Virtual positioning intelligent detection system and method based on multi-source data fusion

Abstract

The invention discloses a virtual positioning intelligent detection system and method based on multi-source data fusion, and relates to the technical field of virtual positioning detection, and the method comprises the steps: analyzing the data trend conditions of behavior indexes of different users in an intelligent detection platform, and obtaining the behavior data of a target user; acquiring a user positioning record of a user in the intelligent detection platform, acquiring coordinate data and sensor data from the user positioning record, and analyzing an abnormal condition of user positioning; obtaining a user behavior record of the user in the current period, evaluating the abnormal degree of the user behavior of the user according to the target user behavior data, and detecting the positioning virtual condition of the user in combination with the positioning abnormal data to obtain a virtual positioning user; the method comprises the following steps: acquiring historical positioning detection data of a virtual positioning user, acquiring account processing system data in an intelligent detection platform, acquiring account data of the virtual positioning user, and performing authority management on an account of the virtual positioning user.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual positioning detection, and in particular to a virtual positioning intelligent detection system and method based on multi-source data fusion. Background Art

[0002] Virtual positioning is when users use software and other technical means to fake their real location, so that the geographic location displayed by the device does not match its actual location. Legal virtual positioning can help developers simulate different location scenarios and test whether the software functions normally in unavailable areas. However, the arbitrary use of virtual positioning can easily lead to a series of abnormal behaviors. Users who use virtual positioning are likely to use fake location information to seek improper benefits or engage in abnormal behaviors. For example, fake users will sign in and consume at specific merchants, thereby improving the store's ranking. It is even possible to bypass the risk control system of financial institutions and evade the constraints of laws or platform rules.

[0003] Therefore, it is very important to detect the user's virtual positioning. Conventional methods for detecting virtual positioning include: 1. IP address positioning, inferring the user's geographic location through the user's IP address, and comparing it with the reported geographic location to detect the virtual positioning, but the IP address can be hidden and forged through VPN or proxy server, resulting in inaccuracy; 2. Base station detection method, by detecting information such as the signal strength of the base station connected to the device, to infer the user's real location, but if the user forges the base station information, it may cause detection failure; 3. GPS verification, by analyzing the logical consistency of the user's GPS signal strength, number of satellites, and location data, to detect the virtual positioning, but there are currently high-quality virtual positioning tools to simulate the movement trajectory of the real GPS, affecting the detection results; therefore, there is currently a lack of an intelligent virtual positioning detection method to detect the user's virtual positioning. Summary of the Invention

[0004] The purpose of the present invention is to provide a virtual positioning intelligent detection system and method based on multi-source data fusion to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a virtual positioning intelligent detection method based on multi-source data fusion, the method comprising: Step S100: Obtain historical user behavior records of different users in the intelligent detection platform, obtain user behavior data from the historical user behavior records, analyze the data trends of behavior indicators of different users in the intelligent detection platform, and obtain target user behavior data; Step S200: Obtain the user location records of the users in the intelligent detection platform, obtain coordinate data and sensor data from the user location records, analyze the abnormal conditions of the user locations, and obtain location abnormal data; Step S300: Obtain the user behavior records of the users in the current cycle, evaluate the degree of abnormality of the user behavior according to the target user behavior data, and combine the location abnormal data to detect the virtual location situation of the users, and obtain virtual location users; Step S400: Obtain the historical location detection data of the virtual location users, obtain the account processing system data in the intelligent detection platform, obtain the account data of the virtual location users, and perform permission management on the accounts of the virtual location users.

[0006] Further, step S100 includes: Step S101: Obtain the previous historical cycle of the current cycle, and record the previous historical cycle as the acquisition historical cycle of the intelligent detection platform; Step S102: Obtain the historical user behavior records of each user in the intelligent detection platform during the acquisition historical cycle, and preprocess the historical user behavior records, and obtain user behavior data from the historical user behavior records. The user behavior data includes the values corresponding to the various behavior indicators of the user; Step S103: Analyze the data trend of the behavior indicators of each user in the intelligent detection platform. Among them, the analysis process of the data trend of the a-th behavior indicator of each user is as follows: Obtain the average value μ of the a-th behavior indicator of each user a and the standard deviation σ a , obtain the m-th range C of the a-th behavior indicator (a,m) = [μ a - m×σ a , μ a + m×σ a , obtain the total number U of several users whose values of the a-th behavior indicator are within the m-th range C (a,m) , obtain the total number U of each user in the intelligent detection platform (a,m) , calculate the user coverage ratio B of the m-th range C SUM = U (a,m) / U (a,m) ; (a,m) ; SUM ; Set the coverage ratio threshold B´. When B (a,m) < B´ and the user coverage ratio B of the (m + 1)-th range C of the a-th behavior indicator (a,m+1) > B´, set the (m + 1)-th range C (a,m+1) >B´, set the (m + 1)-th range C (a,m+1)Recorded as the target range of the a-th behavior indicator, it is determined whether the data trend of the a-th behavior indicator of each user in the current period is within the target range; Step S104: Obtain target ranges of various behavioral indicators of the intelligent detection platform in the current cycle and aggregate them to obtain target user behavior data; The target behavior data mentioned above is to obtain the data trend of the overall user behavior of each user in the platform. In the platform, the user behavior of general users should be similar to that of the overall users, and there will not be a large deviation. When the behavior indicators of a certain user deviate significantly from the data of the overall users in the historical collection period, the more abnormal the user is, the more likely its positioning is virtual. By analyzing the users, the accuracy of virtual positioning detection can be greatly improved.

[0007] Furthermore, step S200 includes: Step S201: Use the intelligent detection platform to obtain the user's location records in the current period and analyze the abnormal situation of the user's location. The specific process is as follows: Obtaining coordinate data from user location records, the coordinate data including the user's GPS coordinate location set and base station location set; Step S202: Time-aligning the GPS coordinate positioning set and the base station positioning set according to the timestamps. After the time alignment, the total number of elements in the GPS coordinate positioning set and the base station positioning set is the same. The coordinates corresponding to the elements in the user's GPS coordinate positioning set and the base station positioning set are converted into angle distances, wherein the coordinates corresponding to the e-th element in the GPS coordinate positioning set (lat （e,gps） ,lon （e,gps） )Angle distance conversion, get (φ （e,gps） ,λ （e,gps） ), the specific formula is: （e,gps） =lat （e,gps） × (π / 180), λ （e,gps） =lon （e,gps） × (π / 180); Step S203: Calculate the positioning distance difference between each element in the GPS coordinate positioning set and the base station positioning set, where the coordinate corresponding to the gth element in the GPS coordinate positioning set (φ （g,gps） ,λ （g,gps） ), and the coordinates corresponding to the gth element in the base station location set (φ （g,lbs） ,λ （g,lbs） ) between the positioning distance difference L g , the specific calculation process is: Get the cumulative value Q of the semi-sine function of the difference in longitude and latitude corresponding to the gth element between the GPS coordinate positioning set and the base station positioning setg ; Calculate the angular distance S between the gth element of the GPS coordinate positioning set and the base station positioning set g : , Calculate the positioning distance difference L g =S g ×R, R is the radius of the earth; Obtain an area containing the coordinates corresponding to the g-th element in the GPS coordinate positioning set and the base station positioning set, and obtain a preset positioning distance difference threshold in the area; When the positioning distance difference L g If the value is greater than the positioning distance difference threshold, it is determined that the user has a coordinate error, and the total number of coordinate errors of the user in the current cycle is obtained. sum ; Get the total number of elements W in the GPS coordinate positioning set sum , calculate the user's coordinate anomaly score V=α in the current cycle sum / W sum ; Step S204: acquiring sensor data from the user location record, where the sensor data includes a set of direction angles collected by a direction sensor in the user's device during a current period; Time align the GPS coordinate positioning set and the direction angle set; Step S205: The coordinates corresponding to the βth element in the GPS coordinate positioning set (φ （β,gps） ,λ （β,gps） ) to transform the plane coordinate system and obtain the coordinates (x （β,gps） ,y （β,gps） ), where x （β,gps） =K·λ （β,gps） ,y （β,gps） =K·φ （β,gps） , K is the preset conversion coefficient, and the coordinates corresponding to each element in the GPS coordinate positioning set are converted into a plane coordinate system; Get the direction angle θ of the βth element β : Calculate the change in the direction angle of the βth element △θ β =θ β+1 -θ β ,θ β+1 is the direction angle of the β+1th element, and obtains the change in the direction angle of each element in the GPS coordinate positioning set; Step S206: Obtain the change in the direction angle of each element in the direction angle set, and calculate the direction angle change error between the direction angle set and each element in the GPS coordinate positioning set, wherein the change in the direction angle of the hth element in the direction angle set is Δθ´h , and the direction angle △θ of the hth element in the GPS coordinate positioning set h The angular error between h =|△θ h -△θ´ h |; When the direction angle change error ε is greater than the preset error threshold, it is determined that the user has a direction change error. The ratio between the total number of direction change errors in the current cycle and the total number of elements in the direction angle set is obtained and recorded as the trajectory anomaly score Z; Step S207: Obtain the user's trajectory anomaly score Z and coordinate anomaly score V in the current period and aggregate them to obtain positioning anomaly data.

[0008] Furthermore, step S300 includes: Step S301: Obtain the values of various behavior indicators of the user in the current period from the user behavior record; Obtain the target user behavior data of the intelligent detection platform in the current cycle, and obtain the target range of each behavior indicator from the target user behavior data; When the value of a certain behavior indicator of a user is not within the target range corresponding to the behavior indicator, the behavior indicator is determined to be abnormal; Obtain the ratio of the total number of the user's behavioral indicators that are judged to be abnormal to the total number of each behavioral indicator to obtain the user's behavioral abnormality score I; Detect the user's virtual positioning situation. The specific detection process is as follows: Obtain the trajectory anomaly score Z and coordinate anomaly score V from the user's positioning anomaly data, and calculate the user's virtual positioning score P = η in the current cycle Z ×Z+η I ×I+η V ×V, where η Z ,η I and η V These are the preset proportional coefficients for the trajectory abnormality score Z, the behavior abnormality score I, and the coordinate abnormality score V; When the virtual score P is greater than the preset virtual score threshold, the user's positioning in the current period is determined to be virtual positioning, and the user is recorded as a virtual positioning user.

[0009] Furthermore, step S400 includes: Step S401: Obtain historical positioning detection data of the virtual positioning user from the intelligent detection platform, and obtain the total number of times N when the positioning of the virtual positioning user is determined to be virtual positioning from the historical positioning detection data; Step S402: Obtain the account data of the virtual positioning user, which includes the data corresponding to the various account parameters of the virtual positioning user, and combine it with the account processing system information in the intelligent detection platform to generate a penalty strategy for the account of the virtual positioning user, and perform authority management on the account of the virtual positioning user, and report the virtual positioning user to the staff in the intelligent detection platform.

[0010] In order to better implement the above method, a virtual positioning intelligent detection system based on multi-source data fusion is also proposed. The system includes a data trend analysis module, a positioning anomaly analysis module, a virtual positioning detection module and an account management module; The data trend analysis module is used to obtain historical user behavior records of different users in the intelligent detection platform, analyze the data trend of the behavioral indicators of different users in the intelligent detection platform, and obtain the target user behavior data; The positioning anomaly analysis module is used to obtain the user's user positioning records, analyze the abnormal conditions of the user's positioning, and obtain positioning anomaly data; The virtual positioning detection module is used to evaluate the abnormal degree of user behavior and, in combination with positioning abnormality data, detect the user's virtual positioning situation to obtain a virtual positioning user; The account management module is used to obtain the historical positioning detection data of the virtual positioning user, and to manage the permissions of the virtual positioning user's account based on the account processing system data pre-screened by the intelligent detection platform.

[0011] Furthermore, the data trend analysis module includes a data acquisition unit and a data trend analysis unit; The data collection unit is used to obtain the historical collection period of the intelligent detection platform, obtain the historical user behavior records of each user within the historical collection period, and obtain user behavior data from the historical user behavior records; The data trend analysis unit is used to analyze the data trend of the behavior indicators of each user in the intelligent detection platform based on the user behavior data to obtain the target user behavior data.

[0012] Furthermore, the positioning anomaly analysis module includes a coordinate anomaly scoring unit and a positioning anomaly analysis unit; A coordinate anomaly scoring unit is used to obtain coordinate data from the user's user location record and calculate the user's coordinate anomaly score in the current period based on the coordinate data; The positioning anomaly analysis unit is used to obtain sensor data from the user's positioning records, calculate the user's trajectory anomaly score, and aggregate the trajectory anomaly score and coordinate anomaly score to obtain positioning anomaly data.

[0013] Furthermore, the virtual location detection module includes a behavior anomaly scoring unit and a virtual location detection unit; The behavior anomaly scoring unit is used to obtain the user's behavior records in the current cycle and calculate the user's behavior anomaly score; The virtual positioning detection unit is used to detect the user's virtual positioning situation based on the behavior abnormality score and combined with the positioning abnormality data to obtain a virtual positioning user.

[0014] Furthermore, the account management module includes an account management unit; The account management unit is used to obtain the historical positioning detection data of the virtual positioning user, and to manage the permissions of the virtual positioning user's account in combination with the account processing system information in the intelligent detection platform.

[0015] Compared with the prior art, the beneficial effects of the present invention are: the present invention realizes the intelligent detection of the user's virtual positioning. In the process of virtual positioning detection of the user, not only the user positioning record in the current period is analyzed, but also the matching between the user's behavior and the overall users in the platform is considered. When the user's positioning is abnormal and the user's behavior deviates significantly from the behavior of the overall users in the platform, then there is a great probability that the user has used virtual positioning in the platform. Therefore, the detection method of the present invention is not only fast, but also integrates multi-source data, which greatly improves the accuracy of virtual positioning detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a method flow chart of a virtual positioning intelligent detection method based on multi-source data fusion of the present invention; Figure 2 It is a module schematic diagram of a virtual positioning intelligent detection system based on multi-source data fusion of the present invention. DETAILED DESCRIPTION

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0018] Example: Figure 1-Figure 2 As shown, the present invention provides a technical solution, a virtual positioning intelligent detection method based on multi-source data fusion, the method comprising: Step S100: Obtain the historical user behavior records of different users in the intelligent detection platform, obtain user behavior data from the historical user behavior records, analyze the data trend of the behavior indicators of different users in the intelligent detection platform, and obtain the target user behavior data; Among them, step S100 includes: Step S101: Obtain the previous historical cycle of the current cycle, and record the previous historical cycle as the acquisition historical cycle of the intelligent detection platform; Step S102: Obtain the historical user behavior records of each user in the intelligent detection platform during the acquisition historical cycle, preprocess the historical user behavior records, obtain user behavior data from the historical user behavior records, and the user behavior data includes the values corresponding to each behavior indicator of the user; For example, the process of preprocessing the historical user behavior records includes handling missing values, handling outliers, etc.; For example, each behavior indicator includes browsing duration, number of likes / comments, frequency of geographical location changes, etc.; Step S103: Analyze the data trend of the behavior indicators of each user in the intelligent detection platform. Among them, the analysis process of the data trend of the a-th behavior indicator of each user is as follows: Obtain the average value μ of the a-th behavior indicator of each user a and the standard deviation σ a , obtain the m-th range C of the a-th behavior indicator (a,m) =[μ a -m×σ a , μ a +m×σ a , obtain the total number U of several users whose values of the a-th behavior indicator are within the m-th range C (a,m) , obtain the total number U of all users in the intelligent detection platform (a,m) , calculate the user coverage ratio B of the m-th range C SUM =U (a,m) / U (a,m) ; (a,m) / U SUM ; Set the coverage ratio threshold B´. When B (a,m) <B´ and the user coverage ratio B of the (m + 1)-th range C of the a-th behavior indicator (a,m+1) >B´, record the (m + 1)-th range C (a,m+1) as the target range of the a-th behavior indicator, and determine that the data trend of the a-th behavior indicator of each user in the current cycle is within the target range; (a,m+1) ; Step S104: Obtain target ranges of various behavioral indicators of the intelligent detection platform in the current cycle and aggregate them to obtain target user behavior data; Step S200: obtaining user location records of the user in the intelligent detection platform, obtaining coordinate data and sensor data from the user location records, analyzing abnormal conditions of the user location, and obtaining location abnormality data; Wherein, step S200 includes: Step S201: Use the intelligent detection platform to obtain the user's location records in the current period and analyze the abnormal situation of the user's location. The specific process is as follows: Obtaining coordinate data from user location records, the coordinate data including the user's GPS coordinate location set and base station location set; Step S202: Time-aligning the GPS coordinate positioning set and the base station positioning set according to the timestamps. After the time alignment, the total number of elements in the GPS coordinate positioning set and the base station positioning set is the same. For example, a specific time alignment method is to obtain the timestamps corresponding to the coordinates of each element in the GPS coordinate positioning set and the base station positioning set, set a time window, generally ±5 seconds, and retain the timestamps of the coordinates of the two elements when they are within the preset time window and have the shortest time distance, so that each element in the GPS coordinate positioning set corresponds one to one with each element in the base station positioning set; The coordinates corresponding to the elements in the user's GPS coordinate positioning set and the base station positioning set are converted into angle distances, wherein the coordinates corresponding to the e-th element in the GPS coordinate positioning set (lat （e,gps） ,lon （e,gps） )Angle distance conversion, get (φ （e,gps） ,λ （e,gps） ), the specific formula is: （e,gps） =lat （e,gps） × (π / 180), λ （e,gps） =lon （e,gps） × (π / 180); Step S203: Calculate the positioning distance difference between each element in the GPS coordinate positioning set and the base station positioning set, where the coordinate corresponding to the gth element in the GPS coordinate positioning set (φ （g,gps） ,λ （g,gps） ), and the coordinates corresponding to the gth element in the base station location set (φ （g,lbs） ,λ （g,lbs） ) between the positioning distance difference L g , the specific calculation process is: Get the cumulative value Q of the semi-sine function of the difference in longitude and latitude corresponding to the gth element between the GPS coordinate positioning set and the base station positioning set g ; For example, in the accumulated value Q g The specific calculation formula is: , Among them, △φ g =φ （g,lbs） -φ （g,gps） ; △λ g =λ （g,lbs） -λ （g,gps） ; Calculate the angular distance S between the gth element of the GPS coordinate positioning set and the base station positioning set g : , Calculate the positioning distance difference L g =S g ×R, R is the radius of the earth; Obtain an area containing the coordinates corresponding to the g-th element in the GPS coordinate positioning set and the base station positioning set, and obtain a preset positioning distance difference threshold in the area; When the positioning distance difference L g If the value is greater than the positioning distance difference threshold, it is determined that the user has a coordinate error, and the total number of coordinate errors of the user in the current cycle is obtained. sum ; Get the total number of elements W in the GPS coordinate positioning set sum , calculate the user's coordinate anomaly score V=α in the current cycle sum / W sum ; Step S204: acquiring sensor data from the user location record, where the sensor data includes a set of direction angles collected by a direction sensor in the user's device during a current period; For example, direction sensors include gyroscopes, electronic compasses, etc. Time align the GPS coordinate positioning set and the direction angle set; Step S205: The coordinates corresponding to the βth element in the GPS coordinate positioning set (φ （β,gps） ,λ （β,gps） ) to transform the plane coordinate system and obtain the coordinates (x （β,gps） ,y （β,gps） ), where x （β,gps） =K·λ （β,gps） ,y （β,gps） =K·φ （β,gps） , K is the preset conversion coefficient, and the coordinates corresponding to each element in the GPS coordinate positioning set are converted into a plane coordinate system; For example, the conversion factor K is generally 1 degree ≈ 111,319.491 meters; Get the direction angle θ of the βth elementβ ; For example, the direction angle θ of the βth element β The specific calculation formula is: , Among them, (x （β+1,gps） ,y （β+1,gps） ) is the coordinate corresponding to the β+1th element in the GPS coordinate positioning set; Calculate the change in the direction angle of the βth element △θ β =θ β+1 -θ β ,θ β+1 is the direction angle of the β+1th element, and obtains the change in the direction angle of each element in the GPS coordinate positioning set; Step S206: Obtain the change in the direction angle of each element in the direction angle set, and calculate the direction angle change error between the direction angle set and each element in the GPS coordinate positioning set, wherein the change in the direction angle of the hth element in the direction angle set is Δθ´ h , and the direction angle △θ of the hth element in the GPS coordinate positioning set h The angular error between h =|△θ h -△θ´ h |; When the direction angle change error ε is greater than the preset error threshold, it is determined that the user has a direction change error. The ratio between the total number of direction change errors in the current cycle and the total number of elements in the direction angle set is obtained and recorded as the trajectory anomaly score Z; Step S207: Obtain the user's trajectory anomaly score Z and coordinate anomaly score V in the current period and aggregate them to obtain positioning anomaly data; Step S300: Obtain the user behavior record of the user in the current cycle, evaluate the abnormality of the user behavior based on the target user behavior data, and detect the user's virtual positioning situation in combination with the positioning abnormality data to obtain a virtual positioning user; Wherein, step S300 includes: Step S301: Obtain the values of various behavior indicators of the user in the current period from the user behavior record; Obtain the target user behavior data of the intelligent detection platform in the current cycle, and obtain the target range of each behavior indicator from the target user behavior data; When the value of a certain behavior indicator of a user is not within the target range corresponding to the behavior indicator, the behavior indicator is determined to be abnormal; Obtain the ratio of the total number of the user's behavioral indicators that are judged to be abnormal to the total number of each behavioral indicator to obtain the user's behavioral abnormality score I; Detect the user's virtual positioning situation. The specific detection process is as follows: Obtain the trajectory anomaly score Z and coordinate anomaly score V from the user's positioning anomaly data, and calculate the user's virtual positioning score P = η in the current cycle Z ×Z+η I ×I+η V ×V, where η Z ,η I and η V These are the preset proportional coefficients for the trajectory abnormality score Z, the behavior abnormality score I, and the coordinate abnormality score V; When the virtual score P is greater than the preset virtual score threshold, the user's positioning in the current period is determined to be virtual positioning, and the user is recorded as a virtual positioning user; Step S400: Acquire historical positioning detection data of the virtual positioning user, obtain account processing system data in the intelligent detection platform, obtain the account data of the virtual positioning user, and perform permission management on the account of the virtual positioning user; Wherein, step S400 includes: Step S401: Obtain historical positioning detection data of the virtual positioning user from the intelligent detection platform, and obtain the total number of times N when the positioning of the virtual positioning user is determined to be virtual positioning from the historical positioning detection data; Step S402: Acquire the account data of the virtual positioning user, which includes data corresponding to various account parameters of the virtual positioning user. Combined with the account processing system information in the intelligent detection platform, generate a penalty policy for the virtual positioning user's account, perform permission management on the virtual positioning user's account, and report the virtual positioning user to the staff of the intelligent detection platform; For example, various account parameters include registration time, account level, etc. For example, account handling system information includes clear rules that users need to follow during use, such as prohibiting the release of illegal information and malicious attacks, and the handling measures for violations of the rules, such as warnings, suspension of use, and account bans; For example, if a user's location is judged to be virtual nine times in total and no identity verification is performed, and the threshold for determining the location of users without identity verification as virtual is eight times in the account handling system information, the corresponding action is to ban the account. Therefore, the generated penalty policy is to ban the user's account. In order to better implement the above method, a virtual positioning intelligent detection system based on multi-source data fusion is also proposed. The system includes a data trend analysis module, a positioning anomaly analysis module, a virtual positioning detection module and an account management module; The data trend analysis module is used to obtain historical user behavior records of different users in the intelligent detection platform, analyze the data trend of the behavioral indicators of different users in the intelligent detection platform, and obtain the target user behavior data; The positioning anomaly analysis module is used to obtain the user's user positioning records, analyze the abnormal conditions of the user's positioning, and obtain positioning anomaly data; The virtual positioning detection module is used to evaluate the abnormal degree of user behavior and, in combination with positioning abnormality data, detect the user's virtual positioning situation to obtain a virtual positioning user; The account management module is used to obtain the historical positioning detection data of virtual positioning users and manage the permissions of virtual positioning users' accounts based on the account processing system data pre-screened by the intelligent detection platform; Wherein, the data trend analysis module includes a data acquisition unit and a data trend analysis unit; The data collection unit is used to obtain the historical collection period of the intelligent detection platform, obtain the historical user behavior records of each user within the historical collection period, and obtain user behavior data from the historical user behavior records; A data trend analysis unit is used to analyze the data trend of the behavior indicators of each user in the intelligent detection platform based on the user behavior data to obtain the target user behavior data; Among them, the positioning anomaly analysis module includes a coordinate anomaly scoring unit and a positioning anomaly analysis unit; A coordinate anomaly scoring unit is used to obtain coordinate data from the user's user location record and calculate the user's coordinate anomaly score in the current period based on the coordinate data; A positioning anomaly analysis unit is used to obtain sensor data from user positioning records, calculate the user's trajectory anomaly score, and aggregate the trajectory anomaly score and coordinate anomaly score to obtain positioning anomaly data; Among them, the virtual positioning detection module includes a behavior abnormality scoring unit and a virtual positioning detection unit; The behavior anomaly scoring unit is used to obtain the user's behavior records in the current cycle and calculate the user's behavior anomaly score; A virtual positioning detection unit is used to detect the user's virtual positioning situation based on the behavior abnormality score and combined with the positioning abnormality data to obtain a virtual positioning user; Wherein, the account management module includes an account management unit; The account management unit is used to obtain the historical positioning detection data of the virtual positioning user, and to manage the permissions of the virtual positioning user's account in combination with the account processing system information in the intelligent detection platform.

[0019] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A virtual positioning intelligent detection method based on multi-source data fusion, characterized in that: The method comprises: Step S100: Obtain historical user behavior records of different users in the intelligent detection platform, obtain user behavior data from the historical user behavior records, analyze data trends of behavior indicators of different users in the intelligent detection platform, and obtain target user behavior data; Step S200: obtaining user location records of the user in the intelligent detection platform, obtaining coordinate data and sensor data from the user location records, analyzing abnormal conditions of the user location, and obtaining location abnormality data; Step S300: Obtaining the user behavior record of the user in the current period, evaluating the abnormality of the user behavior based on the target user behavior data, and detecting the virtual positioning of the user in combination with the positioning abnormality data to obtain a virtual positioning user; Step S400: Acquire historical positioning detection data of the virtual positioning user, acquire account processing system data in the intelligent detection platform, acquire account data of the virtual positioning user, and perform authority management on the account of the virtual positioning user.

2. The method for intelligent virtual positioning detection based on multi-source data fusion according to claim 1, characterized in that: The step S100 includes: Step S101: obtaining a previous historical period of the current period, and recording the previous historical period as a collection historical period of the intelligent detection platform; Step S102: Obtaining historical user behavior records of each user in the intelligent detection platform within the historical collection period, preprocessing the historical user behavior records, and obtaining user behavior data from the historical user behavior records, wherein the user behavior data includes values corresponding to various behavior indicators of the user; Step S103: Analyze the data trend of the behavior indicators of each user in the intelligent detection platform, wherein the data trend analysis process of the behavior indicator item a of each user is as follows: Get the average value μ of the a-th behavior indicator of each user a and standard deviation σ a , get the mth range C of the ath behavior indicator (a,m) =[μ a -m×σ a ,μ a +m×σ a ], obtain the value of the a-th behavior indicator in the m-th range C (a,m) The total number of users in U (a,m) , obtain the total number U of each user in the intelligent detection platform SUM , calculate the mth range C (a,m) User coverage ratio B (a,m) =U (a,m) / U SUM ; Set a coverage ratio threshold B´, when B (a,m) < B´ and the user coverage ratio B of the (m + 1)-th range C of the a-th behavior indicator (a,m+1) (a,m+1) > B´, record the (m + 1)-th range C (a,m+1) as the target range of the a-th behavior indicator, and determine that the data trend of the a-th behavior indicator of each user in the current period is within the target range;​ Step S104: Obtain target ranges of various behavioral indicators of the intelligent detection platform in the current cycle and aggregate them to obtain target user behavior data.

3. The method for intelligent virtual positioning detection based on multi-source data fusion according to claim 2 is characterized in that: The step S200 includes: Step S201: using the intelligent detection platform to obtain the user location record of the user in the current period and analyze the abnormal situation of the user location. The specific process is as follows: Acquire coordinate data from the user location record, the coordinate data including the user's GPS coordinate location set and base station location set; Step S202: Time-aligning the GPS coordinate positioning set and the base station positioning set according to timestamps, so that after time alignment, the total number of elements in the GPS coordinate positioning set and the base station positioning set are the same; The coordinates corresponding to the elements in the user's GPS coordinate positioning set and the base station positioning set are converted into angle distances, wherein the coordinates corresponding to the e-th element in the GPS coordinate positioning set (lat （e,gps） ,lon （e,gps） )Angle distance conversion, get (φ （e,gps） ,λ （e,gps） ), the specific formula is: （e,gps） =lat （e,gps） × (π / 180), λ （e,gps） =lon （e,gps） × (π / 180); Step S203: Calculate the positioning distance difference between each element in the GPS coordinate positioning set and the base station positioning set, wherein the coordinate corresponding to the gth element in the GPS coordinate positioning set (φ （g,gps） ,λ （g,gps） ), and the coordinates corresponding to the gth element in the base station location set (φ （g,lbs） ,λ （g,lbs） ) between the positioning distance difference L g , the specific calculation process is: Obtain the cumulative value Q of the semi-sine function of the difference in longitude and latitude corresponding to the g-th element between the GPS coordinate positioning set and the base station positioning set g ; Calculate the angular distance S between the gth element of the GPS coordinate positioning set and the base station positioning set g : ， Calculate the positioning distance difference L g =S g ×R, R is the radius of the earth; Obtain an area containing the coordinates corresponding to the g-th element in the GPS coordinate positioning set and the base station positioning set, and obtain a preset positioning distance difference threshold in the area; When the positioning distance difference L g If the positioning distance difference is greater than the positioning distance difference threshold, it is determined that the user has a coordinate error, and the total number of coordinate errors of the user in the current cycle α is obtained. sum ; Get the total number W of elements in the GPS coordinate positioning set sum , calculate the user's coordinate anomaly score V=α in the current cycle sum / W sum ; Step S204: acquiring sensor data from the user positioning record, the sensor data including a set of direction angles collected by a direction sensor in the user's device in a current period; Time-aligning the GPS coordinate location set with the direction angle set; Step S205: The coordinates corresponding to the βth element in the GPS coordinate positioning set (φ （β,gps） ,λ （β,gps） ) to transform the plane coordinate system and obtain the coordinates (x （β,gps） ,y （β,gps） ), where x （β,gps） =K·λ （β,gps） ,y （β,gps） =K·φ （β,gps） , K is a preset conversion coefficient, and the coordinates corresponding to each element in the GPS coordinate positioning set are converted into a plane coordinate system; Get the direction angle θ of the βth element β ; Calculate the change in the direction angle of the βth element Δθ β =θ β+1 -θ β ,θ β+1 For the direction angle of the β+1th element, obtain the change in the direction angle of each element in the GPS coordinate positioning set; Step S206: Obtain the change in the direction angle of each element in the direction angle set, and calculate the direction angle change error between the direction angle set and each element in the GPS coordinate positioning set, wherein the change in the direction angle of the hth element in the direction angle set is Δθ´ h , and the direction angle △θ of the hth element in the GPS coordinate positioning set h The direction angle change error ε h =|△θ h -△θ´ h |; When the direction angle change error ε is greater than a preset error threshold, it is determined that the user has a direction change error. The ratio between the total number of direction change errors of the user in the current cycle and the total number of elements in the direction angle set is obtained and recorded as the trajectory anomaly score Z; Step S207: Obtain the trajectory anomaly score Z and coordinate anomaly score V of the user in the current period and aggregate them to obtain positioning anomaly data.

4. The method for intelligent virtual positioning detection based on multi-source data fusion according to claim 3 is characterized in that: The step S300 includes: Step S301: obtaining the values of various behavior indicators of the user in the current period from the user behavior record; Obtaining the target user behavior data of the intelligent detection platform in the current period, and obtaining the target range of each behavior indicator from the target user behavior data; When the value of a certain behavior indicator of the user is not within the target range corresponding to the certain behavior indicator, the certain behavior indicator is determined to be abnormal; Obtaining the ratio of the total number of behavioral indicators of the user determined to be abnormal to the total number of each behavioral indicator to obtain the user's behavioral abnormality score I; Detecting the user's virtual positioning situation, the specific detection process is as follows: Obtain the trajectory anomaly score Z and coordinate anomaly score V from the positioning anomaly data of the user, and calculate the virtual score P=η of the user's positioning in the current cycle Z ×Z+η I ×I+η V ×V, where η Z ,η I and η V The proportional coefficients preset for the trajectory abnormality score Z, the behavior abnormality score I, and the coordinate abnormality score V respectively; When the virtual score P is greater than a preset virtual score threshold, it is determined that the positioning of the user in the current period is virtual positioning, and the user is recorded as a virtual positioning user.

5. The method for intelligent virtual positioning detection based on multi-source data fusion according to claim 4 is characterized in that: The step S400 includes: Step S401: Obtain historical positioning detection data of the virtual positioning user from the intelligent detection platform, and obtain the total number of times N when the positioning of the virtual positioning user is determined to be virtual positioning from the historical positioning detection data; Step S402: Obtain the account data of the virtual positioning user, the account data including the data corresponding to the various account parameters of the virtual positioning user, and combine the account processing system information in the intelligent detection platform to generate a penalty strategy for the account of the virtual positioning user, perform authority management on the account of the virtual positioning user, and report the virtual positioning user to the staff in the intelligent detection platform.

6. A virtual positioning intelligent detection system based on multi-source data fusion, used to execute a virtual positioning intelligent detection method based on multi-source data fusion according to any one of claims 1 to 5, characterized in that: The system includes a data trend analysis module, a positioning anomaly analysis module, a virtual positioning detection module and an account management module; The data trend analysis module is used to obtain historical user behavior records of different users in the intelligent detection platform, analyze the data trend status of the behavior indicators of different users in the intelligent detection platform, and obtain target user behavior data; The positioning anomaly analysis module is used to obtain the user positioning record of the user, analyze the abnormal situation of the user positioning, and obtain positioning anomaly data; The virtual positioning detection module is used to evaluate the abnormality of the user's user behavior and, in combination with the positioning abnormality data, detect the user's virtual positioning situation to obtain a virtual positioning user; The account management module is used to obtain the historical positioning detection data of the virtual positioning user and perform authority management on the account of the virtual positioning user according to the account processing system data pre-examined by the intelligent detection platform.

7. The virtual positioning intelligent detection system based on multi-source data fusion according to claim 6 is characterized in that: The data trend analysis module includes a data acquisition unit and a data trend analysis unit; The data collection unit is configured to obtain a historical collection period of the intelligent detection platform, obtain historical user behavior records of each user within the historical collection period, and obtain user behavior data from the historical user behavior records; The data trend analysis unit is used to analyze the data trend status of the behavior indicators of each user in the intelligent detection platform according to the user behavior data to obtain the target user behavior data.

8. The virtual positioning intelligent detection system based on multi-source data fusion according to claim 6 is characterized in that: The positioning anomaly analysis module includes a coordinate anomaly scoring unit and a positioning anomaly analysis unit; The coordinate anomaly scoring unit is configured to obtain coordinate data from the user's location record and calculate a coordinate anomaly score of the user in a current period based on the coordinate data; The positioning anomaly analysis unit is configured to obtain sensor data from the user positioning record, calculate the user's trajectory anomaly score, and aggregate the trajectory anomaly score and the coordinate anomaly score to obtain positioning anomaly data.

9. The virtual positioning intelligent detection system based on multi-source data fusion according to claim 6 is characterized in that: The virtual positioning detection module includes a behavior abnormality scoring unit and a virtual positioning detection unit; The behavior anomaly scoring unit is used to obtain the user behavior record of the user in the current period and calculate the behavior anomaly score of the user; The virtual positioning detection unit is used to detect the virtual positioning situation of the user according to the behavior abnormality score and in combination with the positioning abnormality data to obtain a virtual positioning user.

10. The virtual positioning intelligent detection system based on multi-source data fusion according to claim 6, characterized in that: The account management module includes an account management unit; The account management unit is used to obtain the historical positioning detection data of the virtual positioning user, and perform authority management on the account of the virtual positioning user in combination with the account processing system information in the intelligent detection platform.

Images