (SSDs)).

[0160] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0161] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0162] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for analyzing user group relationships based on buildings, characterized in that, include: Obtain user group relationship data; The user group relationship data mentioned above includes Scalable Threat Detection and Response (XDR) data, home broadband Wi-Fi soft probe data, and comprehensive data. Based on the user group relationship data, building coverage features are identified to obtain the building's resident users; The first vector data is obtained based on the building's resident users and the XDR data; the first vector data is the WLAN access vector of the XDR user. The second vector data is obtained based on the comprehensive data and the home broadband WIFI soft probe data; the second vector data is the home broadband WIFI soft probe user WLAN access vector. The first vector data is added to the second vector data to form a new spatial matrix; Calculate the matrix correlation coefficient of the new spatial matrix and output the correlation coefficient matrix; Based on the correlation coefficient matrix, the WLAN access vectors of home broadband WIFI soft probe users that are closest to the WLAN access vectors of XDR users are selected, and the IMSI of the selected XDR users is matched with the MAC addresses of the home broadband WIFI soft probe users to obtain a user group relationship profile; wherein the user group relationship profile includes the building network association relationship of the users; The user group relationship profile is analyzed to obtain user group analysis results based on building attributes.

2. The user group relationship profiling analysis method based on buildings according to claim 1, characterized in that, The user group relationship data includes XDR data and electronic map data; The process of identifying building coverage features based on the user group relationship data to obtain resident users of the building includes: Based on the latitude and longitude information of the building layers in the XDR data and the electronic map data, the intersection method is used to determine the preliminary location of the users residing in the building; Based on building attributes, the initially identified resident users were filtered to obtain a partial list of resident users; Based on the aforementioned resident users, extract the building coverage features of resident users; Based on the building coverage features, the initially located resident users are identified and extracted to obtain the building's permanent resident users.

3. The user group relationship profiling analysis method based on buildings according to claim 2, characterized in that, The building attributes include office buildings and residential buildings. The preliminary selection of resident users based on building attributes yields a subset of resident users, including: When the building attribute is an office building, the initially located resident users are filtered according to the user patterns of office buildings to obtain a portion of resident users that meet the characteristics of office buildings; When the building is a residential building, the initially located users are filtered according to the user patterns of the residential building to obtain a portion of the users that meet the characteristics of the residential building.

4. The user group relationship profiling analysis method based on buildings according to claim 1, characterized in that, The building network association includes the association between each user's building, IMIS number, MAC address, and access WLAN name.

5. The user group relationship profiling analysis method based on buildings according to claim 1, characterized in that, The analysis of the user group relationship profile to obtain user group analysis results based on building attributes includes: Based on the user group relationship profile, group relationship mining and analysis are performed to obtain the interpersonal relationship analysis results of the user group in the building; Based on the user group relationship profile, a communication dimension analysis is performed to obtain the communication status analysis results of the building.

6. A user group relationship profiling and analysis device based on buildings, characterized in that, include: The data acquisition module is used to acquire user group relationship data; The user group relationship data mentioned above includes Scalable Threat Detection and Response (XDR) data, home broadband Wi-Fi soft probe data, and comprehensive data. The feature recognition module is used to identify building coverage features based on the user group relationship data to obtain the building's resident users; The relationship association module is used to obtain first vector data based on the building's resident users and the XDR data; the first vector data is the WLAN access vector of the XDR user; to obtain second vector data based on the comprehensive data and the home broadband WIFI soft probe data; the second vector data is the WLAN access vector of the home broadband WIFI soft probe user; to add the first vector data to the second vector data to form a new spatial matrix; to calculate the matrix correlation coefficient of the new spatial matrix and output the correlation coefficient matrix; based on the correlation coefficient matrix, to select the home broadband WIFI soft probe user WLAN access vector that is closest to the XDR user's WLAN access vector, and to match the IMSI of the selected XDR user with the MAC address of the home broadband WIFI soft probe user to obtain a user group relationship profile; wherein the user group relationship profile includes the user's building network association relationship; The user profile analysis module is used to analyze the user group relationship profile and obtain user group analysis results based on building attributes.

7. An electronic device comprising a processor and a memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the building-based user group relationship profiling analysis method according to any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the building-based user group relationship profiling analysis method as described in any one of claims 1 to 5.

Images