Gender prediction method, apparatus, device, storage medium and system

Abstract

The application belongs to the technical field of communication, and particularly relates to a gender prediction method, device, equipment, storage medium and system. The method comprises the following steps: acquiring terminal signaling data, call data and online behavior data of a user to be predicted; acquiring user features of the user to be predicted according to the terminal signaling data, the call data and the online behavior data, wherein the user features comprise location features, call features, online behavior features and identity features; inputting the user features into a gender prediction model to obtain a gender prediction result of the user to be predicted, wherein the gender prediction model is a classification model used for predicting gender; and thus the gender of the user can be accurately predicted, the accuracy of potential user analysis is greatly improved, human resource cost is saved, and the applicability is high.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a gender prediction method, apparatus, device, storage medium, and system. Background Technology

[0002] Currently, it is increasingly common for a single user to own multiple mobile devices. With the implementation of real-name authentication for mobile phone SIM card services, the "one ID, multiple SIM cards" model has emerged. This means that the same valid identity information can be used to register multiple mobile phone SIM cards. However, the existence of this model has led to discrepancies in the gender of some secondary SIM card users.

[0003] Existing methods for confirming a user's gender mainly involve making outbound voice calls to the user, or determining it through questionnaires, door-to-door surveys, etc.

[0004] However, existing methods for determining a user's gender suffer from drawbacks such as high time costs, high human resource consumption, and low accuracy. Summary of the Invention

[0005] This application provides a gender prediction method, apparatus, device, storage medium, and system to address the shortcomings of existing methods for determining user gender, such as high time cost, large human resource consumption, and low accuracy.

[0006] Firstly, this application provides a gender prediction method, including:

[0007] Acquire terminal signaling data, call data, and internet access behavior data of the user to be predicted;

[0008] Based on the terminal signaling data, call data, and internet access behavior data, the user characteristics of the user to be predicted are obtained, including location characteristics, call characteristics, internet access behavior characteristics, and identity characteristics.

[0009] The user features are input into the gender prediction model to obtain the gender prediction result of the user to be predicted; wherein, the gender prediction model is a classification model used to predict gender.

[0010] Optionally, obtaining the user characteristics of the user to be predicted based on the terminal signaling data, call data, and internet access behavior data includes:

[0011] The location features are obtained based on the location signaling in the terminal signaling data;

[0012] Based on the call data, the call features are obtained, and the call features include at least one of male call information, female call information, or gender information of users who meet preset call conditions within a preset time period.

[0013] Based on the internet access behavior data, the internet access behavior characteristics are obtained, including access information for APPs that meet preset conditions, and the access information includes at least one of access duration, access traffic, or access count;

[0014] The identity features are obtained based on the terminal signaling data, call data, and internet access behavior data.

[0015] Optionally, obtaining the identity features based on the terminal signaling data, call data, and internet access behavior data includes:

[0016] Based on the terminal signaling data, the movement speed of the user to be predicted is obtained, and the first candidate identity feature of the user to be predicted is obtained based on the movement speed.

[0017] Based on the call records in the call data, obtain the second candidate identity features of the user to be predicted;

[0018] Based on the access information of different types of apps in the internet behavior data, obtain the third candidate identity features of the user to be predicted;

[0019] The identity features are determined based on the first candidate identity features, the second candidate identity features, and the third candidate identity features.

[0020] Optionally, obtaining the movement speed of the user to be predicted based on the terminal signaling data includes:

[0021] Based on the terminal signaling data, obtain the movement distance and movement duration of the user to be predicted during each movement within a preset time period;

[0022] Based on the distance and duration of each move, obtain the total distance and duration of movement.

[0023] The movement speed is obtained based on the total movement distance and the total movement time.

[0024] Optionally, obtaining the first candidate identity features of the user to be predicted based on the movement speed includes:

[0025] The first candidate identity features of the user to be predicted are obtained based on the speed range to which the movement speed belongs, wherein different speed ranges correspond to different identity features.

[0026] Optionally, obtaining the second candidate identity features of the user to be predicted based on the call records in the call data includes:

[0027] Based on the call records in the call data, obtain the number of calls made through the virtual number within a preset time period;

[0028] Based on the range of the number of calls, the second candidate identity features of the user to be predicted are obtained, wherein different ranges of calls correspond to different identity features.

[0029] Optionally, obtaining the third candidate identity features of the user to be predicted based on access information of different types of apps in the internet behavior data includes:

[0030] Based on the access information of different types of apps in the internet access behavior data, obtain the number of times each type of app is accessed within a preset time period;

[0031] The system identifies APP types that have been accessed more than a preset number of times, and obtains a third candidate identity feature based on the APP type. Different APP types correspond to different identity features.

[0032] Optionally, before determining the identity feature based on the first candidate identity feature, the second candidate identity feature, and the third candidate identity feature, the method further includes:

[0033] Set the value of the element in the first vector corresponding to the first candidate identity feature to 1;

[0034] Set the value of the element in the second vector corresponding to the second candidate identity feature to 1;

[0035] Set the value of the element in the third vector corresponding to the third candidate identity feature to 1;

[0036] The step of determining the identity feature based on the first candidate identity feature, the second candidate identity feature, and the third candidate identity feature includes:

[0037] The identity feature is obtained by multiplying the values ​​of the elements at the same position in the first vector, the second vector, and the third vector.

[0038] Optionally, the classification model is a DeepFM model.

[0039] Secondly, this application provides a gender prediction device, comprising:

[0040] The acquisition module is used to acquire terminal signaling data, call data, and internet access behavior data of the user to be predicted;

[0041] The acquisition module is further configured to acquire user characteristics of the user to be predicted based on the terminal signaling data, call data, and internet access behavior data. The user characteristics include location characteristics, call characteristics, internet access behavior characteristics, and identity characteristics.

[0042] The prediction module is used to input the user features into the gender prediction model to obtain the gender prediction result of the user to be predicted; wherein, the gender prediction model is a classification model used to predict gender.

[0043] Optionally, the acquisition module is specifically used to acquire the location features based on the location signaling in the terminal signaling data;

[0044] The acquisition module is specifically used to acquire the call features based on the call data. The call features include at least one of male call information, female call information, or gender information of users who meet preset call conditions within a preset time period.

[0045] The acquisition module is specifically used to acquire the internet behavior characteristics based on the internet behavior data. The internet behavior characteristics include access information for accessing apps that meet preset conditions. The access information includes at least one of access duration, access traffic, or access count.

[0046] The acquisition module is specifically used to acquire the identity features based on the terminal signaling data, call data, and internet access behavior data.

[0047] Optionally, the acquisition module is specifically used to acquire the movement speed of the user to be predicted based on the terminal signaling data, and to acquire the first candidate identity feature of the user to be predicted based on the movement speed.

[0048] The acquisition module is specifically used to acquire the second candidate identity features of the user to be predicted based on the call records in the call data.

[0049] The acquisition module is specifically used to acquire the third candidate identity features of the user to be predicted based on the access information of different types of APPs in the internet behavior data.

[0050] The gender prediction device further includes: a determination module;

[0051] The determining module is used to determine the identity feature based on the first candidate identity feature, the second candidate identity feature, and the third candidate identity feature.

[0052] Optionally, the acquisition module is specifically used to acquire the movement distance and movement duration of the user to be predicted during each movement within a preset time period based on the terminal signaling data.

[0053] The acquisition module is specifically used to acquire the total moving distance and total moving time based on the moving distance and moving time of each move;

[0054] The acquisition module is specifically used to acquire the movement speed based on the total movement distance and the total movement time.

[0055] Optionally, the acquisition module is specifically used to acquire the first candidate identity features of the user to be predicted based on the speed range to which the movement speed belongs, wherein different speed ranges correspond to different identity features.

[0056] Optionally, the acquisition module is specifically used to acquire the number of calls made through the virtual number within a preset time period based on the call records in the call data;

[0057] The acquisition module is specifically used to acquire the second candidate identity features of the user to be predicted based on the range of the number of calls, wherein different ranges of calls correspond to different identity features.

[0058] Optionally, the acquisition module is specifically used to acquire the number of times each type of APP is accessed within a preset time period based on the access information of different types of APP in the Internet access behavior data;

[0059] The acquisition module is specifically used to acquire APP types that have been accessed more than a preset number of times, and to acquire a third candidate identity feature based on the APP type, wherein different APP types correspond to different identity features.

[0060] Optionally, the gender prediction device further includes: a setting module;

[0061] The setting module is used to set the element in the first vector corresponding to the first candidate identity feature to 1; set the element in the second vector corresponding to the second candidate identity feature to 1; and set the element in the third vector corresponding to the third candidate identity feature to 1.

[0062] The acquisition module is specifically used to multiply the values ​​of the elements at the same position in the first vector, the second vector, and the third vector to obtain the identity feature.

[0063] Optionally, the classification model is a DeepFM model.

[0064] Thirdly, this application provides a gender prediction device, comprising:

[0065] Memory;

[0066] processor;

[0067] The memory stores computer-executed instructions;

[0068] The processor executes computer execution instructions stored in the memory to implement the gender prediction method as described in the first aspect and various possible implementations of the first aspect above.

[0069] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the gender prediction method as described in the first aspect and various possible implementations of the first aspect above.

[0070] The gender prediction method provided in this application acquires terminal signaling data, call data, and internet behavior data of the user to be predicted; based on the terminal signaling data, call data, and internet behavior data, it obtains the user characteristics of the user to be predicted, including location characteristics, call characteristics, internet behavior characteristics, and identity characteristics; the user characteristics are input into a gender prediction model to obtain the gender prediction result of the user to be predicted; wherein, the gender prediction model is a classification model for predicting gender; this method can accurately predict the gender of users, greatly improve the accuracy of potential user analysis, save human resource costs, and has strong applicability. Attached Figure Description

[0071] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0072] Figure 1 The process of the gender prediction method provided in this application Figure 1 ;

[0073] Figure 2 The process of the gender prediction method provided in this application Figure 2 ;

[0074] Figure 3 The process of the gender prediction method provided in this application Figure 3 ;

[0075] Figure 4 A schematic diagram of the gender prediction device provided in this application;

[0076] Figure 5 A schematic diagram of the gender prediction device provided in this application.

[0077] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0078] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0079] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented, for example, in orders other than those illustrated or described herein.

[0080] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0081] Currently, it is increasingly common for a single user to own multiple mobile devices. With the implementation of real-name authentication for mobile phone SIM card services, the "one ID, multiple SIM cards" model has emerged. This means that the same valid identity information can be used to register multiple mobile phone SIM cards. However, the existence of this model has led to discrepancies in the gender of some secondary SIM card users.

[0082] Existing methods for confirming a user's gender mainly involve making outbound voice calls to the user, or using methods such as questionnaires and door-to-door surveys.

[0083] However, confirming a user's gender through voice calls can cause unnecessary disturbance and may result in misjudgments. Confirming a user's gender through questionnaires or door-to-door surveys requires a large number of staff, which consumes a lot of human resources. In addition, this method is difficult to cover all users, and the accuracy of the statistics is low, making it less applicable.

[0084] To address the aforementioned issues, this application provides a gender prediction method. This method obtains user characteristics of the user to be predicted based on terminal signaling data, call data, and internet browsing behavior data. These user characteristics are then input into a classification model for gender prediction to obtain the gender of the user. This method improves the accuracy of gender prediction while saving significant human resource costs and exhibits strong adaptability.

[0085] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0086] Figure 1 The flow chart of the gender prediction method provided in the embodiments of this application Figure 1 .like Figure 1 As shown in this embodiment, the gender prediction method includes:

[0087] S101: Obtain terminal signaling data, call data, and internet access behavior data of the user to be predicted.

[0088] Terminal signaling data refers to the communication data between the terminal user and the transmitting base station or micro-station. As long as the terminal is powered on and the operator's name is displayed on the screen, terminal signaling data will be generated. Terminal signaling data may include, for example, the user's location, travel distance, and travel time.

[0089] Call data may include, for example, the number of calls a user makes, the duration of the calls, and the gender of the person they are talking to.

[0090] Internet behavior data may include, for example, the type of app a user visits, the number of times they visit the app, and the duration of their visits.

[0091] Users to be predicted could be those who have been registered for a long time and have stable monthly voice calls, internet data usage, and location changes.

[0092] S102: Based on the terminal signaling data, call data, and internet access behavior data, obtain the user characteristics of the user to be predicted, including location characteristics, call characteristics, internet access behavior characteristics, and identity characteristics.

[0093] Among these features, location features can reflect the user's location and frequently visited areas; call features can indicate the gender distribution of the user's call partners; internet behavior features can indicate the types, durations, data usage, and frequency of apps accessed by the user; and identity features can indicate the user's identity settings. Identity settings can include, for example, delivery drivers, ride-hailing drivers, food bloggers, etc.

[0094] Understandably, the identity characteristics of the user to be predicted can be determined based on terminal signaling data, call data, and internet access behavior data, and the user to be predicted can possess multiple identity characteristics simultaneously. The purpose of this step is to obtain the user characteristics of the user to be predicted, so that the gender of the user to be predicted can be predicted subsequently based on these user characteristics.

[0095] S103: Input the user features into the gender prediction model to obtain the gender prediction result of the user to be predicted; wherein, the gender prediction model is a classification model used to predict gender.

[0096] The classification model is used to make decisions, and its output is a Boolean value (true or false) or a classification decision (such as "apple", "banana", or "cherry"). Common classification models include decision trees, random forests, GBDT, and XGB. In this embodiment, the classification model can be, for example, the DeepFM model.

[0097] Understandably, the gender prediction model in this embodiment is trained by inputting the user characteristics of the sample users into the classification model.

[0098] In this step, the input is the user characteristics of the user to be predicted, and the output is the classification decision, such as "0" or "1". When the output value is 0, it indicates that the gender of the user to be predicted is male, and when the output value is 1, it indicates that the gender of the user to be predicted is female.

[0099] The gender prediction method provided in this embodiment acquires terminal signaling data, call data, and internet behavior data of the user to be predicted; based on the terminal signaling data, call data, and internet behavior data, it obtains the user characteristics of the user to be predicted, including location characteristics, call characteristics, internet behavior characteristics, and identity characteristics; the user characteristics are input into a gender prediction model to obtain the gender prediction result of the user to be predicted; wherein, the gender prediction model is a classification model used to predict gender; this method can accurately predict the gender of users, greatly improve the accuracy of potential user analysis, save human resource costs, and has strong applicability.

[0100] Figure 2 The flow chart of the gender prediction method provided in the embodiments of this application Figure 2.like Figure 2 As shown, this embodiment is... Figure 1 Based on the embodiments, the gender prediction method is described in detail. The gender prediction method shown in this embodiment includes:

[0101] S201: Obtain terminal signaling data, call data, and internet access behavior data of the user to be predicted.

[0102] Step S201 is similar to step S101 above, and will not be repeated here.

[0103] S202: Obtain the location feature based on the location signaling in the terminal signaling data.

[0104] Specifically, the location characteristics of the user to be predicted can be obtained based on the location information in the terminal signaling data. These location characteristics are used to characterize the areas frequently visited by the user, such as shopping malls they frequent.

[0105] S203: Obtain the call characteristics based on the call data.

[0106] The call characteristics include at least one of the following: male call information, female call information, or gender information of users who meet the preset call conditions within a preset time period.

[0107] Male call information may include, for example, the number of calls, call duration, and call time of the user to be predicted to make to male targets within a preset time period; female call information may include, for example, the number of calls, call duration, and call time of the user to be predicted to make to female targets within a preset time period; the gender information of users who meet the preset call conditions may include, for example, the gender of the N users who made the most calls to the user to be predicted within the preset time period, where N≥1.

[0108] In this step, the call characteristics of the user to be predicted can be obtained based on the user's call data. For example, the number of times the user made calls with men and women within a month, as well as the genders of the three most frequent callers.

[0109] S204: Obtain the internet behavior characteristics based on the internet behavior data.

[0110] Among them, the online behavior characteristics include: access information of the user to be predicted to access APPs that meet preset conditions, and the access information includes at least one of access duration, access traffic or access frequency.

[0111] Apps that meet the preset conditions can be, for example, specific types of apps, the apps with the most visits, or the top M apps with the most downloads, where M≥1.

[0112] In this step, the user's app access information can be obtained based on the user's online behavior data. For example, at least one of the following can be obtained: access duration, data usage, or number of visits to 100 commonly used apps by the user.

[0113] S205: Obtain the identity features based on the terminal signaling data, call data, and internet access behavior data.

[0114] Here, identity characteristics refer to the identity settings of the user to be predicted. Examples of identity characteristics include: delivery driver, ride-hailing driver, food blogger, student, etc. Understandably, identity settings matching the user to be predicted can be determined from preset identity settings based on terminal signaling data, call data, and internet behavior data, and used as the identity characteristics of the user to be predicted.

[0115] This step can, for example, obtain the identity characteristics of the user to be predicted based on the apps they frequently visit or their call records. For instance, if the user frequently visits food delivery apps like Meituan Waimai or Ele.me, and their location frequently changes, this can identify them as a delivery driver. It is understood that a user to be predicted can possess multiple identity characteristics simultaneously. This embodiment does not impose any special restrictions on the specific implementation method for obtaining identity characteristics.

[0116] Understandably, when using a gender prediction model to predict the gender of a user, relying solely on the user's location, call history, and online behavior characteristics has limited effectiveness in predicting gender, except for online behavior characteristics related to users frequently accessing apps with strong gender distinctions, such as those tracking women's menstrual cycles. However, a user's call history, signaling information, and even online behavior are strongly correlated with their identity or occupation, and many identities or occupations have high gender distinctions; for example, occupational gender statistics show that 88% of delivery drivers are male, and 84% of sales assistants are female. Therefore, in this embodiment, by including the user's identity characteristics in the input features, the accuracy of gender prediction is improved.

[0117] S206: Input the location features, call features, internet behavior features, and identity features into the gender prediction model to obtain the gender prediction result of the user to be predicted.

[0118] Step S206 is similar to step S103 above, and will not be described again here.

[0119] The gender prediction method provided in this embodiment obtains identity features based on terminal signaling data, call data, and internet behavior data. It then inputs location features, call features, internet behavior features, and identity features into a gender prediction model. By using both low-order and high-order features, it predicts the user's gender, thereby improving the accuracy of the prediction and making it highly applicable.

[0120] Figure 3 The flow chart of the gender prediction method provided in the embodiments of this application Figure 3 .like Figure 3 As shown, this embodiment is... Figure 2 Based on the embodiments, this embodiment provides a detailed explanation of how to obtain the identity features based on the terminal signaling data, call data, and internet access behavior data. The gender prediction method shown in this embodiment includes:

[0121] S301: Based on terminal signaling data, obtain the movement speed of the user to be predicted, and obtain the first candidate identity feature of the user to be predicted based on the movement speed.

[0122] Specifically, the movement speed of the user to be predicted can be obtained based on the location information in the terminal signaling data, and then the first candidate identity feature of the user to be predicted can be obtained based on the movement speed.

[0123] Specifically, obtaining the movement speed of the user to be predicted based on terminal signaling data includes:

[0124] Based on the terminal signaling data, obtain the movement distance and movement duration of the user to be predicted during each movement within a preset time period; based on the movement distance and movement duration of each movement, obtain the total movement distance and total movement duration; based on the total movement distance and total movement duration, obtain the movement speed.

[0125] The preset time period can be, for example, one day or one month, and the specific value can be determined according to the actual situation.

[0126] In this step, the total moving distance and total moving time can be obtained based on the moving distance and moving time of the user to be predicted each time within a preset time period, and then the average moving speed of the user to be predicted within the preset time period can be calculated.

[0127] Specifically, obtaining the first candidate identity features of the user to be predicted based on the movement speed includes: obtaining the first candidate identity features of the user to be predicted based on the speed range to which the movement speed belongs.

[0128] Different speed ranges correspond to different identity characteristics. For example, the speed range for a food delivery worker is 15-30 km / h, while the speed range for a ride-hailing driver is 30-100 km / h. The first candidate identity characteristic for a user to be predicted can be determined based on the range of their average movement speed.

[0129] S302: Based on the call records in the call data, obtain the second candidate identity features of the user to be predicted.

[0130] Among them, there is a correlation between the identity characteristics of the user to be predicted and the call records in the call data. For example, if a user makes too many calls within a preset time period, it can be determined that the user works in the service industry.

[0131] Specifically, the number of calls made through a virtual number within a preset time period can be obtained from the call records in the call data; then, based on the range of the number of calls made through the virtual number, the second candidate identity features of the user to be predicted can be obtained.

[0132] It is understandable that users need to contact customers through virtual numbers to provide services for those with identities that require frequent phone calls, such as ride-hailing drivers and food delivery workers.

[0133] This step first obtains the number of calls made through the virtual number within a preset time period, and then determines the user's second candidate identity feature based on the range of these calls. Different ranges of calls correspond to different identity features.

[0134] For example, if it is found that a user made 30 calls through a virtual number in a day, then the user's second candidate identity characteristic can be determined to be a food delivery worker and / or a ride-hailing driver.

[0135] S303: Based on the access information of different types of APPs in the internet behavior data, obtain the third candidate identity features of the user to be predicted.

[0136] Users with different identity characteristics visit different types of apps. For example, users who are food delivery workers frequently visit food delivery apps, while users who are ride-hailing drivers frequently visit ride-hailing apps. Therefore, a third candidate identity characteristic of the user to be predicted can be obtained based on the user's access information to different types of apps in the user's online behavior data.

[0137] Specifically, based on the access information of different types of apps in the online behavior data, the number of times each type of app is accessed within a preset time period can be obtained; the types of apps with more than a preset number of accesses can be obtained, and a third candidate identity feature can be obtained based on the app type.

[0138] The process involves first obtaining the number of times the user to be predicted accesses different types of apps within a preset time period. Then, app types with access counts exceeding a preset number are selected. Based on these app types, a third candidate identity characteristic for the user to be predicted is determined. Understandably, different app types correspond to different identity characteristics. For example, if the app type is food delivery, the corresponding third candidate identity characteristic is a food delivery worker; similarly, if the app type is ride-hailing, the corresponding third candidate identity characteristic is a ride-hailing driver.

[0139] S304: Determine the identity feature based on the first candidate identity feature, the second candidate identity feature, and the third candidate identity feature.

[0140] After obtaining the first candidate identity features, the second candidate identity features, and the third candidate identity features, the identity features of the user to be predicted can be determined based on the above three candidate identity features.

[0141] For example, the intersection of the first, second, and third candidate identity features can be taken, that is, the identity features that are present in all three candidate identity features can be determined as the identity features of the user to be predicted.

[0142] Optionally, before determining the identity features of the user to be predicted, the element at the position corresponding to the first candidate identity feature in the first vector can be set to 1; the element at the position corresponding to the second candidate identity feature in the second vector can be set to 1; the element at the position corresponding to the third candidate identity feature in the third vector can be set to 1; and the values ​​of the elements at the same position in the first vector, the second vector, and the third vector can be multiplied together to obtain the identity feature.

[0143] In this vector, different candidate identity features correspond to different positions. By setting the values ​​of the elements at the corresponding positions of the first, second, and third candidate identity features in the vector to 1, we can obtain three vectors representing the first, second, and third candidate identity features, respectively. For example, if the first candidate identity feature is at position "4", the second candidate identity feature is at position "3", and the third candidate identity feature is at position "1", the first vector would be [0, 0, 0, 1], the second vector would be [0, 0, 1, 1], and the third vector would be [1, 0, 0, 1].

[0144] Multiplying the values ​​of the elements at the same position in the first, second, and third vectors yields the vector [0, 0, 0, 1] corresponding to the identity feature. The identity feature of the user to be predicted can then be determined based on the vector corresponding to this identity feature.

[0145] The gender prediction method provided in this embodiment improves the accuracy of identifying identity features by determining the candidate identity features of the user to be predicted from multiple dimensions, and then determining the identity features of the user to be predicted from multiple candidate identity features.

[0146] Figure 4 A schematic diagram of the gender prediction device provided in this application. Figure 4 As shown, this application provides a gender prediction device, the gender prediction device 300 including:

[0147] The acquisition module 301 is used to acquire terminal signaling data, call data, and internet access behavior data of the user to be predicted;

[0148] The acquisition module 301 is further configured to acquire user characteristics of the user to be predicted based on the terminal signaling data, call data and internet access behavior data, wherein the user characteristics include location characteristics, call characteristics, internet access behavior characteristics and identity characteristics;

[0149] The prediction module 302 is used to input the user features into the gender prediction model to obtain the gender prediction result of the user to be predicted; wherein, the gender prediction model is a classification model for predicting gender.

[0150] Optionally, the acquisition module 301 is specifically used to acquire the location features based on the location signaling in the terminal signaling data;

[0151] The acquisition module 301 is specifically used to acquire the call features based on the call data. The call features include at least one of male call information, female call information, or gender information of users who meet preset call conditions within a preset time period.

[0152] The acquisition module 301 is specifically used to acquire the internet behavior characteristics based on the internet behavior data. The internet behavior characteristics include access information for accessing APPs that meet preset conditions. The access information includes at least one of access duration, access traffic, or access count.

[0153] The acquisition module 301 is specifically used to acquire the identity features based on the terminal signaling data, call data, and internet access behavior data.

[0154] Optionally, the acquisition module 301 is specifically used to acquire the movement speed of the user to be predicted based on the terminal signaling data, and to acquire the first candidate identity feature of the user to be predicted based on the movement speed.

[0155] The acquisition module 301 is specifically used to acquire the second candidate identity features of the user to be predicted based on the call records in the call data.

[0156] The acquisition module 301 is specifically used to acquire the third candidate identity features of the user to be predicted based on the access information of different types of APPs in the Internet access behavior data.

[0157] The gender prediction device further includes: a determination module 303;

[0158] The determining module 303 is used to determine the identity feature based on the first candidate identity feature, the second candidate identity feature, and the third candidate identity feature.

[0159] Optionally, the acquisition module 301 is specifically used to acquire the movement distance and movement duration of the user to be predicted during each movement within a preset time period based on the terminal signaling data.

[0160] The acquisition module 301 is specifically used to acquire the total moving distance and total moving time based on the moving distance and moving time of each movement;

[0161] The acquisition module 301 is specifically used to acquire the movement speed based on the total movement distance and the total movement time.

[0162] Optionally, the acquisition module 301 is specifically used to acquire the first candidate identity features of the user to be predicted based on the speed range to which the movement speed belongs, wherein different speed ranges correspond to different identity features.

[0163] Optionally, the acquisition module 301 is specifically used to acquire the number of calls made through the virtual number within a preset time period based on the call records in the call data;

[0164] The acquisition module 301 is specifically used to acquire the second candidate identity features of the user to be predicted based on the range of the number of calls, wherein different ranges of calls correspond to different identity features.

[0165] Optionally, the acquisition module 301 is specifically used to acquire the number of times each type of APP is accessed within a preset time period based on the access information of different types of APP in the Internet access behavior data;

[0166] The acquisition module 301 is specifically used to acquire APP types with access counts greater than a preset number, and to acquire third candidate identity features based on the APP types, wherein different APP types correspond to different identity features.

[0167] Optionally, the gender prediction device further includes: a setting module 304;

[0168] The setting module 304 is used to set the element in the first vector corresponding to the first candidate identity feature to 1; set the element in the second vector corresponding to the second candidate identity feature to 1; and set the element in the third vector corresponding to the third candidate identity feature to 1.

[0169] The acquisition module 301 is specifically used to multiply the values ​​of the elements at the same position in the first vector, the second vector, and the third vector to obtain the identity feature.

[0170] Optionally, the classification model is a DeepFM model.

[0171] Figure 5 A schematic diagram of the gender prediction device provided in this application. Figure 5 As shown, this application provides a gender prediction device 400, which includes a receiver 401, a transmitter 402, a processor 403, and a memory 404.

[0172] Receiver 401 is used to receive instructions and data;

[0173] Transmitter 402 is used to send commands and data;

[0174] Memory 404 is used to store instructions executed by the computer;

[0175] Processor 403 is used to execute computer execution instructions stored in memory 404 to implement the various steps performed by the gender prediction method in the above embodiments. For details, please refer to the relevant descriptions in the foregoing embodiments of the gender prediction method.

[0176] Alternatively, the memory 404 can be either standalone or integrated with the processor 403.

[0177] When the memory 404 is set up independently, the electronic device also includes a bus for connecting the memory 404 and the processor 403.

[0178] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the gender prediction method performed by the gender prediction device described above.

[0179] It will be understood by those skilled in the art that all or some of the steps, systems, or apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0180] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

Claims

1. A method for predicting gender, characterized in that, include: Acquire terminal signaling data, call data, and internet access behavior data of the user to be predicted; Based on the terminal signaling data, call data, and internet access behavior data, the user characteristics of the user to be predicted are obtained, including location characteristics, call characteristics, internet access behavior characteristics, and identity characteristics. The step of obtaining the identity features based on the terminal signaling data, call data, and internet access behavior data includes: Based on the terminal signaling data, the movement speed of the user to be predicted is obtained, and the first candidate identity feature of the user to be predicted is obtained based on the movement speed. Based on the call records in the call data, obtain the second candidate identity features of the user to be predicted; Based on the access information of different types of apps in the internet behavior data, obtain the third candidate identity features of the user to be predicted; The identity feature is determined based on the first candidate identity feature, the second candidate identity feature, and the third candidate identity feature, wherein the identity feature is the identity setting of the user to be predicted; The user features are input into the gender prediction model to obtain the gender prediction result of the user to be predicted; wherein, the gender prediction model is a classification model used to predict gender.

2. The method according to claim 1, characterized in that, The step of obtaining the user characteristics of the user to be predicted based on the terminal signaling data, call data, and internet access behavior data includes: The location features are obtained based on the location signaling in the terminal signaling data; Based on the call data, the call features are obtained, and the call features include at least one of male call information, female call information, or gender information of users who meet preset call conditions within a preset time period. Based on the internet access behavior data, the internet access behavior characteristics are obtained, including access information for APPs that meet preset conditions, and the access information includes at least one of access duration, access traffic, or access count; The identity features are obtained based on the terminal signaling data, call data, and internet access behavior data.

3. The method according to claim 1, characterized in that, The step of obtaining the movement speed of the user to be predicted based on the terminal signaling data includes: Based on the terminal signaling data, obtain the movement distance and movement duration of the user to be predicted during each movement within a preset time period; Based on the distance and duration of each move, obtain the total distance and duration of movement. The movement speed is obtained based on the total movement distance and the total movement time.

4. The method according to claim 1, characterized in that, The step of obtaining the first candidate identity features of the user to be predicted based on the movement speed includes: The first candidate identity features of the user to be predicted are obtained based on the speed range to which the movement speed belongs, wherein different speed ranges correspond to different identity features.

5. The method according to claim 1, characterized in that, The step of obtaining the second candidate identity features of the user to be predicted based on the call records in the call data includes: Based on the call records in the call data, obtain the number of calls made through the virtual number within a preset time period; Based on the range of the number of calls, the second candidate identity features of the user to be predicted are obtained, wherein different ranges of calls correspond to different identity features.

6. The method according to claim 1, characterized in that, The step of obtaining the third candidate identity features of the user to be predicted based on access information of different types of apps in the internet access behavior data includes: Based on the access information of different types of apps in the internet access behavior data, obtain the number of times each type of app is accessed within a preset time period; The system identifies APP types that have been accessed more than a preset number of times, and obtains a third candidate identity feature based on the APP type. Different APP types correspond to different identity features.

7. The method according to claim 1, characterized in that, Before determining the identity feature based on the first candidate identity feature, the second candidate identity feature, and the third candidate identity feature, the method further includes: Set the value of the element in the first vector corresponding to the first candidate identity feature to 1; Set the value of the element in the second vector corresponding to the second candidate identity feature to 1; Set the value of the element in the third vector corresponding to the third candidate identity feature to 1; The step of determining the identity feature based on the first candidate identity feature, the second candidate identity feature, and the third candidate identity feature includes: The identity feature is obtained by multiplying the values ​​of the elements at the same position in the first vector, the second vector, and the third vector.

8. The method according to claim 1, characterized in that, The classification model is the DeepFM model.

9. A gender prediction device, characterized in that, The device includes: The acquisition module is used to acquire terminal signaling data, call data, and internet access behavior data of the user to be predicted; The acquisition module is further configured to acquire user characteristics of the user to be predicted based on the terminal signaling data, call data, and internet access behavior data. The user characteristics include location characteristics, call characteristics, internet access behavior characteristics, and identity characteristics. The acquisition module is specifically used to acquire the movement speed of the user to be predicted based on the terminal signaling data, and to acquire the first candidate identity feature of the user to be predicted based on the movement speed. Based on the call records in the call data, obtain the second candidate identity features of the user to be predicted; Based on the access information of different types of apps in the internet behavior data, obtain the third candidate identity features of the user to be predicted; The identity feature is determined based on the first candidate identity feature, the second candidate identity feature, and the third candidate identity feature, wherein the identity feature is the identity setting of the user to be predicted; The prediction module is used to input the user features into the gender prediction model to obtain the gender prediction result of the user to be predicted; wherein, the gender prediction model is a classification model used to predict gender.

10. A gender prediction device, characterized in that, include: Memory; processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the gender prediction method as described in any one of claims 1-8.

11. A computer storage medium, characterized in that, The computer storage medium stores computer execution instructions, which, when executed by a processor, are used to implement the gender prediction method as described in any one of claims 1-8.

Images