Anti-fraud method, device and equipment based on car networking card, medium and product

Abstract

The invention discloses an anti-fraud method, device and equipment based on an Internet of Vehicles card, a medium and a product. The method comprises the following steps: acquiring Internet of Vehicles card data and mobile phone network card data of the same user; performing feature extraction on the car networking card data and the mobile phone network card data to obtain feature data; wherein the feature data comprises first feature data used for indicating time-space consistency of the car networking card data and the mobile phone network card data; based on the feature data, performing fraud detection by adopting a preset fraud detection model to obtain a fraud detection result; wherein the fraud detection model is obtained based on historical feature data training, and the historical feature data comprises historical first feature data. According to the invention, whether the actual user of the vehicle machine is consistent with the submitted identity information for real name or not can be perceived at the operator level, and the fraud risk of the real name of the vehicle networking card is reduced.

Description

Technical Field

[0001] The present application relates to the field of risk control technology for Internet of Vehicles cards, and in particular to an anti-fraud method, device, equipment, medium and product based on Internet of Vehicles cards. Background Art

[0002] The Internet of Vehicles card refers to a special IoT card installed in road motor vehicles to realize communication service functions such as vehicle networking, data transmission, status monitoring, emergency response, etc. Different from ordinary IoT cards, the Internet of Vehicles card has strong personal attributes. Its real-name registration can prevent and combat various illegal and criminal activities from the source, which is of great significance.

[0003] At present, fraud risk prevention and control is usually carried out by re-verifying the real name of the Internet of Vehicles card. Specifically, the information re-entered by the user is verified through manual verification or AI model to verify whether the user is consistent with the identity information submitted for real name, and then confirm whether there is a fraud risk. However, the existing method can only identify whether it is consistent at the user level, but it cannot be perceived at the operator level. There is a problem that it cannot prevent the actual user from stealing other people's identities for real name, which has a greater risk of fraud. Therefore, an anti-fraud method based on the Internet of Vehicles card is urgently needed. Summary of the invention

[0004] In order to solve the above technical problems, the present application provides an anti-fraud method, device, equipment, medium and product based on the Internet of Vehicles card, which can sense at the operator level whether the actual user of the vehicle machine is consistent with the identity information submitted for real-name registration, thereby reducing the fraud risk of the Internet of Vehicles card real-name registration.

[0005] The embodiment of the present application provides an anti-fraud method based on an Internet of Vehicles card, comprising:

[0006] Get the car network card data and mobile phone network card data of the same user;

[0007] Extracting features from the vehicle network card data and the mobile phone network card data to obtain feature data; wherein the feature data includes: first feature data, the first feature data is used to indicate the spatiotemporal consistency of the vehicle network card data and the mobile phone network card data;

[0008] Based on the feature data, a preset fraud detection model is used to perform fraud detection to obtain a fraud detection result; wherein the fraud detection model is trained based on historical feature data, and the historical feature data includes historical first feature data.

[0009] As an improvement of the above solution, the vehicle network card data includes vehicle network card base station data, and the mobile phone network card data includes mobile phone network card base station data;

[0010] The extracting features of the vehicle network card data and the mobile phone network card data to obtain feature data includes:

[0011] According to the vehicle network card base station data and the mobile network card base station data, obtaining the vehicle network card base station set and the mobile network card base station set of the user in the same time period;

[0012] According to the vehicle network card base station set and the mobile phone network card base station set, the trajectory overlap degree of the user's vehicle network card and mobile phone network card in the same time period is calculated to obtain the first feature data.

[0013] As an improvement of the above solution, the calculation of the trajectory overlap of the user's vehicle network card and mobile network card in the same time period based on the vehicle network card base station set and the mobile network card base station set to obtain the first feature data includes:

[0014] The similarity between the vehicle network card base station set and the mobile phone network card base station set is calculated to obtain the trajectory overlap of the user's vehicle network card and mobile phone network card in the same time period, and use the trajectory overlap as the first feature data.

[0015] As an improvement of the above solution, the Internet of Vehicles card data also includes: Internet of Vehicles card user data, device data of the device used by the user in the real-name process and the generated behavior data; the mobile network card data also includes: mobile network card user data;

[0016] The characteristic data also includes: second characteristic data and third characteristic data, wherein the second characteristic data is used to indicate the behavioral characteristics of the user during the real-name process, and the third characteristic data is used to indicate the real-name consistency between the vehicle network card data and the mobile phone network card data.

[0017] As an improvement of the above solution, the feature extraction of the vehicle network card data and the mobile phone network card data to obtain feature data further includes:

[0018] Extracting the second feature data in chronological order based on the device data and the behavior data;

[0019] Based on the Internet of Vehicles card user data and the mobile network card user data, extracting Internet of Vehicles card user characteristics and mobile network card user characteristics;

[0020] The similarity between the user characteristics of the Internet of Vehicles card and the user characteristics of the mobile phone network card is calculated to obtain the third characteristic data.

[0021] As an improvement of the above solution, the fraud detection model is obtained by fusing the first sub-model, the second sub-model and the third sub-model;

[0022] Among them, the first sub-model is trained based on the historical first feature data; the second sub-model is trained based on the historical second feature data; and the third sub-model is trained based on the historical third feature data.

[0023] The embodiment of the present application also provides an anti-fraud device based on a vehicle networking card, including:

[0024] A data acquisition module, used to acquire the car network card data and mobile phone network card data of the same user;

[0025] A feature extraction module is used to extract features from the vehicle network card data and the mobile phone network card data to obtain feature data; wherein the feature data includes: first feature data, the first feature data is used to indicate the temporal and spatial consistency of the vehicle network card data and the mobile phone network card data;

[0026] A fraud detection module is used to perform fraud detection based on the feature data using a preset fraud detection model to obtain a fraud detection result; wherein the fraud detection model is trained based on historical feature data, and the historical feature data includes historical first feature data.

[0027] An embodiment of the present application also provides an anti-fraud device based on a vehicle networking card, comprising a processor and a memory, wherein a computer program is stored in the memory, and the computer program is configured to be executed by the processor, and when the processor executes the computer program, the anti-fraud method based on the vehicle networking card described above is implemented.

[0028] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any of the above-mentioned anti-fraud methods based on the Internet of Vehicles card.

[0029] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned anti-fraud methods based on the Internet of Vehicles card.

[0030] Compared with the prior art, the anti-fraud method, device, equipment, medium and product based on the Internet of Vehicles card provided by the embodiments of the present application have the following beneficial effects: by obtaining the Internet of Vehicles card data and the mobile network card data corresponding to the same user, and performing feature extraction to obtain feature data including the time-space consistency indicating the Internet of Vehicles card and the mobile network card, the feature data is input into a preset fraud detection model to obtain a fraud detection result, which can identify the real-name accuracy of the Internet of Vehicles card and the safety of the card usage behavior, confirm the consistency of the person and the certificate of the Internet of Vehicles card, and realize the risk prevention and control of real-name information fraud of the Internet of Vehicles card at the operator level;

[0031] Through the behavior data generated by the user during the real-name registration process on the Internet of Vehicles card real-name registration platform and the device data of the device used, the behavior characteristics indicating the user during the real-name registration process are extracted to perform fraud detection. It can be predicted whether the user skips the liveness detection during the real-name registration process and whether the identity is collected through normal channels, thereby improving the accuracy of fraud detection;

[0032] By extracting the features corresponding to the real-name consistency and the spatiotemporal consistency, it is possible to determine whether the real-name information and spatiotemporal data of the same user's Internet of Vehicles card and mobile network card are consistent, and then it is possible to further confirm the identity consistency of the Internet of Vehicles card, thereby improving the comprehensiveness and accuracy of Internet of Vehicles card fraud detection. The embodiment of the present application realizes the real-name verification of Internet of Vehicles cards and the prediction of Internet of Vehicles card fraud risks at the operator level, thereby preventing and controlling the fraud risks caused by false identity information, and improving the comprehensiveness and accuracy of Internet of Vehicles card real-name fraud detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a flowchart of an anti-fraud method based on a vehicle networking card provided in an embodiment of the present application;

[0034] Figure 2 It is a structural schematic diagram of an anti-fraud device based on a vehicle networking card provided in an embodiment of the present application;

[0035] Figure 3 It is a structural schematic diagram of an anti-fraud device based on an Internet of Vehicles card provided in an embodiment of the present application. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0037] The Internet of Vehicles card refers to a special IoT card installed in road motor vehicles to realize communication service functions such as vehicle networking, data transmission, status monitoring, and emergency response. The mobile network card is a mobile data card opened by operators for individuals. Both can communicate and both belong to the card number opened by the operator. They have great similarities in characteristic attributes, but they are different in functional use and contract scenarios. The data of the Internet of Vehicles card is managed by the Internet of Things company, and the data of the mobile network card is managed by the provincial companies of the operator in various places. The customer groups of these two types of companies have a lot of overlap. A user has a mobile network card and may also have an Internet of Vehicles card.

[0038] If a user's mobile network card is located in a certain area in the recent period, and the user's Internet of Vehicles card is also located in the same area recently, then the two maintain data consistency in a specific time and space. In this case, it can be considered that the user is using his real identity for real-name registration, and is thus identified as having the consistency of person and certificate of the Internet of Vehicles card. If a user's mobile network card is often located in area A, but the Internet of Vehicles card is often located in area B, for such inconsistent time and space data cards, it can be determined that there is a lack of consistency of person and certificate, and fraud may occur, such as separation of the card and the device, and theft by others.

[0039] In view of this, an embodiment of the present application provides an anti-fraud method based on an Internet of Vehicles card, which uses the Internet of Vehicles real-name platform built by the Internet of Things company as a data center, exchanges data with various provincial companies through the Internet of Vehicles real-name platform, and then builds a fraud detection model between the Internet of Things company and various provincial companies to achieve fraud detection.

[0040] See also Figure 1 , Figure 1 : is a flow chart of an anti-fraud method based on a vehicle networking card provided in an embodiment of the present application. The anti-fraud method based on a vehicle networking card includes:

[0041] S1: Obtain the car network card data and mobile phone network card data of the same user;

[0042] S2: extracting features from the vehicle network card data and the mobile phone network card data to obtain feature data; wherein the feature data includes: first feature data, the first feature data is used to indicate the spatiotemporal consistency of the vehicle network card data and the mobile phone network card data;

[0043] S3: Based on the feature data, a preset fraud detection model is used to perform fraud detection to obtain a fraud detection result; wherein the fraud detection model is trained based on historical feature data, and the historical feature data includes historical first feature data.

[0044] In step S1, whether the user is the same user can be confirmed through the user ID. Optionally, the user ID is an ID number. For example, a car network card and a mobile phone network card with the same ID number can be identified as belonging to the same user, and the car network card data of the car network card and the mobile phone network card data of the mobile phone network card can be obtained.

[0045] In an optional embodiment, before obtaining the vehicle network card data and the mobile phone network card data of the same user, the method further includes:

[0046] Performing hash processing on the user IDs of several Internet of Vehicles cards to obtain several first hash values;

[0047] receiving a plurality of second hash values, where the plurality of second hash values ​​are obtained by hashing user IDs of a plurality of mobile phone network cards;

[0048] Matching the first hash values ​​with the second hash values ​​to obtain a target first hash value and a target second hash value having a matching relationship;

[0049] The user of the Internet of Vehicles card corresponding to the target first hash value and the user of the mobile phone network card corresponding to the target second hash value are confirmed to be the same user.

[0050] It is understandable that the Internet of Things company and each provincial company hash their local user IDs to obtain the corresponding hash values, and then determine the same user based on the hash values, so that fraud detection can be performed based on the Internet of Vehicles card data and mobile network card data of the same user. In order not to expose data here, the Internet of Vehicles real-name platform and each provincial company hold each other's public keys, and realize data transmission through private key encryption and public key decryption. For example, the RSA asymmetric encryption protocol is used to encrypt and transmit user information (Internet of Vehicles card data / mobile network card data) to ensure the security of user information.

[0051] It is understandable that after confirming the same user, the car network card data and mobile phone network card data of the same user are obtained;

[0052] Among them, the Internet of Vehicles card data is stored in the Internet of Vehicles real-name registration platform and can be directly obtained. The mobile network card data is sent by each provincial company. Furthermore, each provincial company has trained a model based on the mobile network card data locally. The model can learn the original mobile network card data and obtain the mobile network card data in a specified format after learning. The mobile network card data is transmitted or encrypted through the 5G base station, which effectively solves the timeliness and communication overhead required in the fraud detection process. Furthermore, after receiving the mobile network card data, the Internet of Vehicles real-name registration platform splices and aligns the mobile network card data with the Internet of Vehicles card data for feature extraction.

[0053] As one of the optional embodiments, the vehicle network card data includes vehicle network card base station data, and the mobile phone network card data includes mobile phone network card base station data;

[0054] The extracting features of the vehicle network card data and the mobile phone network card data to obtain feature data includes:

[0055] According to the vehicle network card base station data and the mobile network card base station data, obtaining the vehicle network card base station set and the mobile network card base station set of the user in the same time period;

[0056] According to the vehicle network card base station set and the mobile phone network card base station set, the trajectory overlap degree of the user's vehicle network card and mobile phone network card in the same time period is calculated to obtain the first feature data.

[0057] In an embodiment of the present application, the spatiotemporal consistency of the vehicle network card data and the mobile phone network card data can be characterized by the overlap of the trajectories of the vehicle network card and the mobile phone network card in the same time period.

[0058] The vehicle network card base station data at least includes: a first positioning position information and a first positioning time, and the mobile network card base station data at least includes a second positioning position information and a second positioning time, and then based on the vehicle network card base station data and the mobile network card base station data, the vehicle network card base station set and the mobile network card base station set of the user in the same time period are obtained, wherein the vehicle network card base station set includes a number of first positioning position information, and the mobile network card base station set includes a number of second positioning position information; then, based on the vehicle network card base station set and the mobile network card base station set, the trajectory overlap of the vehicle network card and the mobile network card in the same time period is calculated to obtain the first feature data.

[0059] For example, when the base station to which the mobile network card or the Internet of Vehicles card is connected changes, it means that the position of the mobile network card or the Internet of Vehicles card has changed. The positioning information of the base station to which it is connected at each moment can be obtained from the base station data of the mobile network card or the Internet of Vehicles card, thereby forming the trajectory timing information of the mobile network card or the Internet of Vehicles card. These trajectory timing information constitute the corresponding base station set, and then the trajectory overlap of the Internet of Vehicles card and the mobile network card in the same time period is calculated based on the base station set.

[0060] Optionally, the first positioning location information includes at least one of GPS information, base station positioning information, Mac address, WIFI information and network registration location; the second positioning location information includes at least one of GPS information, base station positioning information, Mac address, WIFI information and network registration location. Through this information, the location where the vehicle network card or mobile network card has passed can be confirmed, and then the trajectory within a period of time can be obtained.

[0061] As one of the optional embodiments, the calculating the trajectory overlap of the user's vehicle network card and mobile network card in the same time period according to the vehicle network card base station set and the mobile network card base station set to obtain the first feature data includes:

[0062] The similarity between the vehicle network card base station set and the mobile phone network card base station set is calculated to obtain the trajectory overlap of the user's vehicle network card and mobile phone network card in the same time period, and use the trajectory overlap as the first feature data.

[0063] It is understandable that since the base station data of the mobile network card and the Internet of Vehicles card are massive discrete data, when calculating the similarity between the Internet of Vehicles card base station set and the mobile network card base station set, it is preferred to use MinHash (minimum hash) to calculate the Jaccard similarity to measure the set similarity. If the Internet of Vehicles card and the mobile network card of the same real-name user have common base station information in the same time period, then the corresponding Jaccard similarity is higher. For massive information, the use of MinHash dimensionality reduction can find most of the similar set pairs in a short time when there is less similar set information to extract the first feature data.

[0064] Specifically, the specific process of using MinHash to calculate Jaccard similarity includes:

[0065] The data in the base station set is converted into corresponding characteristic rows. The obtained characteristic rows form a corresponding matrix. The rows of the matrix represent each element in the set, and the columns of the matrix represent each set.

[0066] In order to obtain the minimum hash value of each set, these characteristic rows are reordered, that is, the matrix is ​​randomly shuffled, and then the first non-zero value of each column after reordering is taken as the minimum hash value of the set corresponding to the column;

[0067] Repeat the previous step to perform multiple reorderings. After each reordering, each set has a minimum hash value. Write the hash values ​​generated by each set into a column vector to obtain the MinHash signature vector.

[0068] Based on the obtained MinHash signature vector, the Jaccard similarity formula is used to calculate the similarity between sets. The specific formula is as follows:

[0069]

[0070] Among them, X and Y are MinHash signature vectors corresponding to different sets, Jac(X,Y) is the Jaccard similarity, and the Jaccard similarity is the trajectory overlap.

[0071] As one of the optional embodiments, after calculating the trajectory overlap of the vehicle network card and the mobile phone network card in the same time period, the method further includes:

[0072] Based on the data of the base station of the Internet of Vehicles card and the data of the base station of the mobile network card, the behavior and location characteristics of the Internet of Vehicles card and the mobile network card of the same user when they are connected to and disconnected from the base station within a preset time period are mined to obtain the spatiotemporal characteristics of the base station;

[0073] According to the spatiotemporal characteristics of the base station, the probability that the actual user corresponding to the Internet of Vehicles card and the mobile phone network card is the same is determined, and the trajectory overlap is corrected according to the probability.

[0074] Specifically, in order to improve the prediction accuracy of the subsequent fraud detection model, the location and time data of the base stations to which the Internet of Vehicles card and the mobile network card are connected are mined to obtain the temporal and spatial characteristics of the base stations. The preset time period for data mining is one day, one week or one month, and the temporal and spatial characteristics of the base stations include: the same number of base station positioning, the same base station access duration, the same base station access time difference, the same base station access time difference, the base station positioning distance of the two cards at the same time, the base station positioning distance within the time difference, etc.

[0075] Furthermore, by comparing the spatiotemporal characteristics of the base station with the corresponding preset threshold, the probability that the actual user corresponding to the car network card and the mobile network card is the same person can be obtained, and the trajectory overlap can be corrected according to the obtained probability. For example, if the number of times the base station positioning of the car network card and the mobile network card is the same is less than the preset threshold, it means that the probability that the two cards belong to the same actual user is low. Then, after obtaining the trajectory overlap, a smaller weight is given to the trajectory overlap to correct the trajectory overlap, thereby improving the accuracy of the model prediction.

[0076] As one of the optional embodiments, the Internet of Vehicles card data also includes: Internet of Vehicles card user data, device data of the device used by the user in the real-name process and the generated behavior data; the mobile network card data also includes: mobile network card user data;

[0077] The characteristic data also includes: second characteristic data and third characteristic data, wherein the second characteristic data is used to indicate the behavioral characteristics of the user during the real-name process, and the third characteristic data is used to indicate the real-name consistency between the vehicle network card data and the mobile phone network card data.

[0078] Specifically, the Internet of Vehicles card user data includes: at least one of the Internet of Vehicles card number, province of origin, real-name status, activation time, first use time, vehicle status, ID card number, user name, age, home address, and rate type;

[0079] The device data and behavioral data generated by the device used by the user in the real-name process, where the device data includes at least one of the device brand, operating system, device ID, and device IP address of the device used by the user in the real-name process; the behavioral data includes at least one of the behavioral data of account login, page entry, button click, real-name operation, and real-name unbinding operation generated by the user in the real-name process.

[0080] Specifically, for the acquisition of device data and behavior data, the Internet of Vehicles real-name registration platform provides real-name registration related interfaces in advance, and the corresponding real-name registration APP or website implements these interfaces. When the user is performing real-name registration, these interfaces are called to send the user behavior data related to real-name registration collected by the real-name registration APP or website, such as activation events, registration events, real-name events, etc., to the Internet of Vehicles real-name registration platform, and the Internet of Vehicles real-name registration platform stores the user behavior data in the database table. Therefore, device data and behavior data with time information can be obtained from the database table of the Internet of Vehicles real-name registration platform. Through subsequent analysis of the data, it can be predicted whether the user skips the liveness detection during the real-name process and whether the identity is collected through normal channels, which improves the accuracy of the Internet of Vehicles card anti-fraud detection.

[0081] Specifically, the mobile network card user data includes: mobile phone number, province of origin, ID number, user name, age, home address, service type, tariff type, directional traffic, function activation type and at least one of the traffic limit amount.

[0082] As one of the optional embodiments, the extracting features of the vehicle network card data and the mobile phone network card data to obtain feature data further includes:

[0083] Extracting the second feature data in chronological order based on the device data and the behavior data;

[0084] Based on the Internet of Vehicles card user data and the mobile network card user data, extracting Internet of Vehicles card user characteristics and mobile network card user characteristics;

[0085] The similarity between the user characteristics of the Internet of Vehicles card and the user characteristics of the mobile phone network card is calculated to obtain the third characteristic data.

[0086] Specifically, the user has a certain time sequence in the real-name process, for example, the real-name can only be performed after registering, logging in, and filling in the identity information. Based on these behavior data and the device data involved in the real-name process, the behavior characteristics of the user in the real-name process can be extracted, that is, the second feature data. Among them, the second feature data is a sequential code in time order, which maps various types of data with discrete characteristics into integer serial numbers.

[0087] Further, extracting the user characteristics of the Internet of Vehicles card and the user characteristics of the mobile network card based on the user data of the Internet of Vehicles card and the user data of the mobile network card specifically includes:

[0088] Performing one-hot encoding on the Internet of Vehicles card user data and the mobile phone network card user data to obtain Internet of Vehicles card user encoding data and mobile phone network card user encoding data;

[0089] The PCA method is used to reduce the dimension of the vehicle network card user coding data and the mobile phone network card user coding data to obtain vehicle network card user characteristics and mobile phone network card user characteristics.

[0090] Specifically, first, one-hot encoding is performed on the vehicle network card user data and the mobile network card user data, such as age, home address, name and other discrete feature data, and then the encoded data is reduced in dimension using PCA (principal components analysis) to obtain the reduced-dimensional features, and finally the similarity between the reduced-dimensional features of the vehicle network card and the mobile network card is calculated to obtain the real-name consistency feature, i.e., the third feature data. Preferably, the Euclidean distance is used to calculate the similarity between the reduced-dimensional features of the vehicle network card and the mobile network card.

[0091] As one of the optional embodiments, the fraud detection model is obtained by fusing the first sub-model, the second sub-model and the third sub-model;

[0092] Among them, the first sub-model is trained based on the historical first feature data; the second sub-model is trained based on the historical second feature data; and the third sub-model is trained based on the historical third feature data.

[0093] It is understandable that the historical first characteristic data is used to indicate the temporal and spatial consistency of the historical vehicle network card data and the historical mobile network card data, the historical second characteristic data is used to indicate the behavioral characteristics of historical users in the real-name process, and the historical third characteristic data is used to indicate the real-name consistency of the historical vehicle network card data and the historical mobile network card data. These historical characteristic data can refer to the above-mentioned method for obtaining characteristic data, which will not be repeated here.

[0094] Preferably, the first sub-model is a LightGBM model, the second sub-model is a Bilstm recurrent neural network model, and the third sub-model is a CNN convolutional neural network model.

[0095] Specifically, this application adopts the Stacking method, combines the Bilstm recurrent neural network model, the CNN convolutional neural network model, and the LightGBM model, and obtains an integrated learning model (i.e., fraud detection model) for real-name fraud detection of Internet of Vehicles cards. Among them, the LightGBM model targets the spatiotemporal consistency feature, i.e., the first feature data; the Bilstm recurrent neural network model targets the behavioral features with time sequence, i.e., the second feature data; and the CNN convolutional neural network model targets the real-name consistency feature, i.e., the third feature data.

[0096] It should be noted that Stacking has two layers. The first layer is different sub-models, which are different weak classifiers after training. The second layer is a classifier used to combine sub-models.

[0097] In a specific example, the training and prediction process of the ensemble learning model is as follows:

[0098] First, select three sub-models: Bilstm model, CNN model, and LightGBM model;

[0099] Get the data for model training and prediction, divide nine-tenths of it into the total training set and the remaining one-tenth into the prediction set;

[0100] The total training set is then divided into 10 data sets, and these 10 data sets are used to perform ten-fold cross validation on each sub-model. The specific process is: in each cross validation, a different data set from the 10 data sets is selected as the test set, and the remaining 9 data sets are used as training sets. After the model is trained using the training set, the test set is predicted to obtain the first prediction value of the test set in this cross validation; then the prediction set is predicted to obtain the second prediction value of the prediction set; each sub-model undergoes the above cross validation process 10 times, and 10 first prediction values ​​and 10 second prediction values ​​corresponding to each sub-model are obtained;

[0101] Then, the 10 second prediction values ​​of each sub-model are averaged, and the obtained result is used as the prediction set of the second-layer classifier, and the first prediction values ​​of all sub-models are used as the training set of the second-layer classifier to train and predict the second-layer classifier. The prediction results of each sub-model are quadratically fitted by the second-layer classifier to obtain the fraud detection results output by the integrated learning model, thereby realizing the risk prevention and control of real-name fraud of Internet of Vehicles cards at the operator level. Optionally, the fraud detection result includes the fraud probability. The higher the fraud probability, the lower the consistency between the person and the certificate, and the greater the fraud risk.

[0102] In the prior art, the adopted Internet of Vehicles card risk control solutions include: one solution is to use the massive data stored in the operator platform to associate the province, city and three-level administrator information to which the Internet of Vehicles card belongs, and send the relevant tasks to specific verification personnel for risk prevention and control. This method is time-consuming and labor-intensive, and it is impossible to identify whether the user uses a forged identity for real-name registration. In the case where the user uses illegal means to skip the liveness detection and uses forged information for real-name registration, it is impossible to perform person-document consistency detection, and there is a great risk of fraud; another solution is to use an AI model to check whether the text content uploaded by the user, the personal bare-head photo picture and the secondary collected ID card information are consistent, to perform real-name detection, but it can only determine whether the real-name information and the ID card information are correct, and cannot identify whether the actual user and the ID card information are consistent, which poses a risk of fraud.

[0103] The embodiments of the present application can make full use of the operator's own natural data volume advantage and a large number of mobile phone users to mine the user behavior data, real-name data, base station data and time information data of mobile network cards and Internet of Vehicles cards. The data volume advantage of mobile network cards is used to make up for the problems of small data volume and insufficient features in the anti-fraud modeling of Internet of Vehicles cards, and the advanced 5G network is used to accelerate data transmission, solving problems such as communication overhead and data privacy. Finally, the consistency of the real name of the car card is derived from the user's real-name behavior, the real-name consistency of the car card and the mobile phone card, and the consistency of spatiotemporal data, so as to conduct a real-name review of the Internet of Vehicles card, thereby conducting risk prevention and control of the Internet of Vehicles card. This application enables operators to perceive at their own level whether the actual user of the locomotive is consistent with the identity information submitted for real-name, and can prevent fraudulent users from stealing other people's identities and forging false information, thereby improving the comprehensiveness and accuracy of the fraud detection model.

[0104] Correspondingly, the present application also provides an anti-fraud device based on a vehicle networking card, which can implement all processes of the anti-fraud method based on a vehicle networking card in the above-mentioned embodiment.

[0105] See also Figure 2 , Figure 2 : is a schematic diagram of the structure of an anti-fraud device based on a vehicle networking card provided in an embodiment of the present application. The anti-fraud device based on a vehicle networking card includes:

[0106] The data acquisition module 201 is used to acquire the vehicle network card data and the mobile phone network card data of the same user;

[0107] The feature extraction module 202 is used to extract features from the vehicle network card data and the mobile phone network card data to obtain feature data; wherein the feature data includes: first feature data, the first feature data is used to indicate the temporal and spatial consistency of the vehicle network card data and the mobile phone network card data;

[0108] The fraud detection module 203 is used to perform fraud detection based on the feature data using a preset fraud detection model to obtain a fraud detection result; wherein the fraud detection model is trained based on historical feature data, and the historical feature data includes historical first feature data.

[0109] Preferably, the vehicle network card data includes vehicle network card base station data, and the mobile phone network card data includes mobile phone network card base station data;

[0110] The feature extraction module 202 includes:

[0111] A base station collection unit, configured to obtain a set of base stations of the Internet of Vehicles card and a set of base stations of the mobile network card of the user in the same time period according to the base station data of the Internet of Vehicles card and the base station data of the mobile network card;

[0112] The trajectory overlap unit is used to calculate the trajectory overlap of the user's vehicle network card and mobile phone network card in the same time period according to the vehicle network card base station set and the mobile phone network card base station set to obtain the first feature data.

[0113] Preferably, the trajectory overlap unit is specifically used for:

[0114] The similarity between the vehicle network card base station set and the mobile phone network card base station set is calculated to obtain the trajectory overlap of the user's vehicle network card and mobile phone network card in the same time period, and use the trajectory overlap as the first feature data.

[0115] Preferably, the Internet of Vehicles card data also includes: Internet of Vehicles card user data, device data of the device used by the user in the real-name process and the generated behavior data; the mobile network card data also includes: mobile network card user data;

[0116] The characteristic data also includes: second characteristic data and third characteristic data, wherein the second characteristic data is used to indicate the behavioral characteristics of the user during the real-name process, and the third characteristic data is used to indicate the real-name consistency between the vehicle network card data and the mobile phone network card data.

[0117] Preferably, the feature extraction module 202 further includes:

[0118] A second feature unit, configured to extract the second feature data in chronological order based on the device data and the behavior data;

[0119] The third feature unit is used to extract the user characteristics of the Internet of Vehicles card and the user characteristics of the mobile phone network card based on the user data of the Internet of Vehicles card and the user data of the mobile phone network card; calculate the similarity between the user characteristics of the Internet of Vehicles card and the user characteristics of the mobile phone network card to obtain the third feature data.

[0120] Preferably, the fraud detection model is obtained by fusing the first sub-model, the second sub-model and the third sub-model;

[0121] Among them, the first sub-model is trained based on the historical first feature data; the second sub-model is trained based on the historical second feature data; and the third sub-model is trained based on the historical third feature data.

[0122] In the specific implementation, the working principle, control process and technical effect of the anti-fraud device based on the Internet of Vehicles card provided in the embodiment of the present application are the same as those of the anti-fraud method based on the Internet of Vehicles card in the above-mentioned embodiment, and will not be repeated here.

[0123] See also Figure 3 , Figure 3 : is a schematic diagram of the structure of an anti-fraud device based on a vehicle networking card provided in an embodiment of the present application, and the anti-fraud device based on a vehicle networking card includes: a processor 301, a memory 302, and a computer program stored in the memory 302 and executable on the processor 301. When the processor 301 executes the computer program, the steps in the above-mentioned anti-fraud method embodiment based on the vehicle networking card are implemented. Alternatively, when the processor 301 executes the computer program, the functions of each module / unit in the above-mentioned device embodiments are implemented.

[0124] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory 302 and executed by the processor 301 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the computer program in the anti-fraud device based on the Internet of Vehicles card.

[0125] The anti-fraud device based on the Internet of Vehicles card may include, but is not limited to, a processor 301 and a memory 302. Those skilled in the art will appreciate that the schematic diagram is merely an example of an anti-fraud device based on the Internet of Vehicles card, and does not constitute a limitation on the anti-fraud device based on the Internet of Vehicles card, and may include more or fewer components than shown in the diagram, or a combination of certain components, or different components, for example, the anti-fraud device based on the Internet of Vehicles card may also include input and output devices, network access devices, buses, and the like.

[0126] The processor 301 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor 301 is the control center of the anti-fraud device based on the Internet of Vehicles card, and uses various interfaces and lines to connect the various parts of the entire anti-fraud device based on the Internet of Vehicles card.

[0127] The memory 302 can be used to store the computer program and / or module, and the processor 301 realizes various functions of the anti-fraud device based on the Internet of Vehicles card by running or executing the computer program and / or module stored in the memory 302, and calling the data stored in the memory 302. The memory 302 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory 302 can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0128] Among them, if the module / unit integrated in the anti-fraud device based on the Internet of Vehicles card is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor 301, it can implement the steps of the above-mentioned various method embodiments. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc.

[0129] An embodiment of the present application also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the anti-fraud method based on the Internet of Vehicles card described in any of the above embodiments.

[0130] An embodiment of the present application also provides a computer program product, including a computer program / instruction, which, when executed by a processor of the computer program / instruction, implements the anti-fraud method based on the Internet of Vehicles card described in any of the above embodiments.

[0131] The method in the present application can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instruction is loaded and executed on a computer, the process or function described in the present application is executed in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user device, a core network device, an OAM or other programmable device.

[0132] The computer program or instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program or instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless means. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; it may also be an optical medium, such as a digital video disk; it may also be a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or non-volatile storage medium, or may include both volatile and non-volatile types of storage media.

[0133] The computer programs / instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing / processing device.

[0134] The computer program instructions for performing the operation of the present application can be assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions can be executed completely on a user's computer, partially on a user's computer, executed as an independent software package, partially on a user's computer, partially on a remote computer, or completely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet). In some embodiments, by using the state information of a computer-readable program instruction to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA) or a programmable logic array (PLA), the electronic circuit can execute a computer-readable program instruction, thereby realizing various aspects of the present application.

[0135] Various aspects of the present application are described herein with reference to flowcharts and / or block diagrams according to embodiments of the present application. It should be understood that each block of the flowcharts and / or block diagrams and the combination of each block in the flowcharts and / or block diagrams can be implemented by computer-readable program instructions.

[0136] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0137] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0138] Each square frame in the flow chart or block diagram can represent a part of a module, a program segment or an instruction, and a part of the module, program segment or an instruction comprises one or more executable instructions for realizing the logical function of the regulation. In some implementations as replacement, the function marked in the square frame can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square frames can actually be executed substantially in parallel, and they can also be executed in the opposite order sometimes, depending on the function involved. It should also be noted that each square frame in the block diagram and / or the flow chart, and the combination of the square frames in the block diagram and / or the flow chart can be realized by a dedicated hardware-based system that performs the function or action of the regulation, or can be realized by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that it is equivalent to realize by hardware, realize by software, and realize by the combination of software and hardware.

[0139] The anti-fraud method, device, equipment, medium and product based on the Internet of Vehicles card provided by the embodiment of the present application have the following beneficial effects: by obtaining the Internet of Vehicles card data and the mobile network card data corresponding to the same user, and performing feature extraction to obtain feature data including the time-space consistency of the Internet of Vehicles card and the mobile network card, the feature data is input into a preset fraud detection model to obtain a fraud detection result, so as to identify the real-name accuracy of the Internet of Vehicles card and the safety of the card usage behavior, confirm the consistency of the person and the certificate of the Internet of Vehicles card, and realize the risk prevention and control of fraud of the real-name information of the Internet of Vehicles card at the operator level; through the behavior data generated by the user in the process of real-name registration on the Internet of Vehicles card real-name registration platform and the device data of the device used, the behavior features used to indicate the user in the real-name process are extracted to perform fraud detection, so as to predict whether the user skips the liveness detection and whether the identity is collected through normal channels in the real-name process, thereby improving the accuracy of fraud detection; by extracting the features corresponding to the real-name consistency and the time-space consistency, it is possible to judge whether the real-name information and the time-space data of the Internet of Vehicles card and the mobile network card of the same user are consistent, and then it is possible to further confirm the consistency of the person and the certificate of the Internet of Vehicles card, thereby improving the comprehensiveness and accuracy of fraud detection of the Internet of Vehicles card. The embodiments of the present application realize the real-name verification of Internet of Vehicles cards and the prediction of fraud risks of Internet of Vehicles cards at the operator level, thereby preventing and controlling the fraud risks caused by false identity information and improving the comprehensiveness and accuracy of real-name fraud detection of Internet of Vehicles cards.

[0140] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications are also considered to be within the scope of protection of the present application.

Claims

1. An anti-fraud method based on Internet of Vehicles card, characterized in that: include: Get the car network card data and mobile phone network card data of the same user; Extracting features from the vehicle network card data and the mobile phone network card data to obtain feature data; wherein the feature data includes: first feature data, the first feature data is used to indicate the spatiotemporal consistency of the vehicle network card data and the mobile phone network card data; Based on the feature data, a preset fraud detection model is used to perform fraud detection to obtain a fraud detection result; wherein the fraud detection model is trained based on historical feature data, and the historical feature data includes historical first feature data.

2. The anti-fraud method based on the Internet of Vehicles card as claimed in claim 1, characterized in that: The vehicle network card data includes vehicle network card base station data, and the mobile network card data includes mobile network card base station data; The extracting features of the vehicle network card data and the mobile phone network card data to obtain feature data includes: According to the vehicle network card base station data and the mobile network card base station data, obtaining the vehicle network card base station set and the mobile network card base station set of the user in the same time period; According to the vehicle network card base station set and the mobile phone network card base station set, the trajectory overlap degree of the user's vehicle network card and mobile phone network card in the same time period is calculated to obtain the first feature data.

3. The anti-fraud method based on the Internet of Vehicles card as claimed in claim 2, characterized in that: The calculating, based on the set of vehicle network card base stations and the set of mobile network card base stations, the trajectory overlap of the vehicle network card and the mobile network card of the user in the same time period to obtain the first feature data includes: The similarity between the vehicle network card base station set and the mobile phone network card base station set is calculated to obtain the trajectory overlap of the user's vehicle network card and mobile phone network card in the same time period, and use the trajectory overlap as the first feature data.

4. The anti-fraud method based on the Internet of Vehicles card as claimed in claim 2, characterized in that: The Internet of Vehicles card data also includes: Internet of Vehicles card user data, device data of the device used by the user in the real-name process and the generated behavior data; the mobile network card data also includes: mobile network card user data; The characteristic data also includes: second characteristic data and third characteristic data, wherein the second characteristic data is used to indicate the behavioral characteristics of the user during the real-name process, and the third characteristic data is used to indicate the real-name consistency between the vehicle network card data and the mobile phone network card data.

5. The anti-fraud method based on the Internet of Vehicles card as claimed in claim 4, characterized in that: The extracting features of the vehicle network card data and the mobile phone network card data to obtain feature data also includes: Extracting the second feature data in chronological order based on the device data and the behavior data; Based on the Internet of Vehicles card user data and the mobile network card user data, extracting Internet of Vehicles card user characteristics and mobile network card user characteristics; The similarity between the user characteristics of the Internet of Vehicles card and the user characteristics of the mobile phone network card is calculated to obtain the third characteristic data.

6. The anti-fraud method based on the Internet of Vehicles card as claimed in claim 4, characterized in that: The fraud detection model is obtained by fusing the first sub-model, the second sub-model and the third sub-model; Among them, the first sub-model is trained based on the historical first feature data; the second sub-model is trained based on the historical second feature data; and the third sub-model is trained based on the historical third feature data.

7. An anti-fraud device based on a vehicle networking card, characterized in that: include: A data acquisition module, used to acquire the car network card data and mobile phone network card data of the same user; A feature extraction module is used to extract features from the vehicle network card data and the mobile phone network card data to obtain feature data; wherein the feature data includes: first feature data, the first feature data is used to indicate the temporal and spatial consistency of the vehicle network card data and the mobile phone network card data; A fraud detection module is used to perform fraud detection based on the feature data using a preset fraud detection model to obtain a fraud detection result; wherein the fraud detection model is trained based on historical feature data, and the historical feature data includes historical first feature data. 8.An anti-fraud device based on a car networking card, characterized in that: It includes a processor and a memory, wherein a computer program is stored in the memory, and the computer program is configured to be executed by the processor, and when the processor executes the computer program, an anti-fraud method based on an Internet of Vehicles card as described in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein when the device where the computer-readable storage medium is located executes the computer program, the anti-fraud method based on the Internet of Vehicles card as described in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the anti-fraud method based on the Internet of Vehicles card as described in any one of claims 1 to 6 is implemented.

Images