Dual-card slot user identification method and system based on terminal on-off signaling flow

Abstract

The embodiment of the application discloses a double-card-slot user identification method and system based on terminal on-off signaling process, identifies double-card-slot users through on-off signaling, and obtains on-off weight characteristic data; combines the obtained on-off weight characteristics with other suspected double-card-slot service characteristics, and corrects double-card users through a weight clustering model in a data mining manner to determine double-card-slot user information of the terminal; determines an analysis result of user service according to the double-card-slot user information, carries out operation analysis according to the analysis result of the user service, and develops application services for the user. The accuracy and timeliness of double-card-slot user identification are improved, so that more effective marketing strategies can be formulated.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of communication operator dual card slot user identification, and particularly relates to a dual card slot user identification method and system based on terminal on-off signaling flow. BACKGROUND

[0002] Operator analysis of dual card users is crucial for improving marketing effectiveness. Understanding user characteristics and insights into customer habits can help operators better target dual card users using mobile terminals for marketing activities. Implementation of marketing activities for dual card users of mobile terminals can improve customer loyalty, retain existing customers and increase their business usage, thereby helping operators increase business revenue. Dual card users in the stock market can be classified in different ways, such as one machine and multiple cards from the terminal perspective, and multiple cards from the operator perspective. The dual card user identification model is established through relevant analysis of dual card users. The model can help operators better understand the characteristics and habits of dual card users, and thus develop more effective marketing strategies.

[0003] Existing mobile terminal dual card slot user identification methods generally include the following two methods.

[0004] Detecting by installing SDK or APP and other terminal side software technology means. On the terminal side, through the SDK application integrated by the mobile operating system or the specific APP installed outside the system, the hardware detection interface API of the mobile operating system is called, such as detecting the terminal heartbeat by installing the SDK on the terminal, judging whether the user is a dual card slot user based on the terminal heartbeat data, detecting whether the current terminal is a dual card slot terminal and whether a mobile card and the imsi information of the card are inserted in each card slot based on the above similar means, and reporting to the operator background service in the background regularly.

[0005] On the operator service side, the dual card slot information on the user's terminal is analyzed by big data mining means. Here, the big data mining means mainly refers to that the identification of dual card slot dual card customers is mainly based on the customer level information, terminal use, mobile trajectory of mobile phone signaling, and call behavior data (including average daily call times, average call duration, average call duration each time, and call duration distribution at each time period) mastered by the operator to determine the possibility of dual card slot dual card use by the customer. Among them, the identity information mainly considers the relevance of the boarding information when the same customer handles the number. Terminal use reflects the use of dual network dual standby terminal and multiple numbers on the same terminal. Call characteristics reflect the convergence of the same customer in call habits. The mobile phone signaling mobile trajectory feature reflects the convergence of the same customer in location movement.

[0006] In the prior technical solution, the factor that plays the most important role in data mining analysis is: according to common sense, the signaling activity track of the two mobile phone cards of the double-card-slot double-card user should be overlapped. However, because there are differences in the sizes of the base station coverage range and the inconsistent triggering time of the two cards, the activity track of the user may be different, so the difference is generally eliminated by trying statistical analysis modeling of different time granularity and different location granularity. In the prior technical solution, most specific implementations are basically through static A and B card user maximum residence base station comparison analysis per time period and dynamic user residence analysis based on an improved clustering method, cross verification, and finally through sending a "dummy short message" to detect the residence position of the A and B card users at the same time to improve the accuracy of the model.

[0007] In the above prior art, the scene of analyzing the double-card-slot user identification method has the following disadvantages.

[0008] If it is based on the terminal side actively providing the determination, such as needing the terminal side to install a specific APP software for detection, the user may not agree to install, resulting in poor data integrity, or needing the terminal to actively report the scheme, the terminal actively reports the phone card information on the two card slots to the operator, and the operator deploys a server to aggregate data. However, the terminal active reporting scheme has the following problems: the scheme needs to be modified for the terminal, and cannot cover all double-card-slot terminals in the existing network, especially inventory terminals and terminals that are not in the operator's own channel; in order to reduce terminal processing consumption and system processing load, a periodic reporting method is generally used, so the network cannot grasp the state of the double-card-slot terminal double card in real time, and the timeliness is poor. Based on the installation of an APP on the mobile phone to detect double-card users, but this method is easy to cause user resistance and resistance, and affects the user experience.

[0009] On the operator service side, analysis is performed through big data data mining means, which is based on the clustering of mobile phone base station positioning track of the operator big data, but this method has poor timeliness and accuracy, and cannot meet the needs of actual application. For example, it cannot determine that the user carries two mobile phones and installs a mobile card on each mobile phone, and misjudgment is easy to occur. SUMMARY

[0010] In order to improve the accuracy and timeliness of double-card-slot user identification, and thus develop more effective marketing strategies, embodiments of the present application provide a double-card-slot user identification method and system based on terminal on-off signaling flow.

[0011] In a first aspect, the present application provides a double-card-slot user identification method based on terminal on-off signaling flow, which comprises:

[0012] obtain signaling information of the terminal, extract first data in the signaling information and service identification of the terminal and the user; the first data includes the power-on and power-off signaling flow of the MME interface and / or N1, N2 interface of the terminal;

[0013] filter the first data according to the type of the power-on and power-off signaling flow and a preset rule; analyze the filtered first data based on the service identification of the terminal and the user, and obtain a dual-card slot analysis result of the terminal; determine a dual-card slot weight feature of the terminal according to the dual-card slot analysis result;

[0014] obtain a dual-card slot service feature of the terminal, and correct the dual-card slot weight feature according to the dual-card slot service feature through a weight clustering model to determine dual-card slot user information of the terminal.

[0015] In another aspect, an embodiment of the present application provides a dual-card slot user identification system based on terminal power-on and power-off signaling flow, which is used to execute the above method, and the system comprises:

[0016] a signaling extraction module, which is used to obtain signaling information of the terminal, extract first data in the signaling information and service identification of the terminal and the user; the first data includes the power-on and power-off signaling flow of the MME interface and / or N1, N2 interface of the terminal;

[0017] an identification service module, which filters the first data according to the type of the power-on and power-off signaling flow and a preset rule; analyzes the filtered first data based on the service identification of the terminal and the user, and obtains a dual-card slot analysis result of the terminal; and determines a dual-card slot weight feature of the terminal according to the dual-card slot analysis result

[0018] a correction service module, which is used to obtain a dual-card slot service feature of the terminal, and correct the dual-card slot weight feature according to the dual-card slot service feature through a weight clustering model to determine dual-card slot user information of the terminal.

[0019] In another aspect, an embodiment of the present application provides an electronic device, which comprises a memory and a processor, and the memory stores a computer program capable of running on the processor; the processor implements the above evaluation method for algorithm implementation in chip design when executing the computer program.

[0020] In another aspect, an embodiment of the present application provides a computer readable medium having non-volatile program code executable by a processor, and the program code causes the processor to execute the above evaluation method for algorithm implementation in chip design.

[0021] Compared with the prior art, the application provides a double-card slot user identification method and system based on terminal on-off signaling process, which identifies double-card slot users through on-off signaling and obtains on-off weight characteristic data; the obtained on-off weight characteristics are combined with other suspected double-card slot service characteristics, and then the double-card users are corrected through a weight clustering model in a data mining manner to determine double-card slot user information of the terminal; the analysis result of user service is determined according to the double-card slot user information, business analysis is performed according to the analysis result of user service, and application services are developed for the users. The accuracy and timeliness of double-card slot user identification are improved, and therefore more effective marketing strategies can be developed. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 It is a flowchart of the double-card slot user identification method based on terminal on-off signaling process.

[0023] Figure 2 It is a flowchart of obtaining on-off weight characteristic data by identifying double-card slot users based on signaling.

[0024] Figure 3 It is a schematic diagram of a 4G networking topology.

[0025] Figure 4 It is a schematic diagram of a 5G networking topology.

[0026] Figure 5 It is a 4G network initial attachment signaling process.

[0027] Figure 6 It is a flowchart of 4G network detachment signaling.

[0028] Figure 7 It is a flowchart of 5G registration signaling.

[0029] Figure 8 It is a flowchart of 5G network detachment signaling.

[0030] Figure 9 It is a flowchart of correcting double-card users through a weight clustering model.

[0031] Figure 10 It is a system architecture diagram of the double-card slot user identification system based on terminal on-off signaling process. DETAILED DESCRIPTION

[0032] The terms "first", "second", and the like in the description and in the claims of the present application and above-described drawings are used to distinguish similar objects and are not necessarily used to describe a particular sequential or chronological order. It should be understood that terms so used are interchangeable under appropriate circumstances and are merely employed in the descriptions of embodiments of the present application for purposes of the description. Additionally, the terms "comprising", "having", "including", and "containing" and any variations thereof are intended to cover a non-exclusive inclusion such that processes, methods, systems, products, or apparatuses that comprise, have, include, or contain a list of elements are not required to only those elements but can include other elements not expressly listed or inherent to such processes, methods, systems, products, or apparatuses.

[0033] In order to improve the accuracy and timeliness of the dual card slot user identification, so as to formulate more effective marketing strategies, the embodiments of the present application provide a dual card slot user identification method and system based on terminal on-off signaling flow. The dual card slot user is identified by collecting terminal simultaneous on-off signaling records and auxiliary weighting big data mining, aiming to correct the actual number of terminals and users in the existing network, and solve the problems of identification accuracy and timeliness in the prior art.

[0034] The traditional big data mining form may have the following problems in identifying dual card slot users.

[0035] Firstly, it cannot be determined whether the user carries two mobile phones at the same time and inserts dual cards, although the user static maximum resident base station comparison analysis and dual card signaling track dimension clustering analysis are performed, but even if the running track is the same, the big data mining may still misjudge, because it cannot be ruled out that the user may carry two mobile terminals at the same time and insert corresponding mobile cards.

[0036] Secondly, for a small number of users, if only the mobile phone signaling on-off information is used to judge whether the single mobile phone dual card slot inserts dual cards, misjudgment may also occur, for example, two people have a meeting at the same time, in order to avoid disturbance, the mobile phones are simultaneously put into flight mode, and then the flight mode is canceled at the same time. For the base station server, it will also consider that in the same short time window, the two cards belong to simultaneous on-off behavior, and misjudgment occurs.

[0037] Therefore, pure mobile phone on-off signaling detection still needs to be combined with the form of back-end big data mining for analysis, so as to obtain more accurate analysis.

[0038] In summary, the pure big data mining form cannot accurately determine whether the user's mobile phone is inserted with dual cards in the way of statistical or clustering of dual card slot users, and the pure on-off detection of signaling in a short time window to determine whether the dual cards are simultaneously turned on and off also has misjudgment and omissions.

[0039] Therefore, the core technical process of the application is to combine two means to enhance the identification of dual card slot users.

[0040] The idea of the core technical process includes determining the feature of the dual cards existing in the same base station (in a close distance range) and being turned on and off at the same time (in a short time window) based on the signaling recognition of the mobile phone turning on and off, and further screening in combination with the data mining form of weight clustering to obtain more accurate results. The main process of the core technical process in the embodiment of the application is as follows.

[0041] Identifying the dual card slot users based on the signaling and obtaining the on-off dual card slot weight feature data;

[0042] Combining the on-off dual card slot weight feature data obtained in the above process with other suspected dual card slot business features (such as identity registration, mobile trajectory, and base station residence statistics), and identifying the dual card users through the data mining method of weight clustering.

[0043] Figure 1 The flowchart of the dual card slot user identification method based on the terminal on-off signaling process is shown in the figure, and the method is implemented as follows.

[0044] 10: Obtain the signaling information of the terminal, extract the on-off signaling process of the MME interface and / or N1, N2 interface of the terminal in the signaling information (first data), and the business identification of the terminal and the user. In the embodiment of the application, the business identification of the terminal and the user includes the IMSI number and IMEI number, IMEI_TAC number, and card slot identification.

[0045] 20: Filter the on-off signaling process of the MME interface and / or N1, N2 interface of the terminal according to the preset condition and the preset rule on the first data; analyze the filtered first data based on the business identification of the IMSI number and IMEI number, IMEI_TAC number, and card slot identification of the terminal and the user, obtain the dual card slot analysis result of the terminal, and determine the dual card slot weight feature of the terminal according to the dual card slot analysis result.

[0046] In one possible embodiment, the preset condition is that the interval time of the two consecutive first data is less than 2 seconds or less than a preset specified time, which represents a suspected dual card slot terminal, and the on-off behavior occurs at the same time or the flight mode is entered or canceled at the same time.

[0047] Step 10 and step 20 are the specific process of identifying the dual card slot users based on the signaling to obtain the on-off weight feature data, Figure 2 The flowchart of identifying the dual card slot users based on the signaling to obtain the on-off weight feature data is shown in the figure, and the specific implementation is as follows.

[0048] Figure 3 is a schematic diagram of 4G network topology, as shown in the figure, the 4G network mainly includes GPRS, UMTS, E-UTRAN and CDMA transceiver base station and base station corresponding base station controller; these base station controllers will access the evolved packet system (EPS) data, wherein the evolved packet system mainly includes mobility management entity (MME), home subscriber server, policy and charging control function, service gateway and packet data network gateway, etc., the evolved packet system will access the operator service network, the Internet and enterprise network, when accessing the 4G network, the signaling information of the on-off of MME will be obtained.

[0049] Figure 4 is a schematic diagram of 5G network topology, as shown in the figure, the 5G network mainly includes user equipment (N1), (wireless) access network (N2), user plane function, data network, access and mobile management function, NEW access and mobile management function, OLD access and mobile management function, policy control function, session management function, authentication server function device and unified data management device, etc. When accessing the 5G network, the signaling information of the on-off of N1 and N2 interface will be obtained.

[0050] Figure 5 is the initial attachment signaling flow of 4G network, as shown in the figure, through the steps in the figure, the user equipment first enters the connected state from the idle state, then sends a request to the mobility management entity, then the mobility management entity responds to the request and transmits data to the terminal, and finally obtains the on-off signaling flow of the MME interface of the terminal and the service identification of the terminal and the user including IMSI number and IMEI number, IMEI_TAC number and card slot identification.

[0051] Figure 6 is a schematic diagram of 4G network detachment signaling flow, as shown in the figure, after obtaining the above service data, the user equipment requests the service gateway through the steps in the figure, and the service gateway responds to the request to execute the detachment signaling flow.

[0052] Figure 7 is a schematic diagram of 5G registration signaling flow, as shown in the figure, through the steps in the figure, the user equipment requests the unified data management device, and the unified data management device responds to the request to obtain the on-off signaling flow of the N1 and N2 interfaces of the terminal, and the service identification of the terminal and the user including IMSI number and IMEI number, IMEI_TAC number and card slot identification.

[0053] Figure 8is a flowchart of 5G network detachment signaling, as shown, after obtaining the above business data, the user equipment requests the unified data management equipment, and the unified data management equipment responds to the request to execute the detachment signaling flow through the steps in the figure.

[0054] The signaling information of the terminal is obtained, the signaling information is extracted and filtered, and the dual card slot analysis result of the terminal is obtained. The signaling flow of a terminal during power on and off is as follows:

[0055] If the terminal uses a 4G network, the procedure_type=6 (Deattach) flow occurs at the MME interface when the terminal is turned off, and the procedure_type=1 (Attach) flow occurs when the terminal is turned on.

[0056] If the terminal uses a 5G network, the procedure_type=2 (Deregistration) flow occurs at the N1, N2 interface when the terminal is turned off, and the procedure_type=1 (Registration) flow occurs when the terminal is turned on. There are three cases here:

[0057] In one possible embodiment, at the MME interface, assuming mobile phone numbers A and B; when the mobile phone is turned off, procedure_type=6, and when the mobile phone is turned on, procedure_type=1, IMEI_TAC is the same, the corresponding IMSI and IMEI in the two data are different, and the terminal model corresponding to IMEI_TAC is a dual card slot terminal. When the start time and end time of the two data services differ by 2 seconds or within a set rule number of seconds, the dual card slot analysis result of terminals A and B is a dual card slot terminal using a 4G network. The system assigns a high weight value to the characteristic value of the simultaneous on and off of the corresponding dual card data. The weight value can be adjusted according to the size of the time difference. The lower the time difference, the greater the weight value.

[0058] In one possible embodiment, at the N1, N2 interface, assuming mobile phone numbers A and B; when the mobile phone is turned off, procedure_type=2, and when the mobile phone is turned on, procedure_type=1, IMEI_TAC is the same, the corresponding IMSI and IMEI in the two data are different, and the terminal model corresponding to IMEI_TAC is a dual card slot terminal. When the start time and end time of the two data services differ by 2 seconds or within a set rule number of seconds, the dual card slot analysis result of terminals A and B is a dual card slot terminal using a 5G network. The system assigns a high weight value to the characteristic value of the simultaneous on and off of the corresponding dual card data. The weight value can be adjusted according to the size of the time difference. The lower the time difference, the greater the weight value.

[0059] In one possible embodiment, assuming that mobile phone numbers A and B, when the mobile phone is turned off, the procedure_type=6 in the MME interface, the procedure_type=2 in the N1 and N2 interfaces, when the mobile phone is turned on, the procedure_type=1 in the MME and N1 and N2 interfaces, and the IMEI_TAC is the same, A and B are suspected to be the same user with double cards, the corresponding IMSI and IMEI in the two pieces of data are different, and when the start time and the end time of the two pieces of data are within 2 seconds, the double card slot analysis result of the A and B terminals is that one card slot uses a 4G network and the other card slot uses a 5G network, the system gives a high weight value to the characteristic value of the corresponding double card data when the mobile phone is turned on and turned off at the same time, and the weight value can be adjusted according to the size of the time difference, that is, the lower the time difference, the greater the weight value.

[0060] In the embodiment of the application, the signaling record when the same terminal is turned on and turned off at the same time (or enters or cancels the flight mode) is mainly used, and the record can accurately provide characteristic data highly suspected to be a double card user.

[0061] Through the above means, we can obtain the situation of a highly suspected double card slot user, and through the acquisition of characteristic information, it can be judged whether the two cards belong to the user. The possibility of the user turning on and turning off the mobile phone at the same time in a very small time window, and the above data is also used for mining and analyzing the weight clustering data in step 30, which is an important basis for weight characteristic information judgment.

[0062] 30: Obtain the double card slot service characteristics of the terminal, and correct the double card slot weight characteristics of the terminal according to the above double card slot service characteristics through a weight clustering model to determine the double card slot user information of the terminal.

[0063] In the embodiment of the application, the weight characteristics of the two cards with a high probability of turning on and turning off at the same time obtained in the previous step are combined with other judgment characteristics, and data mining is performed based on a weight clustering model to identify and analyze the double card behavior information of the double card slot.

[0064] In the clustering and mining process of traditional big data user information, the clustering algorithm of the weight clustering model can use some information such as identity information dimension characteristics, call dimension characteristics, and signaling trajectory dimension characteristics for correlation mining and analysis.

[0065] The identity information dimension characteristics are registration information of the user, including the correlation characteristics of the user using a double-network double standby terminal and multiple numbers of the same terminal.

[0066] The call dimension feature is a convergence feature of the user in a voice call conversation habit, and includes an average daily call times, an average call duration, an average call duration each time, and a correlation feature of a call duration distribution of each time period.

[0067] The signaling trajectory dimension feature is a convergence feature of a terminal of the user in position movement, and includes a signaling activity trajectory of a dual card slot user, a same time staying position, and a correlation feature of a static maximum staying base station comparison analysis dimension of the user terminal.

[0068] In the embodiment of the application, the collection mode of the static maximum staying base station comparison analysis dimension of the user is that one day is divided into 8 time periods according to every 3 hours, and the maximum staying base stations of the A card user and the B card user in each time period are counted respectively; whether the maximum staying base station positions of the A card user and the B card user in each time period are less than a threshold value is marked, and it is considered that the positions of the A card user and the B card user in the time period overlap; and if the number of overlapping time periods in a month is greater than a threshold value, the A card and the B card are circled as a suspected same user.

[0069] However, the above-mentioned means all need to be combined and analyzed, and cannot provide strong evidence to show that the user is definitely a dual card slot mobile phone and simultaneously installs dual cards.

[0070] For example, the calculation of the identity feature dimension may exist the possibility that the card handled by the user is used by a relative, the call feature data cannot be obtained enough to analyze because the voice call scene is less and less, and the dual card user often has only a single card for calling. And based on the analysis of the signaling trajectory feature, whether it is the trajectory movement overlap degree or the static maximum staying and the proportion comparison, if the user simultaneously installs dual cards by two mobile phones, it is also impossible to determine whether the user is a single mobile phone dual card or a double mobile phone dual card.

[0071] Therefore, if the embodiment of the application simply performs data weight clustering mining, it may cause misjudgment. Therefore, the above-mentioned multiple identity features, call features, signaling trajectory features, and the like are combined together for mining to improve the probability possibility of determining the user as a single mobile phone dual card slot dual card behavior.

[0072] Here, the embodiment of the application selects a clustering method based on a weighted feature weight. The core idea of the weighted clustering method is to initialize a weight value for each feature dimension. When the target function converges, the weight corresponding to the noise dimension will tend to zero, so that the influence of the noise dimension on the calculation of the distance between samples is ignored as much as possible.

[0073] In the specific implementation of this invention embodiment, feature weights are initialized for all dual-card mining data. Weighted distance is considered when calculating sample distance. Throughout the training process, this invention embodiment assigns a high feature weight value to one key feature, namely, the simultaneous power-on / off signaling detection.

[0074] The mathematical expression for the weighted clustering model is:

[0075]

[0076] In the mathematical expression, P is the objective function, U is the cluster assignment matrix, Z is the cluster centroid matrix, and W is the dual-slot weight matrix. For weight parameters, The mathematical expression for the clustering model, which gives the distances to all sample points in the j-th dimension and the weights, follows... The constraints.

[0077] As can be seen from the above constraints, compared to the original clustering algorithm (such as KMeans), the weighted clustering function simply adds a weight parameter to the objective function. Its function is to calculate the weighted distance sum of each dimension when minimizing the total intra-cluster distance, that is, to adjust the influence of each dimension on the clustering result by using different weight values. Furthermore, when β = 0, the objective function degenerates into the objective function of the KMeans clustering algorithm.

[0078] In the specific implementation of this invention, once the system detects the user's mobile phone behavior, and based on signaling detection, it finds that two SIM cards are simultaneously powering on and off (within a short time window), then the feature weight will be weighted to a very high value (i.e., the weight parameter). (Assigned a higher value).

[0079] In this way, when calculating the weighted distance for each dimension within the entire minimized cluster, the higher weighting can influence the final analysis dimension's impact on the clustering results. This achieves the intended purpose of this embodiment of the invention. By limiting simultaneous power on / off, the accuracy and analysis results of the impact of weighted clustering data mining can be greatly improved.

[0080] Figure 9 This is a flowchart illustrating the process of correcting dual-SIM users using a weighted clustering model. The specific implementation of the correction is as follows.

[0081] First, initialize the power-on / off weight features obtained in step 20 above;

[0082] Secondly, wait for detected power on / off signaling behavior;

[0083] Again, judge whether the acquisition of simultaneous on-off signaling data collection, if no on-off signaling data is obtained, no processing directly to the next step;

[0084] If the on-off signaling data is obtained, the detected on-off signaling behavior is waited for and received;

[0085] Then, the identity information dimension characteristics, the call dimension characteristics, the signaling trajectory dimension characteristics and the like information are combined, and the mining analysis of the correlation is carried out by using the weight clustering model;

[0086] Finally, the double-card slot weight characteristics are corrected to determine the double-card slot user information of the terminal.

[0087] In the embodiment of the application, the analysis result of the user service is determined according to the double-card slot user information, the business analysis is carried out according to the analysis result of the user service, the user can be served by application, and the business analysis is carried out according to the analysis result of the user service, the characteristics and habits of the double-card user are better understood, so that more effective marketing strategies are formulated, the customer stickiness is improved, the existing customers are retained and the business usage of them is improved.

[0088] The embodiment of the application provides a double-card slot user identification system based on a terminal on-off signaling process, Figure 10 It is a system architecture diagram of a double-card slot user identification system based on a terminal on-off signaling process, and the system comprises:

[0089] S1010: a signaling extraction module

[0090] The signaling extraction module is used for extracting the on-off signaling process of the MME interface and / or N1, N2 interface of the terminal in the signaling information, and the business identification of the terminal and the user, in the embodiment of the application, the business identification of the terminal and the user comprises an IMSI number and an IMEI number, an IMEI_TAC number and a card slot identification.

[0091] S1020: an identification service module

[0092] The identification service module filters the first data according to the type of the on-off signaling process and a preset rule, analyzes the filtered first data based on the business identification of the terminal and the user, obtains the double-card slot analysis result of the terminal, and determines the double-card slot weight characteristics of the terminal according to the double-card slot analysis result.

[0093] S1030: a correction service module

[0094] The correction service module is used for obtaining the double-card slot business characteristics of the terminal, correcting the double-card slot weight characteristics by using a weight clustering model according to the double-card slot business characteristics, and determining the double-card slot user information of the terminal.

[0095] The apparatus embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0096] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and necessary universal hardware platforms, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of software products, and the computer software products can be stored in a computer readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and include a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in the embodiments or some parts of the embodiments.

[0097] Compared with the technical solution of terminal heartbeat data adopted in the prior art, the embodiment of the present application is not limited to whether the terminal installs SDK or a specific APP application, and is based on DPI data statistics, which can better guarantee data integrity. The user's resistance is avoided, and the solution has high timeliness.

[0098] In the analysis dimension of judging the dual-card condition of the user, it is not limited to the traditional identity registration or mobile trajectory based dimension analysis, but a key feature based on signaling switching is added to determine, and is used as a highly weighted weight feature dimension. As this key determining factor, the judgment of the single-machine dual-card behavior of the same user is strengthened. The analysis misjudgment caused by the special behavior of the user carrying two mobile phones at the same time or two people carrying mobile phones at the same time is avoided.

[0099] The above specific embodiments further detail the purposes, technical solutions and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A dual-SIM card slot user identification method based on terminal power-on / off signaling flow, characterized in that, include: Obtain the signaling information of the terminal, and extract the first data and the service identifier of the terminal and the user from the signaling information; the first data includes the power-on / off signaling process of the terminal's MME interface and / or N1, N2 interface; The first data is filtered according to the type of the power-on / off signaling process and preset rules; The filtered first data is analyzed based on the service identifiers of the terminal and the user to obtain the dual SIM card slot analysis results of the terminal. The dual SIM card slot weight characteristics of the terminal are determined based on the dual SIM card slot analysis results; wherein, the dual SIM card slot weight characteristics include the weighted dual SIM card slot power-on / off weight characteristics. Obtain the dual SIM card slot service characteristics of the terminal, and determine the dual SIM card slot user information of the terminal by correcting the weight features of the dual SIM card slots through a weighted clustering model based on the dual SIM card slot service characteristics. The mathematical expression for the weighted clustering model is as follows: P is the objective function, U is the cluster assignment matrix, Z is the cluster center matrix, and W is the dual-slot weight matrix. Here, β is a weighting parameter, and β is not 0. Let the sum of distances between all sample points in the j-th dimension be ; The mathematical expression of the weighted clustering model follows The constraints.

2. The dual-SIM card slot user identification method based on terminal power-on / off signaling flow according to claim 1, characterized in that, The preset rule is that the interval between two consecutive first data points is less than 2 seconds or less than a preset specified time.

3. The dual-SIM card slot user identification method based on terminal power-on / off signaling flow according to claim 1, characterized in that, The analysis results of the dual SIM card slots of the terminal include: When in two consecutive first data, the power-on signaling process of the N1 and N2 interfaces is procedure_type=6, and the power-off signaling process is procedure_type=1, and the interval between the two consecutive first data is less than 2 seconds or less than the preset specified time; Wherein, the terminal and the user's service identifier have different IMSI and IMEI numbers, but the same IMEI_TAC number, and the IMEI_TAC is for dual SIM card slots; then The analysis results of the dual SIM card slots indicate that the terminal uses a 4G network with dual SIM card slots.

4. The dual-SIM card slot user identification method based on terminal power-on / off signaling flow according to claim 1, characterized in that, The analysis results of the dual SIM card slots of the terminal also include: When in two consecutive first data, the power-on signaling process of interface N1 and N2 is procedure_type=2, and the power-off signaling process is procedure_type=1, and the interval between the two consecutive first data is less than 2 seconds or less than the preset specified time; Wherein, the terminal and the user's service identifier have different IMSI and IMEI numbers, but the same IMEI_TAC number, and the IMEI_TAC indicates a dual SIM card slot; then The analysis results of the dual SIM card slots indicate that the terminal uses a 5G network with dual SIM card slots.

5. The dual-SIM card slot user identification method based on terminal power-on / off signaling flow according to claim 1, characterized in that, The analysis results of the dual SIM card slots of the terminal also include: In two consecutive first data sets, the power-on signaling procedure for the MME interface is procedure_type=2, and the power-off signaling procedure is procedure_type=1; the power-on signaling procedure for the N1 and N2 interfaces is procedure_type=2, and the power-off signaling procedure is procedure_type=1; and the interval between the two consecutive first data sets is less than 2 seconds or less than a preset time. Wherein, the terminal and the user's service identifier have different IMSI and IMEI numbers, but the same IMEI_TAC number, and the IMEI_TAC indicates a dual SIM card slot; then The analysis results of the dual SIM card slots of the terminal show that one SIM card slot uses the 4G network and the other SIM card slot uses the 5G network.

6. The dual-SIM card slot user identification method based on terminal power-on / off signaling flow according to claim 1, characterized in that, The dual-SIM card slot service features of the terminal include identity information dimension features, call dimension features, and signaling trajectory dimension features; The identity information dimension features are the registration information of the user's number, including the correlation features of the user using a dual-network dual-standby terminal and using multiple numbers on the same terminal; The call dimension features are the convergence features of the users' voice call habits, including the average number of calls per day, the average call duration, the average call duration per call, and the correlation features of the call duration distribution in different time periods. The signaling trajectory dimension features are the convergence features of the user's terminal in terms of location movement, including the signaling activity trajectory of dual-SIM card slot users, the location where they stay at the same time, and the correlation features of the user terminal's static maximum station base station comparison analysis dimension.

7. A dual-card slot user identification system based on terminal power-on / off signaling flow, used to execute the method described in any one of claims 1-6, characterized in that, The system includes: The signaling extraction module is used to obtain the signaling information of the terminal, extract the first data and the service identifier of the terminal and the user from the signaling information; the first data includes the power-on and power-off signaling flow of the terminal's MME interface and / or N1, N2 interface; The identification service module filters the first data according to the type of the power-on / off signaling process and preset rules; it analyzes the filtered first data based on the service identifiers of the terminal and the user to obtain the dual SIM card slot analysis results of the terminal. The dual SIM card slot weight characteristics of the terminal are determined based on the dual SIM card slot analysis results. The correction service module is used to obtain the dual SIM card slot service characteristics of the terminal, and to determine the dual SIM card slot user information of the terminal by correcting the dual SIM card slot weight features through a weighted clustering model based on the dual SIM card slot service characteristics.

8. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the method described in any one of claims 1 to 6.

9. A computer-readable medium having processor-executable non-volatile program code, characterized in that, The program code causes the processor to execute the method according to any one of claims 1 to 6.

Images