Methods and apparatus for identifying secondary number issuance, and storage media

Abstract

The present disclosure relates to a method and device for identifying secondary number release, and a storage medium, which are used to solve the technical problem of the lack of a method for identifying secondary number release in the related art. The method for identifying secondary number release comprises: constructing a portrait label according to the historical login data of the user's login account; training a coding model according to the constructed portrait label and encoding the portrait label using the trained coding model; training a classification model according to the coding completed by the coding model so that the trained classification model can classify the mobile phone number used by the user's login account.

Description

Technical Field

[0001] The present disclosure relates to the field of Internet technology, and in particular, to a method and device for identifying secondary number release, and a storage medium. Background Art

[0002] Secondary number release refers to the operator freezing the user's abandoned mobile phone number, and then putting it back into the number selection system to be issued to new users after a three-month freezing period. Therefore, if the user does not unbind the bank card or application software when changing the mobile phone number, the new user holding the mobile phone card can log in to the account through the mobile phone number to obtain the previous user's data privacy or make purchases, causing the previous user's information leakage or asset loss. Summary of the invention

[0003] The present invention provides a method and device for identifying secondary number release, and a storage medium to solve the technical problem of lack of a method for identifying secondary number release in the related art.

[0004] To achieve the above-mentioned purpose, a first aspect of an embodiment of the present disclosure provides a method for identifying a secondary number release, the method comprising:

[0005] Build a portrait tag based on the historical login data of the user's login account;

[0006] According to the constructed portrait labels, the encoding model is trained and the portrait labels are encoded using the trained encoding model;

[0007] According to the encoding completed by the encoding model, the classification model is trained so that the trained classification model can classify the mobile phone number used by the user to log in to the account.

[0008] Optionally, constructing a portrait tag based on historical login data of the user's login account includes:

[0009] Calculate the Gini index of different attribute data in the historical login data;

[0010] Determining target attribute data according to the size of the Gini index;

[0011] Constructing a user profile according to the target attribute data;

[0012] Filter out the login data of the account when it wakes up after being silent for more than a preset time from the data of the user portrait;

[0013] Determine whether the login data when the account is awakened appears in the historical login data to obtain a user account portrait tag.

[0014] Optionally, the target attribute data includes login time, and at least one of device ID, IP address, international mobile subscriber identity code, international mobile equipment identity code, and geographic information.

[0015] Optionally, the encoding model is a GBDT model.

[0016] Optionally, it also includes:

[0017] If the mobile phone number used by the user to log in to the account is classified as a mobile phone number that has been re-released by a telecommunications operator, verify the user's identity information;

[0018] If the verification fails, a virtual user ID is assigned to the account to seal the account.

[0019] According to a second aspect of the embodiments of the present disclosure, there is provided a device for identifying a secondary number release, the device comprising:

[0020] A construction module is used to construct a portrait label based on the historical login data of the user's login account;

[0021] A first training module is used to train a coding model according to the constructed portrait labels and encode the portrait labels using the trained coding model;

[0022] The second training module is used to train the classification model according to the encoding completed by the encoding model so that the trained classification model can classify the mobile phone number used by the user to log in to the account.

[0023] Optionally, the building blocks include:

[0024] A calculation submodule, used for calculating the Gini index of different attribute data in the historical login data;

[0025] A determination submodule, used to determine target attribute data according to the size of the Gini index;

[0026] A construction submodule, used to construct a user portrait according to the target attribute data;

[0027] A screening submodule, for screening out the login data of accounts awakened after being silent for more than a preset time from the data of the user portrait;

[0028] The acquisition submodule is used to determine whether the login data when the account is awakened appears in the historical login data to obtain the user account portrait label.

[0029] Optionally, the target attribute data includes login time, and at least one of device ID, IP address, international mobile subscriber identity code, international mobile equipment identity code, and geographic information.

[0030] Optionally, the encoding model is a GBDT model.

[0031] Optionally, it also includes:

[0032] A verification module, used to verify the user's identity information if the mobile phone number used by the user to log in to the account is classified as a mobile phone number that has been re-released by a communication operator;

[0033] The processing module is used for assigning a virtual user ID to the account to seal the account if the verification fails.

[0034] According to a third aspect of an embodiment of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the steps of any one of the methods described in the first aspect are implemented.

[0035] According to a fourth aspect of the embodiments of the present disclosure, there is provided a device for secondary number release based on user portrait identification, including:

[0036] a memory having a computer program stored thereon; and

[0037] A processor is used to execute the computer program in the memory to implement the steps of any one of the methods described in the first aspect above.

[0038] By adopting the above technical solution, at least the following technical effects can be achieved:

[0039] The present invention constructs a portrait label of a user portrait through the historical login data of the user's login account, and then trains a coding model based on the portrait label and a coding training classification model output by the trained coding model, so that the trained classification model can distinguish whether the mobile phone number used by the classified user's login account is a mobile phone number released for the second time by a communication operator, thereby solving the technical problem of the lack of a method for identifying the second-released number in the related technology, and avoiding the leakage of the original user's information and the loss of assets caused by other users using the mobile phone number released for the second time by the communication operator to log in to the application that the original number owner has not unbound.

[0040] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:

[0042] Figure 1 The present invention is a flowchart of a method for identifying a secondary number release, shown in an exemplary embodiment of the present invention.

[0043] Figure 2It is a flowchart of constructing a portrait label in the steps included in a method for identifying secondary number release according to an exemplary embodiment.

[0044] Figure 3 The present invention is a block diagram of a device for identifying a secondary number release, shown in an exemplary embodiment of the present invention.

[0045] Figure 4 The present invention is a block diagram of an apparatus for identifying a secondary number release, shown in an exemplary embodiment of the present invention.

[0046] Figure 5 It is a block diagram of building blocks of a device for identifying secondary number release shown in an exemplary embodiment of the present disclosure.

[0047] Figure 6 The present invention is a block diagram of a device for identifying a secondary number release, shown in an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0048] The specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.

[0049] Most of the solutions for identifying secondary number release in related technologies need to rely on data provided by operators, but this solution has the following problems:

[0050] 1. It is impossible to effectively protect the privacy and financial security of users in scenarios where the user does not unbind the application software or bank card when canceling the mobile phone number with the operator. The relevant technical solutions are mostly aimed at the following scenarios: when a user registers a software account with a new mobile phone number, he finds that the current mobile phone number has been registered by someone else. At this time, the user can log out of the registration, log in to the application software with the mobile phone number and mobile phone verification code, and then obtain the privacy information of the previous user that has not been cancelled, and even use the electronic wallet and bank card bound by the previous user of the mobile phone number for consumption.

[0051] 2. Data depends on communication operators. There is a certain delay between the process of Internet companies applying for data from communication operators and receiving feedback data, which makes it difficult to support real-time secondary number verification when users log in. If the process of requesting from the operator is followed to verify whether it is a secondary number, the resulting delay will inevitably affect the user experience of normal users.

[0052] 3. It is very costly for Internet companies to obtain data interfaces provided by communication operators. The premise for communication operators to provide data is that they cannot infringe on the privacy of users, so Internet companies need to ask users for authorization. Therefore, existing technical solutions such as using users' communication information are extremely unfeasible.

[0053] To solve the above problems, please refer to Figure 1 , Figure 1 The present invention is a flowchart of a method for identifying secondary number release based on user portrait according to an exemplary embodiment, so as to solve the technical problem that there is a lack of a method for identifying secondary number release in the related art. Figure 1 As shown, the method for secondary number release based on user portrait identification includes:

[0054] S11, construct a portrait label based on the historical login data of the user's login account.

[0055] S12: According to the constructed portrait labels, the encoding model is trained and the portrait labels are encoded using the trained encoding model.

[0056] S13, training a classification model according to the encoding completed by the encoding model so that the trained classification model can classify the mobile phone number used by the user to log in to the account.

[0057] The account may be a client account installed on a mobile terminal, and the mobile terminal may be a user device that accesses network services through a mobile communication network, such as a smart phone, a tablet computer, a laptop computer, etc. When accessing a network service, the client initiates a network connection request, and based on the network connection request, the mobile terminal may connect to the mobile communication network and access the network service.

[0058] Before executing step S11, when the user logs in to the client on the mobile terminal through the account, the client server can obtain the login data at the time of login, and the client server can store the historical login data of the user logging in to the client. The login data can be the environmental data at the time of login, such as login time, login location, login device information, user IP information, etc.

[0059] After obtaining the login data, execute step S11 to construct a portrait tag based on the historical login data of the user's login account. When constructing a portrait tag, you need to first construct a user portrait, and then construct a portrait tag. Please refer to Figure 2 , based on the historical login data of the user's login account, a portrait tag is constructed. First, the following steps can be included to construct a user portrait:

[0060] S111, calculating the Gini index of different attribute data in the historical login data.

[0061] S112, determining target attribute data according to the size of the Gini index.

[0062] S113: construct a user portrait according to the target attribute data.

[0063] The Gini index (Gini for short) can represent the purity of data. For a data set D, Gini(D) reflects the probability that two samples are randomly selected from the data set D and their category labels are inconsistent. Therefore, the smaller Gini(D) is, the higher the purity of the data set D is. The data set D can be a certain attribute data, such as geographic information data, and different data sets are different attribute data. The target attribute data can be attribute data whose Gini index is less than a preset value.

[0064] The calculation formula of Gini can be expressed as:

[0065]

[0066] Among them, p k It indicates the probability that the selected sample belongs to the kth class.

[0067] Assume that there are K categories in the data set D, then:

[0068]

[0069] Therefore, by calculating the Gini coefficient of all environmental characteristics, we can select characteristic attributes such as device ID, IP address, IMSI (International Mobile Subscriber Identity), IMEI (International Mobile Equipment Identity), geographic information, login time, etc. associated with the user ID to build a user account portrait.

[0070] Optionally, the target attribute data includes at least login time, and also includes at least one attribute data of device ID, IP address, international mobile user identity code, international mobile equipment identity code, and geographic information.

[0071] For example, based on the above features, the information related to the user account history is counted and a user account portrait with the user ID as the index is generated. The example is shown in Table 1:

[0072] Table 1

[0073]

[0074] Of course, in other embodiments, information gain, information gain rate and other methods may be used instead of the Gini coefficient to select target attribute data.

[0075] After creating a user portrait, you need to build a portrait tag based on the created user portrait. Please refer to Figure 2 After building a user portrait, building a portrait tag can include the following steps:

[0076] S114, filtering out the login data when the account whose silence time exceeds the preset time is awakened from the data of the user portrait.

[0077] S145, determining whether the login data when the account is awakened appears in the historical login data to obtain a user account portrait tag.

[0078] For example, based on the collected user portrait data, filter out all the user login data that has been silent for more than 3 months and has woken up in history (login data statistics end at the time when the user wakes up). By determining whether the login data appears in the user's historical login data, the user's portrait label is obtained. The login data when the user wakes up is shown in Table 2 below:

[0079] Table 2

[0080]

[0081] It can be defined as follows: "Whether the current user ID appears in the user portrait" is attribute A, "Whether the current IP address appears in the user portrait" is attribute B, "Whether the current imsi appears in the user portrait" is attribute C, "Whether the current imei appears in the user portrait" is attribute D, and "Whether the current login city appears in the user portrait" is attribute E. The portrait label dataset is shown in Table 3 below:

[0082] Table 3

[0083] User ID A B C D E Image Tags 13224257 no no yes yes yes No secondary release ...... ...... ...... ...... ...... ...... ...... 23456768 no no no no no Secondary release

[0084] After the portrait label is constructed, execute step S12, train the encoding model according to the constructed portrait label, and use the trained encoding model to encode the portrait label. In machine learning, features determine the upper limit of model performance. For example, deep learning methods are also about how to better express data as features. If the data can be expressed as linearly separable data, then using a simple linear model can achieve good results. In step S12, the encoding model can use a GBDT (Gradient Boosting Decision Tree; gradient boosting decision tree) model. Among them, the data of the user portrait constructed in step S113 can be divided into two, half of the user portrait data is used to train the GBDT model, and the other half of the user portrait data is used to encode the trained GBDT model.

[0085] GBDT is an integrated learning model based on the boosting method. It builds an additive model based on the decision tree as the basic function and generates a boosting tree through the forward distribution algorithm. The process of GBDT model training is to continuously improve the accuracy of the final classifier by reducing the deviation, so the weak classifier is required to have the characteristics of low variance and high deviation. During training, GBDT goes through multiple rounds of iterations, and each round of iteration generates a weak classifier. Each classifier is trained based on the residual of the previous round of classifiers.

[0086] The GBDT model has excellent classification performance, and can find combinations of features based on the expansion process of leaf nodes during training, thereby obtaining more information about the data. Since the system can only obtain very little user login information in the secondary number allocation scenario, deep mining of the hidden meanings between information is particularly important for identifying the secondary number allocation scenario.

[0087] The sample code (python version) for constructing features using GBDT is as follows:

[0088] n_estimator = 10 # number of weak classifiers

[0089] X,y=make_classification(n_samples=10000)#Randomly construct 10000 binary classification data sets

[0090] X_train,X_test,y_train,y_test=train_test_split(X,y,test_size=0.5)#Build training and test sets

[0091] grd=GradientBoostingClassifier(n_estimators=n_estimator)#GBDT model

[0092] grd_enc=OneHotEncoder() #OneHot encoding

[0093] grd.fit(X_train,y_train)#Train GBDT model

[0094] grd_enc.fit(grd.apply(X_test)[:,:,0])#Encode X_test

[0095] Based on the above steps, the collected portrait label data is used to train the GBDT model, and then step S13 is performed to encode all portrait labels using the trained GBDT model to obtain the final training set for classification model training. The trained classification model can then classify whether the mobile phone number used by the user to log in to the account is a mobile phone number that has been re-released by the communication operator.

[0096] The classification model can be a Logistic model. Logistic is a linear model, but its learning ability is very limited, and a lot of feature engineering is required to increase the learning ability of the model. The training set constructed by extracting features using GBDT can solve the feature combination problem of the Logistic model. Based on this, the present disclosure can use the Logistic model to construct a secondary number classification model. The sample code is as follows:

[0097] From sklearn.linear_model import LogisticRegression

[0098] lr=LogisticRegression()

[0099] lr.fit(features_trans,y_lr)

[0100] lr.predict(features_trans)

[0101] lr.predict_proba(features_trans)[:,1]

[0102] Of course, in other embodiments, neural networks or other methods are used instead of GBDT to encode portrait labels, and neural networks or other classification models such as FM (Factorization Machine) model and FFM (Field-aware Factorization Machine), FMM model, etc. are used instead of Logistic to construct a secondary number placement classification model.

[0103] Since the data used by the classification model is only the accumulation of historical data over a period of time, the result obtained according to the classification model is a local optimal result, so the result of the classification model may be misclassified. In order to reduce the impact of misclassification, the present disclosure only performs classification verification when users whose silent period exceeds the freezing period of the communication operator log in, and at the same time performs identity verification on users identified by the classification model as secondary releases, such as: asking users to fill in verification information, asking users to feedback identity information, etc. Verification methods that can be used: ID number filled in by the user when registering; user's historical order records, etc.

[0104] For users whose classification results are not for secondary release or whose classification results are for secondary release but who have passed verification, they are allowed to log in to their accounts; for users whose classification results are for secondary release, they are guided to conduct historical behavior verification. If the user cannot pass the verification, the user's login behavior will be rejected. At the same time, the user's logged-in account will be sealed, and he will be guided to register a new account.

[0105] Optionally, the strategy for sealing an account may be to assign an associated virtual user ID to the account for storage, and when the real user of the account passes identity authentication, the real user's new mobile phone number is used as the account to replace the virtual user ID.

[0106] The present invention constructs a user profile through the historical login data of the user's login account, and then trains a coding model based on the user profile related data and a coding training classification model output by the trained coding model, so that the trained classification model can distinguish whether the mobile phone number used by the classified user's login account is a mobile phone number released for the second time by a communication operator, thereby solving the technical problem of the lack of a method for identifying the second released number in the related technology, and avoiding the leakage of the original user's information and the loss of assets caused by other users using the mobile phone number released for the second time by the communication operator to log in to the application that the original number owner has not unbound.

[0107] In addition, the present disclosure only performs classified verification when users whose silent period exceeds the freezing period of the communication operator log in, avoiding disturbing normal users. Moreover, since the data information used is information that can be obtained by the client, the information acquisition cost is low, and the implementation cost of the technical solution is low.

[0108] It is worth noting that for Figure 1 The method embodiments shown are described as a series of action combinations for the sake of simplicity, but those skilled in the art should be aware that the present disclosure is not limited by the order of the actions described. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the present disclosure.

[0109] Figure 3 FIG. 1 is a block diagram of a device for secondary number placement based on user portrait recognition according to an exemplary embodiment of the present disclosure. Figure 3 As shown, the device 300 for secondary number release based on user portrait identification includes:

[0110] A construction module 310 is used to construct a portrait tag based on historical login data of the user login account;

[0111] A first training module 320 is used to train a coding model according to the constructed portrait label and encode the portrait label using the trained coding model;

[0112] The second training module 330 is used to train the classification model according to the encoding completed by the encoding model so that the trained classification model can classify the mobile phone number used by the user to log in to the account.

[0113] Alternatively, if Figure 4 As shown, the building module 310 includes:

[0114] A calculation submodule 311, used to calculate the Gini index of different attribute data in the historical login data;

[0115] A determination submodule 312 is used to determine target attribute data according to the size of the Gini index;

[0116] A construction submodule 313 is used to construct a user portrait according to the target attribute data;

[0117] A screening submodule 314 is used to screen out the login data of accounts whose silence time exceeds a preset time when they are awakened from the data of the user portrait;

[0118] The acquisition submodule 315 is used to determine whether the login data when the account is awakened appears in the historical login data to obtain the user account portrait label.

[0119] Optionally, the target attribute data includes at least one of device ID, IP address, international mobile subscriber identity, international mobile equipment identity, geographic information, and login time.

[0120] Optionally, the encoding model is a GBDT model, and the classification model is a Logistic model.

[0121] Alternatively, if Figure 5 As shown, the device 300 for secondary number release based on user portrait identification also includes:

[0122] Verification module 340, used to verify user identity information if the trained classification model classifies the mobile phone number used by the user to log in to the account as a mobile phone number that is released by the communication operator for the second time;

[0123] The processing module 350 is used to assign a virtual user ID to the account to seal the account if the verification fails.

[0124] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0125] The present disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method steps of secondary number placement based on user portrait identification as described in any of the above optional embodiments.

[0126] The present disclosure also provides a device for secondary number release based on user portrait identification, including:

[0127] a memory having a computer program stored thereon; and

[0128] A processor is used to execute the computer program in the memory to implement the above-mentioned method steps for secondary number release based on user portrait identification.

[0129] Figure 6 FIG. 4 is a block diagram of an apparatus 400 for secondary number placement based on user portrait identification according to an exemplary embodiment. Figure 6 As shown, the apparatus 400 may include: a processor 401 , a memory 402 , a multimedia component 403 , an input / output (I / O) interface 404 , and a communication component 405 .

[0130] Among them, the processor 401 is used to control the overall operation of the device 400 to complete all or part of the steps in the above-mentioned method for secondary number release based on user portrait identification. The memory 402 is used to store various types of data to support the operation of the device 400, and these data may include, for example, instructions for any application or method used to operate on the device 400, and application-related data. The memory 402 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 403 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 402 or sent through the communication component 405. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 404 provides an interface between the processor 401 and other interface modules, and the above-mentioned other interface modules may be a keyboard, a mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 405 is used for wired or wireless communication between the device 400 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 405 may include: Wi-Fi module, Bluetooth module, NFC module.

[0131] In an exemplary embodiment, the device 400 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned method of secondary number release based on user portrait recognition.

[0132] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, such as a memory 402 including program instructions. The above program instructions can be executed by the processor 401 of the device 400 to complete the above method of secondary number release based on user portrait recognition.

[0133] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.

[0134] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0135] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. A method for identifying secondary number release, characterized in that: The method comprises: Build a portrait tag based on the historical login data of the user's login account; According to the constructed portrait labels, the encoding model is trained and the portrait labels are encoded using the trained encoding model; According to the encoding completed by the encoding model, the classification model is trained so that the trained classification model can classify the mobile phone number used by the user to log in to the account; The step of constructing a portrait tag based on historical login data of the user's login account includes: Calculate the Gini index of different attribute data in the historical login data; Determining target attribute data according to the size of the Gini index; Constructing a user profile according to the target attribute data; Filter out the login data of the account when it wakes up after being silent for more than a preset time from the data of the user portrait; Determine whether the login data when the account is awakened appears in the historical login data to obtain a user account portrait tag.

2. The method according to claim 1, characterized in that The target attribute data includes login time, and at least one of device ID, IP address, international mobile user identity code, international mobile equipment identity code, and geographic information.

3. The method according to claim 1, characterized in that The encoding model is a GBDT model.

4. The method according to any one of claims 1 to 3, characterized in that Also includes: If the mobile phone number used by the user to log in to the account is classified as a mobile phone number that has been re-released by a telecommunications operator, verify the user's identity information; If the verification fails, a virtual user ID is assigned to the account to seal the account.

5. A device for identifying secondary number release, characterized in that: The device comprises: A construction module is used to construct a portrait label based on the historical login data of the user's login account; A first training module is used to train a coding model according to the constructed portrait labels and encode the portrait labels using the trained coding model; A second training module is used to train a classification model according to the encoding completed by the encoding model so that the trained classification model can classify the mobile phone number used by the user to log in to the account; The building blocks include: A calculation submodule, used for calculating the Gini index of different attribute data in the historical login data; A determination submodule, used to determine target attribute data according to the size of the Gini index; A construction submodule, used to construct a user portrait according to the target attribute data; A screening submodule, for screening out the login data of accounts awakened after being silent for more than a preset time from the data of the user portrait; The acquisition submodule is used to determine whether the login data when the account is awakened appears in the historical login data to obtain the user account portrait label.

6. The device according to claim 5, characterized in that The target attribute data includes login time, and at least one of device ID, IP address, international mobile user identity code, international mobile equipment identity code, and geographic information.

7. The device according to claim 5, characterized in that The encoding model is a GBDT model.

8. The device according to any one of claims 5 to 7, characterized in that Also includes: A verification module, used to verify the user's identity information if the mobile phone number used by the user to log in to the account is classified as a mobile phone number that has been re-released by a communication operator; The processing module is used for assigning a virtual user ID to the account to seal the account if the verification fails.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

10. A device for secondary number release based on user portrait recognition, characterized in that: include: a memory having a computer program stored thereon; as well as A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 4.

Images