Method, device, electronic device and storage medium for identifying vehicle owners

Abstract

This application provides a method, device, electronic device, and storage medium for identifying vehicle owners. By acquiring a user's telecommunications business behavior data, internet behavior data, and attribute information, and then inputting this data to be predicted into a pre-trained prediction model, the corresponding user type, either a vehicle owner or a vehicle-free user, is determined and stored. The prediction model, trained on user telecommunications network data with broad coverage and comprehensive features, achieves improved accuracy, making the prediction model more accurate in predicting vehicle owners. This allows vehicle service providers to provide both telecommunications-related services and vehicle-related rights services, effectively serving vehicle owners while avoiding harassment of other users.

Description

Technical Field

[0001] The present application relates to the field of data analysis technology, and in particular to a method, device, electronic device, and storage medium for identifying vehicle owners. Background Art

[0002] The goal of data analysis is to gather and extract information hidden within seemingly disorganized data, thereby uncovering the underlying patterns of the subject under study. In practical applications, data analysis can help people make informed decisions and take appropriate action. By leveraging existing data to identify car owners, targeted services can be provided to these users.

[0003] In the existing technology, the method of predicting car-owning users is usually to obtain customer data from online ride-hailing platforms, such as travel frequency, travel distance, travel type and basic information, and then train and predict through machine learning algorithms to obtain the result of whether the user has a car.

[0004] However, in the above prediction method, the user group of the data sample is small, which makes the training data set small and leads to low accuracy of the prediction results. Summary of the Invention

[0005] The present application provides a method, device, electronic device and storage medium for identifying car users, which are used to solve the problem in the prior art that the training data set is small due to the small data sample size, resulting in low accuracy of prediction results.

[0006] In a first aspect, the present application provides a method for identifying a user who owns a vehicle, comprising:

[0007] Acquiring data to be predicted, the data to be predicted including telecommunications business behavior data, internet behavior data, and attribute information of the user, wherein the telecommunications business behavior data and the internet behavior data are both vehicle-related data;

[0008] Inputting the data to be predicted into a pre-trained prediction model to obtain a user type corresponding to the user, wherein the user type is a car user or a non-car user, and the prediction model is a user-based collaborative filtering model or a content-based collaborative filtering model;

[0009] The user type corresponding to the user is saved.

[0010] Optionally, also include:

[0011] Acquire training data, where the training data includes telecommunication network data of a user and a user type label, where the user type label is a car user or a non-car user;

[0012] The prediction model is obtained by training a preset model using the training data, and the preset model is a user-based collaborative filtering model or a content-based collaborative filtering model.

[0013] Optionally, obtaining training data includes:

[0014] Acquiring telecommunication network data of a traffic management system, the telecommunication network data including user attribute information and text message information sent to the user by the traffic management system;

[0015] Users who have received SMS messages from the traffic management system are marked as car owners.

[0016] Optionally, the prediction model is a user-based collaborative filtering model, and inputting the data to be predicted into a pre-trained prediction model to obtain a user type corresponding to the user includes:

[0017] The telecommunications service behavior data, the Internet behavior data, and the attribute information are each used as a feature;

[0018] Perform one-hot encoding on each feature and construct a multidimensional vector based on the encoding corresponding to each feature of the user;

[0019] Calculating the similarity between the multidimensional vector corresponding to the data to be predicted and each multidimensional vector in the training data, wherein each user corresponds to a multidimensional vector;

[0020] Determine whether the user is a car owner based on the similarity.

[0021] Optionally, the prediction model is a content-based collaborative filtering model, and inputting the data to be predicted into a pre-trained prediction model to obtain a user type corresponding to the user includes:

[0022] Taking the telecommunications service behavior data and the internet behavior data as a feature respectively;

[0023] Determine the frequency of occurrence of each feature;

[0024] One-hot encoding is performed on the attribute information to obtain a code corresponding to the attribute information;

[0025] Constructing a multidimensional vector according to the frequency of occurrence of each feature and the code;

[0026] Calculating the similarity between the multidimensional vector corresponding to the data to be predicted and each multidimensional vector in the training data, wherein each user corresponds to a multidimensional vector;

[0027] Determine whether the user is a car owner based on the similarity.

[0028] Optionally, the telecommunications service behavior data includes the user's traffic call records, traffic text message records, and location changes per unit time;

[0029] The Internet behavior data includes built-in information of vehicle-related applications used by users and vehicle browsing records on shopping websites. The built-in information is the vehicle-related information set by the user in the application;

[0030] The attribute information includes the user's real-name information, telecommunications service consumption limit information, whether broadband services are used, and whether vehicle-related value-added services are subscribed.

[0031] In a second aspect, the present application provides a device for identifying a user who owns a vehicle, comprising:

[0032] A first acquisition module is configured to acquire data to be predicted, wherein the data to be predicted includes telecommunications business behavior data, internet behavior data, and attribute information of the user, wherein both the telecommunications business behavior data and the internet behavior data are vehicle-related data;

[0033] a processing module, configured to input the data to be predicted into a pre-trained prediction model to obtain a user type corresponding to the user, wherein the user type is a car user or a non-car user, and the prediction model is a user-based collaborative filtering model or a content-based collaborative filtering model;

[0034] The saving module is used to save the user type corresponding to the user.

[0035] Optionally, also include:

[0036] A second acquisition module is used to acquire training data, wherein the training data includes the user's telecommunications network data and a user type label, wherein the user type label is a car user or a non-car user;

[0037] The training module is used to train a preset model using the training data to obtain the prediction model, where the preset model is a user-based collaborative filtering model or a content-based collaborative filtering model.

[0038] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0039] The memory stores computer-executable instructions;

[0040] The processor executes the computer-executable instructions stored in the memory to implement the method described in the first aspect.

[0041] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in the first aspect.

[0042] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which implements the method described in the first aspect when executed by a processor.

[0043] This application provides a method, device, electronic device, and storage medium for identifying vehicle owners. By acquiring a user's telecommunications business behavior data, internet behavior data, and attribute information, and then inputting this data to be predicted into a pre-trained prediction model, the corresponding user type, either a vehicle owner or a vehicle-free user, is determined and stored. The prediction model, trained on user telecommunications network data with broad coverage and comprehensive features, achieves improved accuracy, making the prediction model more accurate in predicting vehicle owners. This allows vehicle service providers to provide both telecommunications-related services and vehicle-related rights services, effectively serving vehicle owners while avoiding harassment of other users. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0045] Figure 1 A schematic diagram of an application scenario applicable to this application;

[0046] Figure 2 A flowchart of a method for identifying a user with a car provided in Example 1 of the present application;

[0047] Figure 3 A schematic diagram of the structure of a device for identifying a user with a car provided in Example 2 of the present application;

[0048] Figure 4 This is a schematic diagram of the structure of a device for identifying a user with a car provided in the third embodiment of the present invention.

[0049] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0050] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0051] By leveraging existing data to identify car owners, targeted services can be provided to these users. Existing methods for predicting car owners typically involve obtaining customer data from ride-hailing platforms, such as trip frequency, distance, type, and basic information. Machine learning algorithms are then trained and used to predict the user's type, indicating whether the user owns a car.

[0052] However, the aforementioned prediction methods rely on a smaller user base for data samples, resulting in a smaller training dataset and lower prediction accuracy. Furthermore, due to the limitations of services offered by platforms like ride-hailing, the training data obtained through these platforms has fewer features, which also impacts the accuracy of predictions.

[0053] Therefore, this application provides a method, apparatus, electronic device, and storage medium for identifying vehicle owners. This improves the accuracy of the prediction model by training a model using user telecommunications network data with broad coverage and relatively comprehensive features. This prediction model is then used to predict the user's type, resulting in more accurate predictions. This allows vehicle service providers to provide users with telecommunications-related services while also accurately providing vehicle-related rights services, effectively serving vehicle owners while avoiding harassing other users.

[0054] refer to Figure 1 , Figure 1This is a schematic diagram of an application scenario applicable to the present application, in which information is exchanged between the prediction server 102, the traffic management system 103 and the terminal 101 via the Internet. The prediction server 102 obtains the telecommunications network data of the traffic management system 103, and calibrates the user labels of the users of the telecommunications network data as car users. Then, the telecommunications network data and user labels are used as training data sets to train the preset model to obtain a prediction model. The prediction server 102 then obtains the data to be predicted from the terminal 101, and inputs the data to be predicted into the prediction model to obtain the user type corresponding to the user. The data to be predicted includes the user's telecommunications business behavior data, Internet behavior data and attribute information. After obtaining the user type of the user, the prediction server 102 saves it so that the vehicle service provider can provide targeted services to car users and non-car users. It can be understood that the number of prediction servers 102, traffic management systems 103 and terminals 101 can be multiple, which is not shown in the figure.

[0055] In addition, the embodiment of the present application does not limit the type of terminal 101. Terminal 101 can be an electronic device with call function and / or network function, such as a mobile phone, a tablet computer, and a desktop computer.

[0056] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can exist independently or in combination with each other. For the same or similar concepts or processes, they may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0057] refer to Figure 2 , Figure 2 This is a flow chart of a method for identifying a car user provided in Example 1 of the present application. The method can be performed by Figure 1 The prediction server in the method is executed, hereinafter referred to as the server, and the method includes the following steps.

[0058] S201. Obtain data to be predicted, where the data to be predicted includes the user's telecommunications business behavior data, Internet behavior data, and attribute information.

[0059] The server may obtain the data to be predicted from the user terminal, where the data to be predicted includes the user's telecommunications service behavior data, Internet behavior data, and attribute information.

[0060] Both telecommunications service behavior data and internet behavior data are vehicle-related data. Specifically, telecommunications service behavior data includes the user's traffic call records, traffic text message records, and location change per unit time. For example, the unit time can be 15 minutes, and the location change can be expressed as the distance the user moves within 15 minutes. Traffic call records are records of calls between the user and traffic management departments, including the time and call number. Traffic text message records are detailed text messages sent by the user to traffic management departments, such as violation reminders.

[0061] Internet behavior data includes information built into the applications used by users and vehicle browsing history on shopping websites. This built-in information is vehicle-related information set by users within the application. For example, this information includes whether the user is a member. When the application is a vehicle-related application, such as a vehicle maintenance application or a vehicle shopping mall application, the built-in information also includes basic vehicle information entered by the user within the application, such as the model. Browsing history on shopping websites includes vehicle-related web pages browsed by users, such as visits to items needed for vehicle interior decoration. This internet behavior can also include the number of visits to gas stations and the number of stops at bus stops.

[0062] Attribute information includes the user's real-name information, telecommunications service consumption limit information, whether broadband services are used, and whether vehicle-related value-added services are subscribed. For example, real-name information includes the user's name, occupation, age, gender, income, and address.

[0063] Optionally, the server can set up vehicle-related tracking points, which are used to collect user attribute information, telecommunications business behavior data, and internet behavior data. Specifically, the server sets tracking events and tracking event identifiers. When a tracking event is triggered for a user, the server can receive the corresponding message. The tracking event is a vehicle-related event, such as a user receiving a traffic SMS from the Traffic Management Bureau or a user searching for vehicle accessories on a website. When these tracking events occur, the server can obtain the corresponding message data.

[0064] S202: Input the data to be predicted into a pre-trained prediction model to obtain the user type corresponding to the user.

[0065] After the server obtains the data to be predicted, it inputs the data to be predicted into the pre-trained prediction model to obtain the user type corresponding to the user, that is, car users and non-car users. The prediction model is a user-based collaborative filtering model or a content-based collaborative filtering model.

[0066] The above prediction model is obtained by the server training a preset model based on the acquired training data. The training data includes the user's telecommunications network data and user type label. The user type label is a car user, and the user type label is used to distinguish the user type of the user.

[0067] Specifically, the training data is telecommunications network data acquired by the server from the traffic management system. This data includes user attributes and text messages sent to users by the traffic management system. Similarly, the text messages contain detailed text messages received by users from traffic management departments. The server then identifies users who have received text messages from the traffic management system as car owners. This training data not only has a large sample user base but also contains a wide range of features, which facilitates the training of the pre-set model and improves the accuracy of the prediction model.

[0068] It should be noted that the above-mentioned preset model is a user-based collaborative filtering model or a content-based collaborative filtering model, and the above-mentioned prediction model can be obtained by training both types of models.

[0069] The basic principles of the prediction model obtained by training these two types of models are explained below.

[0070] (1) User-based collaborative filtering model.

[0071] According to the prediction model obtained by training this type of model, when predicting the data to be predicted, the telecommunications business behavior data, Internet behavior data and attribute information in the data to be predicted are used as a feature respectively.

[0072] Each feature is then One-Hot encoded, and a multidimensional vector is constructed based on the encoding corresponding to each feature of the user. The values in the multidimensional vector are the corresponding encodings of the above features. The model then calculates the similarity between the multidimensional vector corresponding to the predicted data and the multidimensional vectors in the training data, where each user corresponds to a multidimensional vector. For example, the similarity can be calculated using the following formula:

[0073]

[0074] Among them, A i is the feature vector of user A, B i is the feature vector of user B, θ is the feature vector of A i and eigenvector B i The cosine value is the angle between the vectors in the multidimensional space of [-1, 1]. The closer the cosine value is to 1, the smaller the vector angle is, indicating a higher correlation between the users. In this embodiment of the present application, for example, when the cosine value is within the preset range of [0.8, 1], it can be considered that user A and user B are similar.

[0075] After calculating the similarity, it can be determined whether the user is a car owner based on the similarity. Specifically, when the similarity between the multidimensional vector corresponding to the predicted data and each multidimensional vector in the training data is within [0.8, 1], it can be determined that the user corresponding to the multidimensional vector is a car owner.

[0076] Alternatively, if the similarity between a preset proportion or a preset number of multidimensional vectors in the training data and the multidimensional vector corresponding to the data to be predicted is within a preset range, then the user corresponding to the multidimensional vector is determined to be a car owner. For example, if the similarity between more than 80% of the multidimensional vectors in the training data and user A in the data to be predicted is within a preset range, then it can be determined that user A is a car owner.

[0077] By predicting user types based on a user-based collaborative filtering model and calculating similarity using a multidimensional vector that describes the user's characteristics, the accuracy of predicting user types is improved.

[0078] (2) Content-based collaborative filtering model.

[0079] According to the prediction model obtained by training this type of model, when predicting the data to be predicted, the telecommunications business behavior data and Internet behavior data in the data to be predicted are used as a feature respectively.

[0080] The frequency of occurrence of each feature of the telecommunications business behavior data and the internet behavior data is then determined, and the attribute information is one-hot encoded to obtain the corresponding code of the attribute information. The model then constructs a multidimensional vector based on the frequency and code. The values in the multidimensional vector are the corresponding frequencies and codes of the above-mentioned features. The model then calculates the similarity between the multidimensional vector corresponding to the predicted data and the multidimensional vectors in the training data, where each user corresponds to a multidimensional vector. Similarly, the similarity can be calculated according to formula (1).

[0081] After calculating the similarity, we can use it to determine whether the user owns a car. For more details, refer to the user-based collaborative filtering model described above. Using the content-based collaborative filtering model to predict user types, the multidimensional vector used to calculate similarity describes the frequency of occurrence of certain features within a user's set of characteristics, making the prediction of user types more accurate.

[0082] It should be noted that the embodiments of the present application do not limit the above-mentioned method for calculating similarity. Similarity can also be calculated using the definition of the Jaccard coefficient, or methods such as the Pearson coefficient and the Euler distance. Furthermore, the above-mentioned method for determining whether a user is a car owner based on similarity can also be to calculate the similarity between the multidimensional vector corresponding to the data to be predicted and a multidimensional vector in the training data. When the similarity is within a preset range, the user can be determined to be a car owner. The embodiments of the present application limit the method for determining whether a user is a car owner based on similarity.

[0083] The loss function is usually used to express the degree of difference between the predicted data and the actual data. When the server uses the training data to train the above two types of models, it can determine the quality of the resulting prediction model by judging the size of the loss function. That is, the smaller the loss function, the higher the accuracy of the prediction model. For example, a loss function in the embodiment of the present application is defined as follows:

[0084]

[0085] in, is the loss function, i is the user number, j is the user's telecommunications business behavior number, Represents user attributes, Indicates Internet behavior, y (i，j) is the frequency of user i performing telecommunication service behavior j, and λ is the regularization parameter.

[0086] S203: Save the user type corresponding to the user.

[0087] After the server obtains the user type of the user, it saves the user type. The server can output the user type, so that the vehicle service provider can provide vehicle-related rights and interests services accurately while providing telecommunications-related services to car users, and avoid harassing other users while providing good user services to car owners.

[0088] In this embodiment, the server obtains the user's telecommunications business behavior data, internet behavior data, and attribute information, then inputs this data into a pre-trained prediction model to determine the user's corresponding user type, either a car owner or a non-car owner, and saves this user type. The prediction model, trained on user telecommunications network data with broad coverage and comprehensive features, has improved accuracy, making it more accurate in predicting car owners. This allows vehicle service providers to provide users with both telecommunications services and vehicle-related benefits, effectively serving car owners while avoiding harassment of other users.

[0089] refer to Figure 3 , Figure 3This is a structural diagram of a device for identifying a user with a car provided in Example 2 of the present application. The device 30 includes: an acquisition module 301, a processing module 302 and a storage module 304.

[0090] The first acquisition module 301 is used to acquire data to be predicted, which includes the user's telecommunications business behavior data, Internet behavior data and attribute information. The telecommunications business behavior data and Internet behavior data are both vehicle-related data.

[0091] The processing module 302 is used to input the data to be predicted into the pre-trained prediction model to obtain the user type corresponding to the user, where the user type is a car user or a non-car user, and the prediction model is a user-based collaborative filtering model or a content-based collaborative filtering model.

[0092] The saving module 303 is used to save the user type corresponding to the user.

[0093] Optionally, also include:

[0094] The second acquisition module is used to acquire training data. The training data includes the user's telecommunications network data and user type label. The user type label is used to distinguish the user type of the user. The user type label is a car user.

[0095] The training module is used to train a preset model through training data to obtain a prediction model. The preset model is a user-based collaborative filtering model or a content-based collaborative filtering model.

[0096] Optionally, the second acquisition module is specifically configured to:

[0097] Obtain telecommunication network data of the traffic management system, which includes user attribute information and text message information sent to users by the traffic management system.

[0098] Users who have received SMS messages from the traffic management system are identified as car owners.

[0099] Optionally, the prediction model is a user-based collaborative filtering model, and the processing module 302 is specifically configured to:

[0100] Telecommunications business behavior data, Internet behavior data and attribute information are respectively regarded as a feature.

[0101] Each feature is encoded with one-hot encoding, and a multidimensional vector is constructed based on the encoding corresponding to each feature of the user.

[0102] Calculate the similarity between the multidimensional vector corresponding to the data to be predicted and each multidimensional vector in the training data, where each user corresponds to a multidimensional vector.

[0103] Determine whether the user owns a car based on the similarity.

[0104] Optionally, the prediction model is a content-based collaborative filtering model, and the processing module 302 is further configured to:

[0105] Telecommunications business behavior data and Internet behavior data are each considered as a feature.

[0106] Determine how often each feature occurs.

[0107] The attribute information is One-Hot encoded to obtain a code corresponding to the attribute information.

[0108] Construct a multidimensional vector based on the frequency and coding of each feature.

[0109] Calculate the similarity between the multidimensional vector corresponding to the data to be predicted and each multidimensional vector in the training data, where each user corresponds to a multidimensional vector.

[0110] Determine whether the user owns a car based on the similarity.

[0111] Optionally, the telecommunications business behavior data includes the user's traffic call records, traffic text message records, and location changes per unit time.

[0112] Internet behavior data includes the built-in information of vehicle-related applications used by users and vehicle browsing records on shopping websites. The built-in information is the vehicle-related information set by the user in the application.

[0113] Attribute information includes the user's real-name information, telecommunications service consumption limit information, whether broadband services are used, and whether vehicle-related value-added services are subscribed.

[0114] The device of this embodiment can be used to execute the steps of a method for identifying a user with a car in the first embodiment. The specific implementation method and technical effects are similar and will not be repeated here.

[0115] refer to Figure 4 , Figure 4 This is a schematic diagram of a structure of a device for identifying a user with a car provided in the third embodiment of the present invention. Figure 4 As shown, the device 40 includes: a processor 401, a memory 402, and a transceiver 403. The memory 402 is used to store instructions, the transceiver 403 is used to communicate with other devices, and the processor 401 is used to execute instructions stored in the memory so that the device 40 performs a method step for identifying a car user as in Example 1. The specific implementation method and technical effect are similar and will not be repeated here.

[0116] Embodiment 4 of the present invention provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement a method and steps for identifying a user with a car as in the above-mentioned embodiment 1. The specific implementation method and technical effect are similar and will not be repeated here.

[0117] Embodiment 5 of the present invention provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the method steps for identifying a user with a car as in the above-mentioned embodiment 1. The specific implementation method and technical effects are similar and will not be repeated here.

[0118] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0119] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A method for identifying a car user, characterized in that: include: Obtaining data to be predicted, the data to be predicted including telecommunications business behavior data, internet behavior data, and attribute information of the user, wherein the telecommunications business behavior data and the internet behavior data are both vehicle-related data; the telecommunications business behavior data includes the user's traffic call records, traffic text message records, and location change per unit time; Inputting the data to be predicted into a pre-trained prediction model to obtain a user type corresponding to the user, wherein the user type is a car user or a non-car user; Saving the user type corresponding to the user; The prediction model is a content-based collaborative filtering model. Inputting the data to be predicted into the pre-trained prediction model to obtain the user type corresponding to the user includes: The telecommunications service behavior data and the internet behavior data are each used as a feature; the frequency of occurrence of each feature is determined; the attribute information is one-hot encoded to obtain a code corresponding to the attribute information; a multidimensional vector is constructed based on the frequency of occurrence of each feature and the code; and the similarity between the multidimensional vector corresponding to the data to be predicted and each multidimensional vector in the training data is calculated, wherein each user corresponds to a multidimensional vector; the users corresponding to the training data are car owners; If the similarity between a preset proportion or a preset number of multidimensional vectors in the training data and the multidimensional vector corresponding to the data to be predicted is within a preset range, the user is determined to be a car owner.

2. The method according to claim 1, characterized in that Also includes: Acquire training data, where the training data includes telecommunication network data of the user and a user type label, where the user type label is a car-owning user; The prediction model is obtained by training a preset model using the training data, and the preset model is a content-based collaborative filtering model.

3. The method according to claim 2, characterized in that The obtaining of training data includes: Acquiring telecommunication network data of a traffic management system, the telecommunication network data including user attribute information and text message information sent to the user by the traffic management system; Users who have received SMS messages from the traffic management system are marked as car owners.

4. The method according to any one of claims 1 to 3, characterized in that The Internet behavior data includes built-in information of vehicle-related applications used by users and vehicle browsing records on shopping websites. The built-in information is the vehicle-related information set by the user in the application; The attribute information includes the user's real-name information, telecommunications service consumption limit information, whether broadband services are used, and whether vehicle-related value-added services are subscribed.

5. A device for identifying a user who owns a car, characterized in that: include: A first acquisition module is configured to acquire data to be predicted, wherein the data to be predicted includes the user's telecommunications business behavior data, internet behavior data, and attribute information, wherein the telecommunications business behavior data and the internet behavior data are both vehicle-related data; the telecommunications business behavior data includes the user's traffic call records, traffic text message records, and location change per unit time; A processing module, configured to input the data to be predicted into a pre-trained prediction model to obtain a user type corresponding to the user, wherein the user type is a car user or a non-car user; A saving module, used for saving the user type corresponding to the user; The processing module is further configured to: use the telecommunications service behavior data and the Internet behavior data as a feature respectively; Determine the frequency of occurrence of each feature; perform one-hot encoding on the attribute information to obtain a code corresponding to the attribute information; construct a multidimensional vector based on the frequency of occurrence of each feature and the code; calculate the similarity between the multidimensional vector corresponding to the data to be predicted and each multidimensional vector in the training data; If the similarity between a preset proportion or a preset number of multidimensional vectors in the training data and the multidimensional vector corresponding to the data to be predicted is within a preset range, the user is determined to be a car owner, where each user corresponds to a multidimensional vector and the user corresponding to the training data is a car owner.

6. The device according to claim 5, characterized in that Also includes: A second acquisition module is used to acquire training data, wherein the training data includes the user's telecommunications network data and a user type label, wherein the user type label is a car user or a non-car user; The training module is used to train a preset model using the training data to obtain the prediction model, where the preset model is a content-based collaborative filtering model.

7. An electronic device, characterized in that: include: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 4.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 4 when executed by a processor.

9. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 4 when the computer program is executed by a processor.

Images