Method, device, electronic device and storage medium for processing user social circle portrait

Abstract

An embodiment of the present invention provides a method, device, electronic device and storage medium for processing a user social circle portrait. The method includes: obtaining user basic data and user behavior data of a target user; inputting the user basic data into a traffic circle model to perform social circle prediction, and obtaining a family-like user group corresponding to the target user, and the family-like user group includes several associated users; dividing the associated users according to the user behavior data, and obtaining a social circle corresponding to the target user, and the social circle includes at least a colleague social circle, a friend social circle, and a family social circle.

Description

Technical Field

[0001] The present invention relates to the technical field of big data analysis, and in particular to a method for processing a user social circle portrait, a device for processing a user social circle portrait, an electronic device, and a computer-readable storage medium. Background Art

[0002] The internet's entry into the big data era has brought about a series of changes and reshapes to businesses and consumer behavior. Perhaps the most significant change is the ability for businesses to predict user needs by analyzing relevant user data. With the in-depth research and application of big data technologies, businesses are increasingly focused on leveraging big data to provide more personalized services that better meet user needs. Consequently, the concept of "user profiles" has emerged. User profiles are a tagged model of user entities abstracted from information such as their social attributes, lifestyle, and consumer behavior. The key to building user profiles is to "label" users. Labels are highly refined, characteristic features derived from analyzing user information. Currently, user profiles are widely used, helping businesses quickly identify precise user groups and their needs, enabling them to provide services tailored to those groups. However, processing user social circle profiles presents challenges such as inaccurate analysis and the high cost of requiring external equipment to collect user data for analysis. Summary of the Invention

[0003] The embodiments of the present invention provide a method, device, electronic device, and computer-readable storage medium for processing a user social circle portrait, so as to solve or partially solve the problems of inaccurate analysis and high cost in the process of analyzing the user social circle portrait.

[0004] An embodiment of the present invention discloses a method for processing a user's social circle portrait, comprising:

[0005] Obtain basic user data and user behavior data of target users;

[0006] Inputting the user basic data into a traffic circle model to perform social circle prediction, and obtaining a family-like user group corresponding to the target user, wherein the family-like user group includes a plurality of associated users;

[0007] The associated users are divided according to the user behavior data to obtain a social circle corresponding to the target user, wherein the social circle includes at least a colleague social circle, a friend social circle, and a family social circle.

[0008] Optionally, the user basic data includes at least package numbers, reserved numbers, complaint and fault reporting information, and communication bill data. Inputting the user basic data into a traffic circle model to perform social circle prediction to obtain a family-like user group corresponding to the target user includes:

[0009] The package numbers, the reserved numbers, the complaint and fault reporting information, and the communication call log data are input into a traffic circle model to perform social circle prediction to obtain a family-like user group corresponding to the target user.

[0010] Optionally, the user behavior data includes communication bill data, first communication data corresponding to social communication performed by the target user on weekdays within a first preset time period, the communication bill data includes communication contacts, and the first communication data includes a base station code of a first base station used by the target user to perform social communication. The dividing the associated users according to the user behavior data to obtain a social circle corresponding to the target user includes:

[0011] According to the usage frequency of the first base station, extract the first base station of the top 1N as the first target base station, and use the base station code of each first target base station as the first working base station data;

[0012] Obtaining second working base station data corresponding to the associated user;

[0013] Inputting the first working base station data and the second working base station data into a bag-of-words model for vector mapping respectively, to obtain a first vector corresponding to the first working base station data and a second vector corresponding to the second working base station data;

[0014] Calculating the similarity between the first vector and the second vector, and taking the associated user corresponding to the second vector whose similarity is greater than or equal to a first preset threshold as the first colleague user of the target user;

[0015] The first colleague user who is successfully matched with the communication contact is formed into a colleague social circle corresponding to the target user.

[0016] Optionally, the user behavior data further includes first application behavior data and a first MAC address and a first IP address of a user terminal to which the target user belongs when executing the social communication. The first colleague user successfully matched with the communication contact is formed into a colleague social circle corresponding to the target user, including:

[0017] Using the first application behavior data, the first MAC address, and the first IP address as the first work behavior data corresponding to the target user;

[0018] Obtaining second work behavior data corresponding to the associated user;

[0019] Inputting the first work behavior data and the second work behavior data into a bag-of-words model for vector mapping respectively, obtaining a third vector corresponding to the first work behavior data and a fourth vector corresponding to the second work behavior data;

[0020] Calculating the similarity between the third vector and the fourth vector, and taking the associated user corresponding to the fourth vector whose similarity is greater than or equal to a second preset threshold as the second colleague user of the target user;

[0021] The first colleague user and / or the second colleague user who are successfully matched with the communication contact form a colleague social circle corresponding to the target user.

[0022] Optionally, the user behavior data includes communication bill data, and second communication data corresponding to social communication performed by the target user on holidays within a second preset time period, the communication bill data includes communication contacts, and the second communication data includes a base station code of a second base station used by the target user to perform social communication. The dividing the associated users according to the user behavior data to obtain a social circle corresponding to the target user includes:

[0023] According to the usage frequency of the second base station, extract the second base station in the top 10 as the second target base station, and use the base station code of each second target base station as the first home base station data;

[0024] Obtaining second home base station data corresponding to the associated user;

[0025] Inputting the first home base station data and the second home base station data into a bag-of-words model for vector mapping, respectively, to obtain a fifth vector corresponding to the first home base station data and a sixth vector corresponding to the second home base station data;

[0026] Calculating the similarity between the fifth vector and the sixth vector, and taking the associated user corresponding to the sixth vector whose similarity is greater than or equal to a third preset threshold as the first relative user of the target user;

[0027] The first relative user who is successfully matched with the communication contact forms the family social circle corresponding to the target user.

[0028] Optionally, the user behavior data further includes second application behavior data and a second MAC address and a second IP address of a user terminal to which the target user belongs when performing the social communication. The first relative user successfully matched with the communication contact is formed into a family social circle corresponding to the target user, including:

[0029] Using the second application behavior data, the second MAC address, and the second IP address as the first entertainment behavior data corresponding to the target user;

[0030] Acquiring second entertainment behavior data corresponding to the associated user;

[0031] Inputting the first entertainment behavior data and the second entertainment behavior data into a bag-of-words model for vector mapping, respectively, to obtain a seventh vector corresponding to the first entertainment behavior data and an eighth vector corresponding to the second entertainment behavior data;

[0032] Calculating the similarity between the seventh vector and the eighth vector, and taking the associated user corresponding to the eighth vector whose similarity is greater than or equal to a fourth preset threshold as the second relative user of the target user;

[0033] The first relative user and / or the second relative user who are successfully matched with the communication contact form the family social circle corresponding to the target user.

[0034] Optionally, the second entertainment behavior data includes target base station data, a third MAC address, and a third IP address corresponding to family users in the family social circle, and the dividing the associated users according to the user behavior data to obtain the social circle corresponding to the target user further includes:

[0035] Obtaining second traffic uplink and downlink communication information, second permanent residence coordinates, and second broadband grid unit coordinates corresponding to associated users in the family social circle;

[0036] Comparing the target base station data, the third MAC address, the third IP address, and the second traffic uplink and downlink communication information corresponding to each family user in the family social circle for overlap, and dividing family users whose overlap is greater than or equal to a fifth preset threshold into the same family group;

[0037] Calculating the communication distance corresponding to each of the family users by using the second permanent residence coordinates corresponding to each of the family users and the second broadband grid unit coordinates;

[0038] Family users whose communication distance is less than or equal to a preset distance threshold and who are in the same family group are formed into a unit family group corresponding to the target user.

[0039] Optionally, it also includes:

[0040] If the user behavior data includes child behavior data related to a child user, inputting the child behavior data into a child user identification model to obtain a child label for the unit family group;

[0041] If the user behavior data includes elderly behavior data related to elderly users, inputting the elderly behavior data into an elderly user identification model to obtain an elderly label for the unit family group;

[0042] The children's behavior data are data related to the daily behavior of children, and the elderly's behavior data are data related to the daily behavior of the elderly.

[0043] Optionally, dividing the associated users according to the user behavior data to obtain a social circle corresponding to the target user includes:

[0044] The associated users except the colleague social circle and the family social circle are used as the friend social circle.

[0045] Optionally, it also includes:

[0046] Acquiring terminal call data of the target user;

[0047] Core users of the social circle are mined according to the terminal call data to obtain core social users in the social circle.

[0048] Optionally, mining core users of the social circle according to the terminal call data to obtain social core users in the social circle includes:

[0049] Extracting from the terminal call data the caller-called relationship between the target user and each associated user in the social circle, the number of calls made by the target user to each associated user in the social circle, and the total number of calls made by the target user to all associated users in the same social circle;

[0050] A Markov matrix is ​​established using the caller-called relationship, and the Markov matrix is ​​iteratively calculated to obtain a weight score corresponding to each associated user in the social circle;

[0051] Calculate the weight coefficient corresponding to each associated user in the social circle using the number of calls and the total number of calls;

[0052] Calculating the result score of each associated user in the social circle using the weight coefficient and the weight score;

[0053] The associated user with the highest result score in each social circle is regarded as the social core user.

[0054] Optionally, the traffic circle model is generated in the following manner:

[0055] Acquire user call history data, wherein the user call history data includes at least calling and calling information of different users, and the calling and calling information includes at least the number of calls and call duration between each pair of calling users;

[0056] A pair of talking users whose number of calls is greater than or equal to a preset number threshold and whose call duration is greater than or equal to a preset duration threshold is considered as active users;

[0057] The mobile phone numbers corresponding to the active users in the integrated package are used as positive samples, and the mobile phone numbers corresponding to users outside the integrated package, who are not the same users and have different IDs, are used as negative samples;

[0058] The positive samples and the negative samples are used as inputs of an input layer of a traffic circle model, and the predicted values ​​output by the input layer are used as feature vectors and input into a logistic regression layer of the traffic circle model for model training;

[0059] The input layer includes random forest algorithm, extremely random tree algorithm and gradient boosting algorithm.

[0060] The embodiment of the present invention further discloses a device for processing a user's social circle portrait, comprising:

[0061] Data acquisition module, used to obtain basic user data and user behavior data of target users;

[0062] A user group determination module is configured to input the user basic data into a traffic circle model to perform social circle prediction, and obtain a family-like user group corresponding to the target user, wherein the family-like user group includes a plurality of associated users;

[0063] The social circle determination module is used to divide the associated users according to the user behavior data to obtain the social circle corresponding to the target user, and the social circle includes at least a colleague social circle, a friend social circle, and a family social circle.

[0064] Optionally, the user basic data includes at least package numbers, reserved numbers, complaint and fault reporting information, and communication bill data, and the user group determination module is specifically used to:

[0065] The package numbers, the reserved numbers, the complaint and fault reporting information, and the communication call log data are input into a traffic circle model to perform social circle prediction to obtain a family-like user group corresponding to the target user.

[0066] Optionally, the user behavior data includes communication bill data, and first communication data corresponding to social communication performed by the target user on weekdays within a first preset time period, the communication bill data includes communication contacts, and the first communication data includes a base station code of a first base station used by the target user to perform social communication. The social circle determination module is specifically configured to:

[0067] According to the usage frequency of the first base station, extract the first base station of the top 1N as the first target base station, and use the base station code of each first target base station as the first working base station data;

[0068] Obtaining second working base station data corresponding to the associated user;

[0069] Inputting the first working base station data and the second working base station data into a bag-of-words model for vector mapping respectively, to obtain a first vector corresponding to the first working base station data and a second vector corresponding to the second working base station data;

[0070] Calculating the similarity between the first vector and the second vector, and taking the associated user corresponding to the second vector whose similarity is greater than or equal to a first preset threshold as the first colleague user of the target user;

[0071] The first colleague user who is successfully matched with the communication contact is formed into a colleague social circle corresponding to the target user.

[0072] Optionally, the user behavior data further includes first application behavior data and a first MAC address and a first IP address of a user terminal to which the target user belongs when performing the social communication. The social circle determination module is further configured to:

[0073] Using the first application behavior data, the first MAC address, and the first IP address as the first work behavior data corresponding to the target user;

[0074] Obtaining second work behavior data corresponding to the associated user;

[0075] Inputting the first work behavior data and the second work behavior data into a bag-of-words model for vector mapping respectively, obtaining a third vector corresponding to the first work behavior data and a fourth vector corresponding to the second work behavior data;

[0076] Calculating the similarity between the third vector and the fourth vector, and taking the associated user corresponding to the fourth vector whose similarity is greater than or equal to a second preset threshold as the second colleague user of the target user;

[0077] The first colleague user and / or the second colleague user who are successfully matched with the communication contact form a colleague social circle corresponding to the target user.

[0078] Optionally, the user behavior data includes communication bill data, and second communication data corresponding to the target user performing social communication on holidays within a second preset time period, the communication bill data includes communication contacts, and the second communication data includes a base station code of a second base station used by the target user to perform social communication. The social circle determination module is specifically configured to:

[0079] According to the usage frequency of the second base station, extract the second base station in the top 10 as the second target base station, and use the base station code of each second target base station as the first home base station data;

[0080] Obtaining second home base station data corresponding to the associated user;

[0081] Inputting the first home base station data and the second home base station data into a bag-of-words model for vector mapping, respectively, to obtain a fifth vector corresponding to the first home base station data and a sixth vector corresponding to the second home base station data;

[0082] Calculating the similarity between the fifth vector and the sixth vector, and taking the associated user corresponding to the sixth vector whose similarity is greater than or equal to a third preset threshold as the first relative user of the target user;

[0083] The first relative user who is successfully matched with the communication contact forms the family social circle corresponding to the target user.

[0084] Optionally, the user behavior data further includes second application behavior data and a second MAC address and a second IP address of a user terminal to which the target user belongs when performing the social communication. The social circle determination module is further configured to:

[0085] Using the second application behavior data, the second MAC address, and the second IP address as the first entertainment behavior data corresponding to the target user;

[0086] Acquiring second entertainment behavior data corresponding to the associated user;

[0087] Inputting the first entertainment behavior data and the second entertainment behavior data into a bag-of-words model for vector mapping, respectively, to obtain a seventh vector corresponding to the first entertainment behavior data and an eighth vector corresponding to the second entertainment behavior data;

[0088] Calculating the similarity between the seventh vector and the eighth vector, and taking the associated user corresponding to the eighth vector whose similarity is greater than or equal to a fourth preset threshold as the second relative user of the target user;

[0089] The first relative user and / or the second relative user who are successfully matched with the communication contact form the family social circle corresponding to the target user.

[0090] Optionally, the second entertainment behavior data includes target base station data, a third MAC address, and a third IP address corresponding to family users in the family social circle, and the social circle determination module is further configured to:

[0091] Obtaining second traffic uplink and downlink communication information, second permanent residence coordinates, and second broadband grid unit coordinates corresponding to associated users in the family social circle;

[0092] Comparing the target base station data, the third MAC address, the third IP address, and the second traffic uplink and downlink communication information corresponding to each family user in the family social circle for overlap, and dividing family users whose overlap is greater than or equal to a fifth preset threshold into the same family group;

[0093] Calculating the communication distance corresponding to each of the family users by using the second permanent residence coordinates corresponding to each of the family users and the second broadband grid unit coordinates;

[0094] Family users whose communication distance is less than or equal to a preset distance threshold and who are in the same family group are formed into a unit family group corresponding to the target user.

[0095] Optionally, it also includes:

[0096] a child label determination module, configured to, if the user behavior data includes child behavior data related to a child user, input the child behavior data into a child user identification model to obtain a child label for the unit family group;

[0097] an elderly label determination module, configured to, if the user behavior data includes elderly behavior data related to elderly users, input the elderly behavior data into an elderly user identification model to obtain an elderly label for the unit family group;

[0098] The children's behavior data are data related to the daily behavior of children, and the elderly's behavior data are data related to the daily behavior of the elderly.

[0099] Optionally, the social circle determination module is specifically configured to:

[0100] The associated users except the colleague social circle and the family social circle are used as the friend social circle.

[0101] Optionally, it also includes:

[0102] A call data acquisition module, configured to acquire terminal call data of the target user;

[0103] The core user mining module is used to perform core user mining on the social circle according to the terminal call data to obtain the social core users in the social circle.

[0104] Optionally, the core user mining module is specifically used to:

[0105] Extracting from the terminal call data the caller-called relationship between the target user and each associated user in the social circle, the number of calls made by the target user to each associated user in the social circle, and the total number of calls made by the target user to all associated users in the same social circle;

[0106] A Markov matrix is ​​established using the caller-called relationship, and the Markov matrix is ​​iteratively calculated to obtain a weight score corresponding to each associated user in the social circle;

[0107] Calculate the weight coefficient corresponding to each associated user in the social circle using the number of calls and the total number of calls;

[0108] Calculating the result score of each associated user in the social circle using the weight coefficient and the weight score;

[0109] The associated user with the highest result score in each social circle is regarded as the social core user.

[0110] Optionally, the traffic circle model is generated by the following modules:

[0111] A call history data acquisition module is used to obtain user call history data, wherein the user call history data includes at least the calling and calling information of different users, and the calling and calling information includes at least the number of calls and the call duration between each pair of calling users;

[0112] An active user determination module, configured to identify a pair of talking users whose number of calls is greater than or equal to a preset number threshold and whose call duration is greater than or equal to a preset duration threshold as active users;

[0113] A sample determination module is configured to use the mobile phone numbers corresponding to the active users in the integrated package as positive samples, and the mobile phone numbers corresponding to users outside the integrated package who are not the same users and do not have the same ID as negative samples;

[0114] A model training module, configured to use the positive samples and the negative samples as inputs of an input layer of a traffic circle model, and to input the predicted values ​​output by the input layer as feature vectors into a logistic regression layer of the traffic circle model for model training;

[0115] The input layer includes random forest algorithm, extremely random tree algorithm and gradient boosting algorithm.

[0116] An embodiment of the present invention further discloses an electronic device, comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0117] The memory is used to store computer programs;

[0118] The processor is configured to implement the method described in the embodiment of the present invention when executing the program stored in the memory.

[0119] An embodiment of the present invention further discloses a computer-readable storage medium having instructions stored thereon. When executed by one or more processors, the processors are enabled to execute the method according to the embodiment of the present invention.

[0120] The embodiments of the present invention include the following advantages:

[0121] In an embodiment of the present invention, it can be applied to big data analysis. By obtaining the user basic data and user behavior data of the target user, the user basic data can be input into a pre-trained call circle model to predict the social circle, and a family-like user group corresponding to the target user can be obtained. The family-like user group includes several associated users corresponding to the target user, and each associated user has at least one social relationship with the target user. Then, the associated users can be divided according to the user behavior data to obtain the social circle corresponding to the target user. The social circle includes at least a colleague social circle, a friend social circle, a family social circle, etc., so that the user's social circle is analyzed through the user basic data to obtain the family-like user group associated with the user. Then, the social relationship of the family-like user group is divided based on the user behavior data to obtain the social circle corresponding to the user. Based on the data analysis method, not only can the accuracy of the analysis be guaranteed, but also the dependence on peripheral equipment can be reduced, thereby reducing costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0122] Figure 1 This is a flowchart of a method for processing a user's social circle portrait provided in an embodiment of the present invention;

[0123] Figure 2 is a schematic diagram of a matrix provided in an embodiment of the present invention;

[0124] Figure 3 is a schematic diagram of a matrix provided in an embodiment of the present invention;

[0125] Figure 4 is a schematic diagram of a matrix provided in an embodiment of the present invention;

[0126] Figure 5 is a schematic diagram of a matrix provided in an embodiment of the present invention;

[0127] Figure 6 is a schematic diagram of a social circle provided in an embodiment of the present invention;

[0128] Figure 7 is a processing flow chart of a social circle provided in an embodiment of the present invention;

[0129] Figure 8 is a schematic diagram of mining a unit household provided in an embodiment of the present invention;

[0130] Figure 9 This is a structural block diagram of a device for processing a user's social circle portrait provided in an embodiment of the present invention;

[0131] Figure 10 This is a block diagram of an electronic device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0132] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0133] It should be noted that the embodiments of this application may involve the collection and use of user data. In actual applications, user-specific personal data can be used in the scheme described herein within the scope permitted by applicable laws and regulations, subject to the requirements of applicable laws and regulations of the country where the user is located (for example, with the user's explicit consent, effective notification to the user, etc.).

[0134] As an example, user profiles are widely used, helping businesses quickly identify precise user groups and their needs, enabling them to provide services tailored to those needs. However, processing user social circle profiles can be inaccurate and require external equipment to collect user data for analysis, leading to high analysis costs.

[0135] In this regard, one of the core invention points of the present invention is to obtain the user basic data and user behavior data of the target user based on big data analysis, and then input the user basic data into a pre-trained call circle model to predict the social circle, and obtain a family-like user group corresponding to the target user. The family-like user group includes several associated users corresponding to the target user, and each associated user has at least one social relationship with the target user. Then, the associated users can be divided according to the user behavior data to obtain the social circle corresponding to the target user. The social circle includes at least a colleague social circle, a friend social circle, a family social circle, etc., so that the user's social circle is analyzed through the user basic data to obtain the family-like user group associated with the user. Then, the social relationship of the family-like user group is divided based on the user behavior data to obtain the social circle corresponding to the user. Based on the data analysis method, not only can the accuracy of the analysis be guaranteed, but also the dependence on peripheral equipment can be reduced, thereby reducing costs.

[0136] Specifically, refer to Figure 1 , which shows a flowchart of a method for processing a user social circle portrait provided in an embodiment of the present invention, which may specifically include the following steps:

[0137] Step 101: Obtain basic user data and user behavior data of the target user;

[0138] For a user, there may be multiple social relationships, such as colleague relationships, friend relationships, family relationships, etc. Among them, family relationships can be further divided into unit family relationships living together and family relationships not living together, etc. The present invention does not limit this.

[0139] Among them, in the process of big data analysis, the user basic data and user behavior data of the target users can be obtained. For the user basic data, it can be data used to characterize the user's basic information. For the user behavior data, it can be the relevant data generated by the user's daily communication through the corresponding mobile phone number, use of the terminal, use of applications, etc., and the information contained is relatively complicated.

[0140] Step 102: Input the user basic data into a traffic circle model to perform social circle prediction to obtain a family-like user group corresponding to the target user, wherein the family-like user group includes a number of associated users.

[0141] In a specific implementation, the basic user data may at least include package numbers, reserved numbers, complaint and fault reporting information, and communication call bill data, etc., among which the package numbers may be the mobile phone numbers corresponding to the service packages provided by the operator; the reserved numbers may be the mobile phone numbers reserved by the users when applying for service packages; the complaint and fault reporting information may be the corresponding information when the users apply for related services; the communication call bill data may be the data collected by the operators in combination with broadband DPI (Deep Packet Inspection) with the user's authorization, etc. After obtaining the above-mentioned basic user data, the package numbers, reserved numbers, complaint and fault reporting information, and call bill data may be input into the traffic circle model for social circle prediction to obtain a family-like user group corresponding to the target user.

[0142] Among them, for the family-like user group, it can include associated users who have at least one social relationship with the target user, such as colleague relationship, friend relationship, relative relationship, etc., and the family-like user group with relevant social relationship with the target user can be screened out through the call circle model and user basic data, so as to make further division based on the family-like user group and obtain the corresponding social circle of the target user.

[0143] In an optional embodiment, the training process of the traffic circle model can be performed by obtaining user call history data, which at least includes the calling and calling information of different users, and the calling and calling information at least includes the number of calls and call duration between each pair of calling users. Then, a pair of calling users whose number of calls is greater than or equal to a preset number threshold and whose call duration is greater than or equal to a preset duration threshold is regarded as active users, and the mobile phone numbers corresponding to active users in the integrated package are used as positive samples, and the mobile phone numbers corresponding to users outside the integrated package, who are not the same user and who do not have the same ID, are used as negative samples. Then, the positive and negative samples are input as the input layer of the traffic circle model, and the predicted value output by the input layer is input as a feature vector into the logistic regression layer of the traffic circle model for model training. The input layer includes a random forest algorithm, an extremely randomized tree algorithm, and a gradient boosting algorithm.

[0144] The user call history data can be historical call data for mobile phone numbers corresponding to different users; the converged package can be a service package provided by the operator corresponding to the mobile phone number; and the ID can be an ID card, residence permit, or other identity document. Furthermore, the traffic circle model can include an input layer, a logistic regression layer, and an output layer, etc., which are not limited in the present invention.

[0145] In one example, after obtaining positive and negative samples, for the training process, a cloud computing-based big data analysis platform can be used to extract processed sample data from a distributed computing cluster using Spark to generate a DataFrame. Due to the imbalance of samples, when training the model, 500,000 samples are randomly selected as training sets, and at the same time, model training is performed based on Pandas and Scikit-learn machine learning lib required for Python big data analysis to obtain a call circle model. Specifically, the Stacking method can be a hierarchical model integration framework. Taking the secondary Stacking as an example, assume that there are three base models M1, M2, and M3. The base model M1 is trained on the training set train, and then used to predict the label columns of train and test, which are P1, T1 (such as Figure 2 As shown). Repeat the same process for M2 and M3, and we will get P2, T2, P3, and T3. Merge P1, P2, P3 and T1, T2, T3 to get a new training set and test set train2, test2 (as shown). Figure 3 、 4 Then use the second layer model M4 to train train2 and predict test2 to get the final label column (as shown). Figure 4 shown).

[0146] The model trained by Stacking can be as described above, but since the model trained with the training set needs to reversely predict the labels of the training set, it is easy to cause serious overfitting. Therefore, the K-fold cross-validation method is used to solve this overfitting problem (for example, 5-fold cross-validation is used), so that the corresponding traffic circle model can be trained through the above method.

[0147] Step 103: Divide the associated users according to the user behavior data to obtain a social circle corresponding to the target user, wherein the social circle includes at least a colleague social circle, a friend social circle, and a family social circle.

[0148] For the target user, since the family-like user group includes different associated users who have at least one social relationship with the target user, it is necessary to classify the associated users belonging to the same social relationship to obtain the social circles of different social relationships corresponding to the target user, such as colleagues' social circle, friends' social circle and family social circle, etc., so as to analyze the user's social circle through the user's basic data to obtain the family-like user group associated with the user, and then divide the social relationships of the family-like user group based on the user's behavior data to obtain the social circle corresponding to the user. The data analysis method can not only ensure the accuracy of the analysis, but also reduce the dependence on peripheral equipment and reduce costs.

[0149] For the colleague social circle, the user behavior data corresponding to the target user may include communication call data, first communication data corresponding to the target user performing social communication on working days within a first preset time period, the communication call data includes communication contacts, and the first communication data includes the base station code of the first base station used by the target user to perform social communication. Then, according to the usage frequency of the first base station, the first base station of TOPN (N can be a positive integer) is extracted as the first target base station, and the base station code of each first target base station is used as the first working base station data, and then the second working base station data corresponding to the associated user is obtained, and then the first working base station data and the second working base station data are respectively input into the word bag model for vector mapping, and the first vector corresponding to the first working base station data and the second vector corresponding to the second working base station data are obtained, and then the similarity between the first vector and the second vector is calculated, and the associated user corresponding to the second vector whose similarity is greater than or equal to the first preset threshold is taken as the first colleague user of the target user, and the first colleague users who are successfully matched with the communication contacts are formed into the colleague social circle corresponding to the target user.

[0150] Among them, for social communication, it can include voice communication, SMS communication, MMS communication and video communication, etc.; the communication contact can be the communication object of the target user when performing social communication; the first base station can be the base station to which the terminal is connected when the target user performs social communication on weekdays. The base station code can be the identifier of the base station, and different base stations can correspond to different identifiers. In an example, assuming that the target user performs related social communications through the terminal during the five working days from Monday to Friday, involving base stations A, B, C and D, and the usage frequencies of each base station are 15 times, 20 times, 30 times and 50 times respectively, then the base station codes corresponding to the TOP3 base stations can be used as the first working base station data corresponding to the target user. Similarly, the second working base station data corresponding to each associated user in the family-like user group can be obtained, and then a similarity comparison can be performed based on the two. According to the comparison results and the possibility that the user communicates with colleagues on weekdays, the first colleague user of the target user can be obtained.

[0151] In one example, the user behavior data of the target user can be counted on a monthly basis, and the base stations frequently used by the user last month can be sorted according to 9-12 o'clock, 14-18 o'clock, etc. on weekdays to obtain the TOP3 base station codes used by the user, and the base station codes corresponding to the base stations with the TOP3 usage frequencies can be regarded as working base station data.

[0152] Secondly, broadband can be used as a carrier to obtain similarity through process processing of base station data of its target users and associated users in the family-like user group, and associated users with similarity greater than or equal to 0.4 are regarded as colleague group A.

[0153] Specifically, to address the issue of a large number of related users and a large number of base station combinations, the user base station codes can be converted into vectors using the natural language processing (NLP) bag of words (BOW) model, and the similarity can be calculated using the Euclidean distance combination. For example:

[0154] import numpy as np

[0155] from sklearn.feature_extraction.text import CountVectorizer

[0156] corpus=[

[0157] "Base station a, base station b, base station c", # of users in the suite 1

[0158] "Base station f, base station a, base station c", #close contacts of user 1 in the suite 1

[0159] "Base station n, base station c, base station b", #close contacts of user 1 in the same suite 2 ]

[0161] vectorizer=CountVectorizer()

[0162] X=vectorizer.fit_transform(corpus)

[0163] vec = X.toarray()

[0164] a=vec[0]

[0165] b=vec[1]

[0166] c=vec[2]

[0167] #Finding the euclidean distance

[0168] dis=np.linalg.norm(ab)

[0169] s=1 / (1+dis)#similarity

[0170] As shown in the example data, by calculating the comparison threshold, close 1 is a colleague of user 1 in the broadband suite.

[0171] Furthermore, the user behavior data also includes first application behavior data and the first MAC address and first IP address of the user terminal to which the target user belongs when performing social communication. The first application behavior data, the first MAC address, and the first IP address can be further used as the first work behavior data corresponding to the target user, and then the second work behavior data corresponding to the associated user is obtained, and the first work behavior data and the second work behavior data are respectively input into the bag-of-words model for vector mapping, and the third vector corresponding to the first work behavior data and the fourth vector corresponding to the second work behavior data are obtained. The similarity between the third vector and the fourth vector is calculated, and the associated user corresponding to the fourth vector whose similarity is greater than or equal to the second preset threshold is used as the second colleague user of the target user, and then the first colleague user and / or the second colleague user who are successfully matched with the communication contact are formed into the colleague social circle corresponding to the target user.

[0172] The first application behavior data may be the behavior data corresponding to office applications used by the user during the workday. The first MAC address and first IP address of the user terminal, the broadband optical modem, and the router may be data collected after the user terminal is connected to the corresponding network with user authorization. Similarly, based on the aforementioned similarity calculation based on base station data, the analysis of the target user's work behavior data during the workday can refer to the description of the relevant process described above and will not be repeated here.

[0173] Specifically, after obtaining the first colleague user corresponding to the target user based on the work base station data, and obtaining the second colleague user corresponding to the target user based on the work behavior data, if only one of them exists, the first colleague user or the second colleague user is divided from the family-like user group as the colleague social circle of the target user. If both exist, the first colleague user and the second colleague user can be used as the colleague social circle of the target user.

[0174] In the above process, based on the similarity, possible colleague users of the target user can be obtained. In order to determine the real colleague users of the target user, the communication contacts included in the call record communication data can be used as the communication contacts, and the possible colleague users who are the same as the communication contacts can be used as the real colleague users of the target user to obtain the corresponding colleague social circle.

[0175] For the family and colleagues circle, the user behavior data corresponding to the target user may also include communication call data, second communication data corresponding to the target user performing social communication during holidays within the second preset time period, the communication call data includes communication contacts, and the second communication data includes the base station code of the second base station used by the target user to perform social communication. Then, according to the usage frequency of the second base station, the second base station of TOPN can be extracted as the second target base station, and the base station code of each second target base station can be used as the first home base station data, and the second home base station data corresponding to the associated user can be obtained. Then, the first home base station data and the second home base station data are respectively input into the word bag model for vector mapping, and the fifth vector corresponding to the first home base station data and the sixth vector corresponding to the second home base station data are obtained. Then, the similarity between the fifth vector and the sixth vector is calculated, and the associated user corresponding to the sixth vector whose similarity is greater than or equal to the third preset threshold is used as the first relative user of the target user, and the first relative users who are successfully matched with the communication contacts form the family social circle corresponding to the target user.

[0176] It should be noted that the difference between work base station data and home base station data is that the former is the data corresponding to the base station that the user terminal corresponding to the user connects to and uses on weekdays, and the latter is the data corresponding to the base station that the user terminal connects to and uses on holidays. For the process of dividing user groups through base station data to obtain family users, you can refer to the aforementioned process of dividing colleague users, which will not be repeated here.

[0177] Furthermore, the user behavior data may also include second application behavior data and the second MAC address and second IP address of the user terminal to which the target user belongs when performing social communication. The second application behavior data, the second MAC address, and the second IP address may be further used as the first entertainment behavior data corresponding to the target user, and the second entertainment behavior data corresponding to the associated user may be obtained. The first entertainment behavior data and the second entertainment behavior data may then be input into the bag-of-words model for vector mapping, respectively, to obtain the seventh vector corresponding to the first entertainment behavior data and the eighth vector corresponding to the second entertainment behavior data. The similarity between the seventh vector and the eighth vector is then calculated, and the associated user corresponding to the eighth vector whose similarity is greater than or equal to the fourth preset threshold is used as the second relative user of the target user. The first relative user and / or the second relative user who are successfully matched with the communication contact may then be used to form the family social circle corresponding to the target user. Similarly, for the division process of the second relative users, reference may be made to the aforementioned division process of the second colleague users, which will not be repeated here.

[0178] In addition, the classification of family users can also be based on the binding relationship between mobile phone numbers and mobile phone terminals. Specifically, when a new SIM (Subscriber Identity Module) card is inserted into the mobile phone terminal, the IMEI (International Mobile Equipment Identity) code of the powered-on terminal will be recorded (International Mobile Equipment Identity, a mark that distinguishes mobile devices, stored in the mobile device, and can be used to monitor stolen or invalid mobile devices), thereby forming a binding relationship between the mobile phone number and the mobile phone terminal. The possible third-party users can be obtained through the terminal's years of use and the circulating mobile phone number information.

[0179] After obtaining the colleague social circle and family social circle through the above process, the associated users in the family-like user group except the colleague social circle and family social circle can be used as the friend social circle, so as to analyze the user's social circle through the user's basic data, and after obtaining the family-like user group associated with the user, the social relationship of the family-like user group is divided based on the user's behavior data to obtain the social circle corresponding to the user. The data analysis method can not only ensure the accuracy of the analysis, but also reduce the dependence on peripheral equipment and reduce costs.

[0180] For the family social circle, the family users involved in the family social circle can be further subdivided to obtain a social circle with finer dimensions. Specifically, the second entertainment behavior data includes the target base station data, the third MAC address, and the third IP address corresponding to the family users in the family social circle. For the family social circle, the second traffic uplink and downlink communication information, the second permanent residence coordinates, and the second broadband grid unit coordinates corresponding to the associated users in the family social circle can be obtained. Then, the target base station data, the third MAC address, the third IP address, and the second traffic uplink and downlink communication information corresponding to each family user in the family social circle can be compared for overlap, and the family users whose overlap is greater than or equal to the fifth preset threshold are divided into the same family group. Then, the second permanent residence coordinates and the second broadband grid unit coordinates corresponding to each family user are used to perform operations to calculate the communication distance corresponding to each family user, and the family users whose communication distance is less than or equal to the preset distance threshold and are in the same family group are formed into a unit family group corresponding to the target user.

[0181] Among them, the permanent residence coordinates can be the longitude and latitude coordinates corresponding to the permanent residence of the target user, etc.; for the broadband grid unit coordinates, the operator can divide the broadband into several different broadband grid units, and the broadband grid unit coordinates can be the longitude and latitude coordinates corresponding to the broadband grid units. Then, during the networking process, the user terminal can obtain the coordinates of the broadband grid unit where it is located when networking, so as to calculate the corresponding communication distance based on the permanent residence coordinates and the broadband grid unit coordinates; the communication distance can be used to characterize the positional relationship between the user terminal and the broadband grid unit. When the distances between two user terminals and the broadband grid units are close, it indicates that the two user terminals can be in basically the same position.

[0182] Specifically, first, data can be collected based on the MAC (each device in the network has a unique network identifier, this address is called a MAC address or network card address, which is written inside the hardware by the network equipment manufacturer when it is produced) of the base station data, broadband optical modems, and routers used by all members of the family social circle. Then, the communication overlap is compared in combination with the upstream and downstream communication conditions. The overlap is used to assign a corresponding member label to each family user, and the family users with the same label are regarded as the same user group. Then, the communication distance can be calculated based on the coordinates of the permanent residence (which can be obtained by analyzing the late-night communication base station where the user has been located for a long time) and the coordinates of the broadband grid unit, and the closest users with the same label are grouped into a unit family group corresponding to the target user. For example, the family users whose communication distance is less than or equal to the preset distance threshold (such as within 2 kilometers) and whose Internet access MAC addresses are often recorded by broadband optical modems and routers are grouped into the unit family group of the target user to indicate that the users in the unit family group are users who live with the target user.

[0183] In addition, based on the customer information in the service package, multiple package situations, etc., families with multiple residences can be found from the family social circle; based on the closeness of the customers in the service package to other families, the families of relatives can be found from the family social circle to obtain the connection between families; and a map can be established by age and gender, that is, a family circle, which is not limited by the present invention.

[0184] In an embodiment of the present invention, after obtaining the social circle corresponding to the target user through the above process, the social core users in each social circle can be further mined based on the circle data corresponding to the social circle (the social core users may be the users who are most closely connected with the target user in the social circle). The terminal call data of the target user can be obtained, and then the core users of the social circle can be mined based on the terminal call data to obtain the social core users in the social circle.

[0185] In a specific implementation, the caller-called relationship between the target user and each associated user in the social circle, the number of calls made by the target user to each associated user in the social circle, and the total number of calls made by the target user to all associated users in the same social circle can be extracted from the terminal call data first. Then, the caller-called relationship is used to establish a Markov matrix, and the Markov matrix is ​​iteratively calculated to obtain the weight score corresponding to each associated user in the social circle. Then, the number of calls and the total number of calls are used to calculate the weight coefficient corresponding to each associated user in the social circle. The weight coefficient and the weight score are used to calculate the result score of each associated user in the social circle. Then, the associated user with the highest result score in each social circle is taken as the social core user.

[0186] Among them, for the caller-called relationship, it can be the target user calling the associated users in the social circle, or the associated users in the social circle calling the target user, etc.; the weight score can be used to characterize the degree of intimacy between the associated users in the social circle and the target user. The larger the score, the more important the associated user is relative to the target user. In addition, for the weight score, since it is related to the number of calls, it needs to be weighted calculated in combination with the corresponding weight coefficient to improve its accuracy.

[0187] In one example, the PageRank algorithm can be used to mine core social users in a social circle. Specifically, a Markov matrix is ​​established based on the caller-caller relationship between users. Then, through continuous iterative calculations, the user weight ranking is obtained. The larger the weight, the more important the user. The calculation formula is: P = (1-d)ee T / n+dB T The core ideas of the PageRank algorithm are as follows:

[0188] (1) The more calls a user makes, the more important the user is;

[0189] (2) If a user is frequently called by an important user, then the user is also important.

[0190] For example, the algorithm implementation in Python can be:

[0191]

[0192]

[0193]

[0194] The above algorithm yields a weighted score for each user. Generally, high scores indicate core social users. However, call frequency factors are factored into the number of calls, which is then weighted using the following formula: Result Score = Weight Score × (1 + Number of Calls Made by the User / Total Number of Calls Made by Users in the Circle). By iterating the algorithm and performing a weighted calculation based on frequency, the most important mobile phone number within the circle is determined. The user with the highest weighted score is considered a core social user. By applying the above algorithm to the social circles of each circle, we can determine the core social users for the corresponding social circles, friend circles, colleague circles, family circles, home circles, and unit family circles.

[0195] In addition, for a unit family group, in addition to determining the core social users through the above method, it is also possible to determine whether the group contains elderly people and / or children. Specifically, if the user behavior data includes child behavior data related to child users, the child behavior data is input into the child user identification model to obtain a child label for the unit family group. The child label can be used to identify that the unit family group contains children; if the user behavior data includes elderly behavior data related to elderly users, the elderly behavior data is input into the elderly user identification model to obtain an elderly label for the unit family group. The elderly label can be used to identify that the unit family group contains elderly people. Among them, child behavior data is data related to children's daily behavior, and elderly behavior data is data related to the daily behavior of the elderly.

[0196] For example, to analyze the family users of a unit family group, 50,000 other members of the smart watch user group are taken as positive samples, and communication behavior data, on-demand animation live broadcast, educational application usage data (including homework, question search, etc.), and campus base station communication data are used as features. A model is obtained by model training using a random forest algorithm, and children's behavior data is predicted based on the model to analyze whether the unit family group contains children. Similarly, to analyze the family users of a unit family group, 50,000 users over 60 years old can be taken as positive samples, and TV watching, mobile phone viewing of news consultation, health and wellness, e-book applications, weekday park and market range base station communication data are used as features. A model is obtained by model training using a random forest algorithm, and elderly behavior data is predicted based on the model to analyze whether the unit family group contains elderly people. The present invention does not impose any restrictions on this.

[0197] After obtaining the target user's corresponding social circles through the above method, you can collect and organize the schedule to form the corresponding social circles at each level. Figure 6, shows a schematic diagram of the social circle provided in an embodiment of the present invention. For a target user, the social circles involved may include a circle of colleagues, a circle of friends, a circle of family, etc. The circle of colleagues may include several colleagues, the circle of friends may include several friends, the circle of family may include several relatives, and the circle of extended family may include a unit family, family members, children, etc. The unit family may be the family circle that is closest to the target user. Therefore, after analyzing the user's social circle through basic user data and obtaining a family-like user group associated with the user, the social relationship of the family-like user group is divided based on the user's behavior data to obtain the social circle corresponding to the user. Based on the data analysis method, not only can the accuracy of the analysis be guaranteed, but also the dependence on peripheral equipment can be reduced, reducing costs, and the social circle can be visualized, presenting the user's social circle in a more intuitive and concise way.

[0198] In addition, the big data middle platform can also establish corresponding tasks and send the social circle data to the local Oracle database for production use to improve the efficiency of targeted marketing and smart operations at all levels. The present invention does not impose any restrictions on this.

[0199] In an embodiment of the present invention, it can be applied to big data analysis. By obtaining the user basic data and user behavior data of the target user, the user basic data can be input into a pre-trained call circle model to predict the social circle, and a family-like user group corresponding to the target user can be obtained. The family-like user group includes several associated users corresponding to the target user, and each associated user has at least one social relationship with the target user. Then, the associated users can be divided according to the user behavior data to obtain the social circle corresponding to the target user. The social circle includes at least a colleague social circle, a friend social circle, a family social circle, etc., so that the user's social circle is analyzed through the user basic data to obtain the family-like user group associated with the user. Then, the social relationship of the family-like user group is divided based on the user behavior data to obtain the social circle corresponding to the user. Based on the data analysis method, not only can the accuracy of the analysis be guaranteed, but also the dependence on peripheral equipment can be reduced, thereby reducing costs.

[0200] In order to help those skilled in the art better understand the technical solutions of the embodiments of the present invention, an example is given below to illustrate:

[0201] Reference Figure 7 , shows a processing flow chart of a social circle provided in an embodiment of the present invention:

[0202] First, we can use basic data such as call behavior data and broadband DPI data for feature processing. We then apply a specific stacking fusion model algorithm (random forest, extremely randomized tree, gradient boosting, and logistic regression) to predict customers and form a call circle. This is then combined with the user's basic data to form a social circle. The social circle output by the call circle model can be supplemented with broadband data to complete the initial social circle construction, resulting in group social circles divided into groups such as colleagues, friends, and family based on relationship events.

[0203] Secondly, for the family circle, we can combine multiple dimensions of data, such as high overlap in travel and home trajectories between users, device circulation, close communication, shared application interests, and device DPI data traces, to identify relatives who have lived together for a long time. This includes:

[0204] 1. Mining the actual residential unit household user situation:

[0205] Reference Figure 8 , showing a schematic diagram of mining a single-family household in an embodiment of the present invention. Using the traffic circle model output, broadband-based social circle information is supplemented to complete the initial social circle construction. Then, by combining multiple dimensions such as high overlap in travel and home trajectories, device transfer, close communication, shared application interests, and device DPI data traces, long-term cohabiting relatives are identified.

[0206] 2. Explore the situation of households with children in the actual residential unit:

[0207] Based on the household data, some existing wristwatch terminal usage is processed; the frequency of educational videos and animation programs played on TV is analyzed; the base station trajectories often go to amusement parks, campuses, etc.; the user terminals contain educational learning software applications for specific groups such as preschool, middle school and primary school students, and other family user characteristic data as positive samples, and an RF (Random Forest) algorithm model is trained to understand the situation of families with children. Through this model, the situation of other families with children can be analyzed.

[0208] 3. Discover the core members of the extended family circle:

[0209] Based on the extended family circle, the PageRank algorithm analyzes the communication stickiness of users in the circle. At the same time, basic information such as user age, financial consumption data, business decision-making behavior and other data are scored and ranked to obtain core user data at the comprehensive application level.

[0210] It should be noted that for the sake of simplicity, the method embodiments are described as a series of actions. However, those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0211] Reference Figure 9 , shows a structural block diagram of a device for processing a user social circle portrait provided in an embodiment of the present invention, which may specifically include the following modules:

[0212] Data acquisition module 901, used to obtain basic user data and user behavior data of target users;

[0213] A user group determination module 902 is configured to input the user basic data into a traffic circle model to perform social circle prediction, and obtain a family-like user group corresponding to the target user, wherein the family-like user group includes a plurality of associated users;

[0214] The social circle determination module 903 is used to divide the associated users according to the user behavior data to obtain the social circle corresponding to the target user, where the social circle includes at least a colleague social circle, a friend social circle, and a family social circle.

[0215] In an optional embodiment, the user basic data includes at least package numbers, reserved numbers, complaint and fault reporting information, and communication bill data, and the user group determination module 902 is specifically configured to:

[0216] The package numbers, the reserved numbers, the complaint and fault reporting information, and the communication call log data are input into a traffic circle model to perform social circle prediction to obtain a family-like user group corresponding to the target user.

[0217] In an optional embodiment, the user behavior data includes communication bill data, first communication data corresponding to the target user performing social communication on weekdays within a first preset time period, the communication bill data includes communication contacts, and the first communication data includes the base station code of the first base station used by the target user to perform social communication. The social circle determination module 903 is specifically configured to:

[0218] According to the usage frequency of the first base station, extract the first base station of the top 1N as the first target base station, and use the base station code of each first target base station as the first working base station data;

[0219] Obtaining second working base station data corresponding to the associated user;

[0220] Inputting the first working base station data and the second working base station data into a bag-of-words model for vector mapping respectively, to obtain a first vector corresponding to the first working base station data and a second vector corresponding to the second working base station data;

[0221] Calculating the similarity between the first vector and the second vector, and taking the associated user corresponding to the second vector whose similarity is greater than or equal to a first preset threshold as the first colleague user of the target user;

[0222] The first colleague user who is successfully matched with the communication contact is formed into a colleague social circle corresponding to the target user.

[0223] In an optional embodiment, the user behavior data further includes first application behavior data and a first MAC address and a first IP address of a user terminal to which the target user belongs when performing the social communication. The social circle determination module 903 is further configured to:

[0224] Using the first application behavior data, the first MAC address, and the first IP address as the first work behavior data corresponding to the target user;

[0225] Obtaining second work behavior data corresponding to the associated user;

[0226] Inputting the first work behavior data and the second work behavior data into a bag-of-words model for vector mapping respectively, obtaining a third vector corresponding to the first work behavior data and a fourth vector corresponding to the second work behavior data;

[0227] Calculating the similarity between the third vector and the fourth vector, and taking the associated user corresponding to the fourth vector whose similarity is greater than or equal to a second preset threshold as the second colleague user of the target user;

[0228] The first colleague user and / or the second colleague user who are successfully matched with the communication contact form a colleague social circle corresponding to the target user.

[0229] In an optional embodiment, the user behavior data includes communication bill data, and second communication data corresponding to the target user performing social communication on holidays within a second preset time period, the communication bill data includes communication contacts, and the second communication data includes the base station code of the second base station used by the target user to perform social communication. The social circle determination module 903 is specifically configured to:

[0230] According to the usage frequency of the second base station, extract the second base station in the top 10 as the second target base station, and use the base station code of each second target base station as the first home base station data;

[0231] Obtaining second home base station data corresponding to the associated user;

[0232] Inputting the first home base station data and the second home base station data into a bag-of-words model for vector mapping, respectively, to obtain a fifth vector corresponding to the first home base station data and a sixth vector corresponding to the second home base station data;

[0233] Calculating the similarity between the fifth vector and the sixth vector, and taking the associated user corresponding to the sixth vector whose similarity is greater than or equal to a third preset threshold as the first relative user of the target user;

[0234] The first relative user who is successfully matched with the communication contact forms the family social circle corresponding to the target user.

[0235] In an optional embodiment, the user behavior data further includes second application behavior data and a second MAC address and a second IP address of a user terminal to which the target user belongs when performing the social communication. The social circle determination module 903 is further configured to:

[0236] Using the second application behavior data, the second MAC address, and the second IP address as the first entertainment behavior data corresponding to the target user;

[0237] Acquiring second entertainment behavior data corresponding to the associated user;

[0238] Inputting the first entertainment behavior data and the second entertainment behavior data into a bag-of-words model for vector mapping, respectively, to obtain a seventh vector corresponding to the first entertainment behavior data and an eighth vector corresponding to the second entertainment behavior data;

[0239] Calculating the similarity between the seventh vector and the eighth vector, and taking the associated user corresponding to the eighth vector whose similarity is greater than or equal to a fourth preset threshold as the second relative user of the target user;

[0240] The first relative user and / or the second relative user who are successfully matched with the communication contact form the family social circle corresponding to the target user.

[0241] In an optional embodiment, the second entertainment behavior data includes target base station data, a third MAC address, and a third IP address corresponding to family users in the family social circle. The social circle determination module 903 is further configured to:

[0242] Obtaining second traffic uplink and downlink communication information, second permanent residence coordinates, and second broadband grid unit coordinates corresponding to associated users in the family social circle;

[0243] Comparing the target base station data, the third MAC address, the third IP address, and the second traffic uplink and downlink communication information corresponding to each family user in the family social circle for overlap, and dividing family users whose overlap is greater than or equal to a fifth preset threshold into the same family group;

[0244] Calculating the communication distance corresponding to each of the family users by using the second permanent residence coordinates corresponding to each of the family users and the second broadband grid unit coordinates;

[0245] Family users whose communication distance is less than or equal to a preset distance threshold and who are in the same family group are formed into a unit family group corresponding to the target user.

[0246] In an optional embodiment, it further includes:

[0247] a child label determination module, configured to, if the user behavior data includes child behavior data related to a child user, input the child behavior data into a child user identification model to obtain a child label for the unit family group;

[0248] an elderly label determination module, configured to, if the user behavior data includes elderly behavior data related to elderly users, input the elderly behavior data into an elderly user identification model to obtain an elderly label for the unit family group;

[0249] The children's behavior data are data related to the daily behavior of children, and the elderly's behavior data are data related to the daily behavior of the elderly.

[0250] In an optional embodiment, the social circle determination module 903 is specifically configured to:

[0251] The associated users except the colleague social circle and the family social circle are used as the friend social circle.

[0252] In an optional embodiment, it further includes:

[0253] A call data acquisition module, configured to acquire terminal call data of the target user;

[0254] The core user mining module is used to perform core user mining on the social circle according to the terminal call data to obtain the social core users in the social circle.

[0255] In an optional embodiment, the core user mining module is specifically used to:

[0256] Extracting from the terminal call data the caller-called relationship between the target user and each associated user in the social circle, the number of calls made by the target user to each associated user in the social circle, and the total number of calls made by the target user to all associated users in the same social circle;

[0257] A Markov matrix is ​​established using the caller-called relationship, and the Markov matrix is ​​iteratively calculated to obtain a weight score corresponding to each associated user in the social circle;

[0258] Calculate the weight coefficient corresponding to each associated user in the social circle using the number of calls and the total number of calls;

[0259] Calculating the result score of each associated user in the social circle using the weight coefficient and the weight score;

[0260] The associated user with the highest result score in each social circle is regarded as the social core user.

[0261] In an optional embodiment, the traffic circle model is generated by the following modules:

[0262] A call history data acquisition module is used to obtain user call history data, wherein the user call history data includes at least the calling and calling information of different users, and the calling and calling information includes at least the number of calls and the call duration between each pair of calling users;

[0263] An active user determination module, configured to identify a pair of talking users whose number of calls is greater than or equal to a preset number threshold and whose call duration is greater than or equal to a preset duration threshold as active users;

[0264] A sample determination module is configured to use the mobile phone numbers corresponding to the active users in the integrated package as positive samples, and the mobile phone numbers corresponding to users outside the integrated package who are not the same users and do not have the same ID as negative samples;

[0265] A model training module, configured to use the positive samples and the negative samples as inputs of an input layer of a traffic circle model, and to input the predicted values ​​output by the input layer as feature vectors into a logistic regression layer of the traffic circle model for model training;

[0266] The input layer includes random forest algorithm, extremely random tree algorithm and gradient boosting algorithm.

[0267] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0268] In addition, an embodiment of the present invention also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and runnable on the processor. When the computer program is executed by the processor, the various processes of the above-mentioned method for processing the user social circle portrait are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0269] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When executed by a processor, the computer program implements the various processes of the above-mentioned method for processing a user social circle portrait, and can achieve the same technical effect. To avoid repetition, the details are not described here. The computer-readable storage medium may be, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0270] Figure 10 A schematic diagram of the hardware structure of an electronic device for implementing various embodiments of the present invention.

[0271] The electronic device 1000 includes but is not limited to components such as a radio frequency unit 1001, a network module 1002, an audio output unit 1003, an input unit 1004, a sensor 1005, a display unit 1006, a user input unit 1007, an interface unit 1008, a memory 1009, a processor 1010, and a power supply 1011. It will be understood by those skilled in the art that the electronic device structure involved in the embodiments of the present invention does not constitute a limitation on the electronic device, and the electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. In the embodiments of the present invention, the electronic device includes but is not limited to a mobile phone, a tablet computer, a laptop computer, a PDA, a vehicle-mounted terminal, a wearable device, and a pedometer.

[0272] It should be understood that in this embodiment of the present invention, the RF unit 1001 can be used to receive and transmit signals during information transmission or calls. Specifically, it receives downlink data from the base station and transmits it to the processor 1010 for processing; in addition, it transmits uplink data to the base station. Typically, the RF unit 1001 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low-noise amplifier, a duplexer, and the like. Furthermore, the RF unit 1001 can communicate with the network and other devices via a wireless communication system.

[0273] The electronic device provides users with wireless broadband Internet access through the network module 1002, such as helping users to send and receive emails, browse web pages, and access streaming media.

[0274] The audio output unit 1003 can convert audio data received by the RF unit 1001 or the network module 1002 or stored in the memory 1009 into an audio signal and output it as sound. In addition, the audio output unit 1003 can also provide audio output related to a specific function performed by the electronic device 1000 (for example, a call signal reception sound, a message reception sound, etc.). The audio output unit 1003 includes a speaker, a buzzer, a receiver, etc.

[0275] The input unit 1004 is used to receive audio or video signals. The input unit 1004 may include a graphics processing unit (GPU) 10041 and a microphone 10042. The graphics processor 10041 processes image data of still pictures or videos obtained by an image capture device (such as a camera) in video capture mode or image capture mode. The processed image frames can be displayed on the display unit 1006. The image frames processed by the graphics processor 10041 can be stored in the memory 1009 (or other storage medium) or transmitted via the radio frequency unit 1001 or the network module 1002. The microphone 10042 can receive sound and process such sound into audio data. In the case of telephone call mode, the processed audio data can be converted into a format that can be sent to a mobile communication base station via the radio frequency unit 1001 for output.

[0276] The electronic device 1000 also includes at least one sensor 1005, such as a light sensor, a motion sensor, and other sensors. Specifically, the light sensor includes an ambient light sensor and a proximity sensor, wherein the ambient light sensor can adjust the brightness of the display panel 10061 according to the brightness of the ambient light, and the proximity sensor can turn off the display panel 10061 and / or the backlight when the electronic device 1000 is moved to the ear. As a type of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in all directions (generally three axes), and can detect the magnitude and direction of gravity when stationary. It can be used to identify the posture of the electronic device (such as horizontal and vertical screen switching, related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tapping), etc.; the sensor 1005 can also include a fingerprint sensor, a pressure sensor, an iris sensor, a molecular sensor, a gyroscope, a barometer, a hygrometer, a thermometer, an infrared sensor, etc., which will not be repeated here.

[0277] The display unit 1006 is used to display information input by the user or information provided to the user. The display unit 1006 may include a display panel 10061, which may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like.

[0278] The user input unit 1007 can be used to receive input digital or character information, and to generate key signal input related to the user settings and function control of the electronic device. Specifically, the user input unit 1007 includes a touch panel 10071 and other input devices 10072. The touch panel 10071, also known as a touch screen, can collect user touch operations on or near it (such as operations performed by the user using any suitable object or accessory such as a finger, stylus, etc. on or near the touch panel 10071). The touch panel 10071 may include two parts: a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch direction, detects the signal caused by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device and converts it into contact point coordinates, which are then sent to the processor 1010, which receives the command sent by the processor 1010 and executes it. In addition, the touch panel 10071 can be implemented using various types such as resistive, capacitive, infrared, and surface acoustic wave. In addition to the touch panel 10071, the user input unit 1007 may also include other input devices 10072. Specifically, the other input devices 10072 may include but are not limited to a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and an operating stick, which will not be described in detail here.

[0279] Furthermore, the touch panel 10071 may be overlaid on the display panel 10061. When the touch panel 10071 detects a touch operation on or near the touch panel 10071, the touch operation is transmitted to the processor 1010 to determine the type of touch event. The processor 1010 then provides a corresponding visual output on the display panel 10061 based on the type of touch event. It will be understood that in one embodiment, the touch panel 10071 and the display panel 10061 are used as two independent components to implement the input and output functions of the electronic device. However, in some embodiments, the touch panel 10071 and the display panel 10061 may be integrated to implement the input and output functions of the electronic device. The specific details are not limited here.

[0280] The interface unit 1008 is an interface for connecting external devices to the electronic device 1000. For example, the external devices may include a wired or wireless headset port, an external power supply (or battery charger) port, a wired or wireless data port, a memory card port, a port for connecting a device with an identification module, an audio input / output (I / O) port, a video I / O port, a headphone port, etc. The interface unit 1008 may be used to receive input (e.g., data information, power, etc.) from the external device and transmit the received input to one or more elements within the electronic device 1000, or may be used to transmit data between the electronic device 1000 and the external device.

[0281] Memory 1009 can be used to store software programs and various data. Memory 1009 may primarily include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (such as a sound playback function or an image playback function); the data storage area may store data generated based on the use of the mobile phone (such as audio data, a phone book, etc.). Furthermore, memory 1009 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0282] The processor 1010 is the control center of the electronic device. It connects the various components of the electronic device using various interfaces and circuits. By running or executing software programs and / or modules stored in the memory 1009 and accessing data stored in the memory 1009, it performs various functions of the electronic device and processes data, thereby monitoring the entire electronic device. The processor 1010 may include one or more processing units; preferably, the processor 1010 may integrate an application processor and a modem processor, wherein the application processor primarily processes the operating system, user interface, and application programs, while the modem processor primarily handles wireless communications. It is understood that the modem processor may not be integrated into the processor 1010.

[0283] The electronic device 1000 may also include a power supply 1011 (such as a battery) to supply power to each component. Preferably, the power supply 1011 may be logically connected to the processor 1010 through a power management system, thereby managing functions such as charging, discharging, and power consumption through the power management system.

[0284] In addition, the electronic device 1000 includes some functional modules not shown, which will not be described here.

[0285] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0286] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0287] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the present invention and the claims, all of which are protected by the present invention.

[0288] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the embodiments of the present invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0289] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0290] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0291] The units described as separate components may or may not be physically separate, and 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 these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0292] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0293] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or 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 for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard drive, ROM, RAM, a magnetic disk, or an optical disk.

[0294] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for processing a user's social circle portrait, characterized in that: include: Obtain basic user data and user behavior data of target users; Inputting the user basic data into a traffic circle model to perform social circle prediction, and obtaining a family-like user group corresponding to the target user, wherein the family-like user group includes a plurality of associated users; Dividing the associated users according to the user behavior data to obtain a social circle corresponding to the target user, wherein the social circle includes at least a colleague social circle, a friend social circle, and a family social circle; The second entertainment behavior data corresponding to the associated user includes the target base station data, the third MAC address, and the third IP address corresponding to the family users in the family social circle. The associated users are divided according to the user behavior data to obtain the social circle corresponding to the target user, including: Obtaining second traffic uplink and downlink communication information, second permanent residence coordinates, and second broadband grid unit coordinates corresponding to associated users in the family social circle; Comparing the target base station data, the third MAC address, the third IP address, and the second traffic uplink and downlink communication information corresponding to each family user in the family social circle for overlap, and dividing family users whose overlap is greater than or equal to a fifth preset threshold into the same family group; Calculating the communication distance corresponding to each of the family users by using the second permanent residence coordinates corresponding to each of the family users and the second broadband grid unit coordinates; Family users whose communication distance is less than or equal to a preset distance threshold and who are in the same family group are formed into a unit family group corresponding to the target user.

2. The method according to claim 1, characterized in that The user basic data includes at least package numbers, reserved numbers, complaint and fault reporting information, and communication bill data. The user basic data is input into the traffic circle model to perform social circle prediction to obtain a family-like user group corresponding to the target user, including: The package numbers, the reserved numbers, the complaint and fault reporting information, and the communication call log data are input into a traffic circle model to perform social circle prediction to obtain a family-like user group corresponding to the target user.

3. The method according to claim 1, characterized in that The user behavior data includes communication bill data and first communication data corresponding to social communication performed by the target user on weekdays within a first preset time period, the communication bill data includes communication contacts, and the first communication data includes a base station code of a first base station used by the target user to perform social communication. The dividing the associated users according to the user behavior data to obtain a social circle corresponding to the target user includes: According to the usage frequency of the first base station, extract the first base station of the top 1N as the first target base station, and use the base station code of each first target base station as the first working base station data; Obtaining second working base station data corresponding to the associated user; Inputting the first working base station data and the second working base station data into a bag-of-words model for vector mapping respectively, to obtain a first vector corresponding to the first working base station data and a second vector corresponding to the second working base station data; Calculating the similarity between the first vector and the second vector, and taking the associated user corresponding to the second vector whose similarity is greater than or equal to a first preset threshold as the first colleague user of the target user; The first colleague user who is successfully matched with the communication contact is formed into a colleague social circle corresponding to the target user.

4. The method according to claim 3, characterized in that The user behavior data also includes first application behavior data and a first MAC address and a first IP address of a user terminal of the target user when executing the social communication. The first colleague user successfully matched with the communication contact is formed into a colleague social circle corresponding to the target user, including: Using the first application behavior data, the first MAC address, and the first IP address as the first work behavior data corresponding to the target user; Obtaining second work behavior data corresponding to the associated user; Inputting the first work behavior data and the second work behavior data into a bag-of-words model for vector mapping respectively, obtaining a third vector corresponding to the first work behavior data and a fourth vector corresponding to the second work behavior data; Calculating the similarity between the third vector and the fourth vector, and taking the associated user corresponding to the fourth vector whose similarity is greater than or equal to a second preset threshold as the second colleague user of the target user; The first colleague user and / or the second colleague user who are successfully matched with the communication contact form a colleague social circle corresponding to the target user.

5. The method according to claim 1, wherein The user behavior data includes communication bill data and second communication data corresponding to the target user performing social communication on holidays within a second preset time period, the communication bill data includes communication contacts, and the second communication data includes a base station code of a second base station used by the target user to perform social communication. The dividing the associated users according to the user behavior data to obtain a social circle corresponding to the target user includes: According to the usage frequency of the second base station, extract the second base station in the top 10 as the second target base station, and use the base station code of each second target base station as the first home base station data; Obtaining second home base station data corresponding to the associated user; Inputting the first home base station data and the second home base station data into a bag-of-words model for vector mapping, respectively, to obtain a fifth vector corresponding to the first home base station data and a sixth vector corresponding to the second home base station data; Calculating the similarity between the fifth vector and the sixth vector, and taking the associated user corresponding to the sixth vector whose similarity is greater than or equal to a third preset threshold as the first relative user of the target user; The first relative user who is successfully matched with the communication contact forms the family social circle corresponding to the target user.

6. The method according to claim 5, characterized in that The user behavior data also includes second application behavior data and a second MAC address and a second IP address of the user terminal to which the target user belongs when performing the social communication. The first relative user successfully matched with the communication contact is formed into a family social circle corresponding to the target user, including: Using the second application behavior data, the second MAC address, and the second IP address as the first entertainment behavior data corresponding to the target user; Acquiring second entertainment behavior data corresponding to the associated user; Inputting the first entertainment behavior data and the second entertainment behavior data into a bag-of-words model for vector mapping, respectively, to obtain a seventh vector corresponding to the first entertainment behavior data and an eighth vector corresponding to the second entertainment behavior data; Calculating the similarity between the seventh vector and the eighth vector, and taking the associated user corresponding to the eighth vector whose similarity is greater than or equal to a fourth preset threshold as the second relative user of the target user; The first relative user and / or the second relative user who are successfully matched with the communication contact form the family social circle corresponding to the target user.

7. The method according to claim 1, characterized in that Also includes: If the user behavior data includes child behavior data related to a child user, inputting the child behavior data into a child user identification model to obtain a child label for the unit family group; If the user behavior data includes elderly behavior data related to elderly users, inputting the elderly behavior data into an elderly user identification model to obtain an elderly label for the unit family group; The children's behavior data are data related to the daily behavior of children, and the elderly's behavior data are data related to the daily behavior of the elderly.

8. The method according to claim 1 or 4 or 5 or 6, characterized in that The dividing the associated users according to the user behavior data to obtain a social circle corresponding to the target user includes: The associated users except the colleague social circle and the family social circle are used as the friend social circle.

9. The method according to claim 1, characterized in that Also includes: Acquiring terminal call data of the target user; Core users of the social circle are mined according to the terminal call data to obtain core social users in the social circle.

10. The method according to claim 9, characterized in that The step of mining core users of the social circle according to the terminal call data to obtain core social users in the social circle includes: Extracting from the terminal call data the caller-called relationship between the target user and each associated user in the social circle, the number of calls made by the target user to each associated user in the social circle, and the total number of calls made by the target user to all associated users in the same social circle; A Markov matrix is ​​established using the caller-called relationship, and the Markov matrix is ​​iteratively calculated to obtain a weight score corresponding to each associated user in the social circle; Calculate the weight coefficient corresponding to each associated user in the social circle using the number of calls and the total number of calls; Calculating the result score of each associated user in the social circle using the weight coefficient and the weight score; The associated user with the highest result score in each social circle is regarded as the social core user.

11. The method according to claim 1 or 2, characterized in that The traffic circle model is generated in the following way: Acquire user call history data, wherein the user call history data includes at least calling and calling information of different users, and the calling and calling information includes at least the number of calls and call duration between each pair of calling users; A pair of talking users whose number of calls is greater than or equal to a preset number threshold and whose call duration is greater than or equal to a preset duration threshold is considered as active users; The mobile phone numbers corresponding to the active users in the integrated package are used as positive samples, and the mobile phone numbers corresponding to users outside the integrated package, who are not the same users and have different IDs, are used as negative samples; The positive samples and the negative samples are used as inputs of an input layer of a traffic circle model, and the predicted values ​​output by the input layer are used as feature vectors and input into a logistic regression layer of the traffic circle model for model training; The input layer includes random forest algorithm, extremely random tree algorithm and gradient boosting algorithm.

12. A device for processing a user's social circle portrait, characterized in that: include: Data acquisition module, used to obtain basic user data and user behavior data of target users; A user group determination module is configured to input the user basic data into a traffic circle model to perform social circle prediction, and obtain a family-like user group corresponding to the target user, wherein the family-like user group includes a plurality of associated users; A social circle determination module, configured to divide the associated users according to the user behavior data to obtain a social circle corresponding to the target user, wherein the social circle includes at least a colleague social circle, a friend social circle, and a family social circle; The second entertainment behavior data corresponding to the associated user includes the target base station data, the third MAC address and the third IP address corresponding to the family users in the family social circle; the social circle determination module is further used to: Obtaining second traffic uplink and downlink communication information, second permanent residence coordinates, and second broadband grid unit coordinates corresponding to associated users in the family social circle; Comparing the target base station data, the third MAC address, the third IP address, and the second traffic uplink and downlink communication information corresponding to each family user in the family social circle for overlap, and dividing family users whose overlap is greater than or equal to a fifth preset threshold into the same family group; Calculating the communication distance corresponding to each of the family users by using the second permanent residence coordinates corresponding to each of the family users and the second broadband grid unit coordinates; Family users whose communication distance is less than or equal to a preset distance threshold and who are in the same family group are formed into a unit family group corresponding to the target user.

13. An electronic device, characterized in that: comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; The memory is used to store computer programs; The processor is configured to implement the method according to any one of claims 1 to 11 when executing a program stored in the memory.

14. A computer-readable storage medium having instructions stored thereon, which, when executed by one or more processors, cause the processors to perform the method according to any one of claims 1 to 11.

Images