User track protection method and device in crowdsourcing service, equipment and storage medium

Abstract

According to the user track protection method and device in crowdsourcing service, the equipment and the storage medium provided by the invention, firstly, the track data generated by the user is acquired, the privacy budget corresponding to the track data is calculated, the target position point in the track data is predicted, and the prediction result corresponding to the target position point is obtained; and then an error between the target position point and a prediction result corresponding to the target position point is calculated to obtain an importance evaluation result corresponding to the target position point, and finally the privacy budget is correspondingly allocated to the target position point according to the importance evaluation result. The privacy of the user can be effectively protected, the maximization of the data value can be ensured, and a safe and credible interaction relationship is established between the user and a service provider. By applying the Kalman filtering technology and comparing a predicted position point with an actual position point, the importance of the position point is evaluated, and a better privacy protection effect and higher data availability are achieved.

Description

Technical Field

[0001] The present application relates to a field, and in particular to a method, apparatus, device and storage medium for protecting user traces in crowdsourcing services. Background Art

[0002] With the popularization of mobile smart devices and the rapid development of crowdsourcing applications, the generated crowdsourcing data is increasingly closely related to the construction of smart cities. In the crowdsourcing service model, task requesters distribute tasks of specific locations or routes through centralized servers, generating a large amount of spatiotemporal data related to worker behavior patterns. Although the above data can effectively support task allocation, it also exposes users' sensitive information, such as location and movement trajectory, posing a privacy threat to users.

[0003] In related technologies, differential privacy is an effective method to protect location and trajectory privacy. It interferes with original data by randomly adding noise to ensure data availability. Furthermore, local differential privacy can process data locally and protect privacy, but over-protection will reduce data availability, especially when noise is added to each location point in the trajectory. Summary of the invention

[0004] In order to overcome the problems existing in the related art, the present application provides a multi-view realistic high-precision map method, device and storage medium, which receives the geographic location information input by the user, generates multiple view data according to the geographic location information and displays them on different display devices, so that the user can observe and analyze the geographic location information from multiple angles and dimensions and make better path planning.

[0005] The first aspect of the present application provides a method for protecting user tracks in a crowdsourcing service, comprising:

[0006] Obtaining trajectory data generated by a user and calculating a privacy budget corresponding to the trajectory data;

[0007] Predicting the target location point in the trajectory data to obtain a prediction result corresponding to the target location point;

[0008] Calculating the error between the target position point and the prediction result corresponding to the target position point to obtain the importance evaluation result corresponding to the target position point;

[0009] The privacy budget is allocated to the target location point according to the importance evaluation result.

[0010] Optionally, it includes: predicting the target position point in the trajectory data based on a Kalman filter algorithm.

[0011] Optionally, the prediction process includes:

[0012] Inputting the state vector of the target position point into the prediction model to obtain an initial prediction result;

[0013] Comparing the error between the initial prediction result and the current target position point to obtain a comparison result;

[0014] The state prediction weight is adjusted based on the comparison result until the iteration stop condition is met, and the prediction result that meets the iteration condition is output.

[0015] Optionally, the condition for stopping iteration is that an error between a prediction result of the current target position point calculated according to the Kalman filter algorithm and the current target position point is less than a preset error threshold.

[0016] Optionally, allocating the privacy budget to the target location point according to the importance evaluation result includes:

[0017] Determine a privacy budget allocation weight value for the target location point according to an importance evaluation result of the target location point;

[0018] The privacy budget is allocated to each target location point according to the weight value.

[0019] Optionally, determining the privacy budget allocation weight of the target location point includes:

[0020] Determine the weight ratio according to the relative size of the importance evaluation result of the target location point and the importance evaluation results of other target location points;

[0021] The privacy budget is allocated to each target location point according to the weight ratio.

[0022] Optionally, after allocating the privacy budget to each target location point according to the weight ratio, the method further includes:

[0023] According to the importance evaluation result of the target location point and the privacy budget allocated to the target location point, the false location point to be inserted is determined.

[0024] A second aspect of the present application is a user track protection device in a crowdsourcing service, comprising:

[0025] An acquisition module, used to acquire trajectory data generated by a user and calculate a privacy budget corresponding to the trajectory data;

[0026] A prediction module, used to predict the target position point in the trajectory data to obtain a prediction result corresponding to the target position point;

[0027] A calculation module, used to calculate the error between the target position point and the prediction result corresponding to the target position point, and obtain the importance evaluation result corresponding to the target position point;

[0028] An allocation module is used to allocate the privacy budget to the target location point according to the importance evaluation result.

[0029] A third aspect of the present application provides a non-transitory machine-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor is caused to execute the method as described above.

[0030] A fourth aspect of the present application provides a computer-readable storage medium having executable code stored thereon. When the executable code is executed by a processor of an electronic device, the processor is caused to execute the method as described above.

[0031] It can be seen that the user trajectory protection method, device, equipment and storage medium in a crowdsourcing service provided by the present application first obtain the trajectory data generated by the user, and calculate the privacy budget corresponding to the trajectory data, predict the target location point in the trajectory data, and obtain the prediction result corresponding to the target location point, and then calculate the error between the target location point and the prediction result corresponding to the target location point to obtain the importance evaluation result corresponding to the target location point, and finally allocate the privacy budget to the target location point according to the importance evaluation result. It can not only effectively protect the privacy of the user, but also ensure the maximization of the data value, and establish a safe and reliable interactive relationship between the user and the service provider. By applying Kalman filtering technology and comparing the predicted location point with the actual location point to evaluate its importance, better privacy protection effect and higher data availability are achieved.

[0032] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The above and other objects, features and advantages of the present application will become more apparent through a more detailed description of exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0034] Figure 1 It is a flowchart of a method for protecting user tracks in a crowdsourcing service shown in an embodiment of the present application.

[0035] Figure 2 It is a structural diagram of a trajectory-aware query and privacy protection framework shown in an embodiment of the present application.

[0036] Figure 3 It is a schematic diagram of the Geolife dataset experiment shown in the embodiment of the present application.

[0037] Figure 4 It is a schematic diagram of the Gowalla data set experiment shown in the embodiment of the present application.

[0038] Figure 5 It is a schematic diagram of the Shanghai data set experiment shown in the embodiment of the present application.

[0039] Figure 6 It is a structural schematic diagram of a user trajectory protection device in a crowdsourcing service shown in an embodiment of the present application.

[0040] Figure 7 It is a schematic diagram of the structure of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0041] The preferred embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0042] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0043] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0044] With the popularity of smartphones, smart bracelets and other mobile smart devices, crowdsourced data generated by individuals plays an increasingly important role in the construction of smart cities. The rapid development of crowdsourcing applications has led to the extensive collection and analysis of user behavior data. The availability of this data is crucial to the realization of various functions of smart cities. However, the above process is also accompanied by serious privacy and security challenges, such as the potential risk of leakage of sensitive privacy data of users or workers, especially their location and movement trajectory information.

[0045] In the crowdsourcing service model, task requesters usually distribute tasks of specific target locations or routes through a centralized third-party server and assign tasks to individual workers. The process generates a large amount of spatiotemporal data related to the behavior patterns of workers. Although these data provide an effective means for the server to optimize task allocation, they may also pose a major threat to user privacy. Even a small amount of mobile path or location data leakage may cause potential attacks on anonymous users and pose risks to personal privacy.

[0046] For example, in the scenario of calling a taxi, both the driver and the passenger need to upload their personal location to the server before using the application, and the server will match the appropriate passenger and driver. For the driver, by uploading relevant location information for a long time to obtain the order tasks provided by the server, it is easy to expose the driver's personal residence area and related information. The server can further calculate the driver's home address and other location information by analyzing the driver's personal interaction range. Therefore, it poses a risk of privacy leakage to both the driver and the passenger.

[0047] To address the above privacy threats, differential privacy is widely regarded as an effective protection measure. Differential privacy adds random perturbation noise to the location points in the trajectory data to ensure that data consumers cannot accurately obtain the original data, while maintaining a certain accuracy in the statistical results, and still providing valuable information to service providers. Common differential privacy mechanisms, such as global differential privacy and local differential privacy, ensure that the privacy budget is controlled within a certain range during the data collection and release process to protect users' sensitive information.

[0048] However, in applications, users still have concerns about centralized data collection, believing that the process may pose a major threat to their privacy. Local differential privacy mechanisms allow users to modify their data locally and then transmit the processed data to an untrusted third-party server, thereby reducing potential privacy risks. However, existing local differential privacy mechanisms often add noise to all trajectory location points in the process of protecting privacy. Although this enhances privacy protection, it also significantly reduces the availability of data and affects the overall utility of the application.

[0049] Therefore, the present application provides a method, device, equipment and storage medium for protecting user tracks in crowdsourcing services. Under the premise of protecting user privacy as much as possible, the availability of data is improved, and the specific needs of application scenarios are taken into consideration to ensure that users with disabilities get a better experience in crowdsourcing services.

[0050] See also Figure 1 , the present application provides a method for protecting user tracks in a crowdsourcing service, comprising:

[0051] S100: Acquire trajectory data generated by a user, and calculate a privacy budget corresponding to the trajectory data.

[0052] In an embodiment of the present application, the trajectory data generated by the user can be collected from the user's mobile smart device (such as a smart phone, smart bracelet, etc.). The trajectory data may include information such as the user's real-time location, movement route, and dwell time. Among them, the trajectory data can collect the user's location data through the user's mobile smart device and connect the location points in chronological order.

[0053] In this embodiment, the privacy budget is a quantitative concept used to manage user privacy during data processing. For example, the distribution standard of user privacy information can be calculated and controlled during data analysis. The privacy budget described below is represented by ε, where a smaller privacy budget ε value is used to represent a higher privacy protection requirement, and a larger privacy budget ε value is used to represent a lower privacy protection requirement.

[0054] For example, in practical applications, user personal information and behavior data are evaluated as highly sensitive, that is, a smaller privacy budget ε value is used, such as 0.1. The Laplace mechanism can be used to add random noise. In the process of calculating the privacy budget ε value, for each query (for example, the total count of user activities), the privacy budget ε1 consumed during the analysis process consumes 0.01 per percentage point; for user behavior pattern analysis, the consumed privacy budget can be ε2, which consumes 0.05 per analysis and a total budget of 0.1. When executing data queries, the total budget is tracked to ensure that the privacy budget ε does not exceed the set 0.1 after each analysis step.

[0055] S101, predicting a target position point in the trajectory data to obtain a prediction result corresponding to the target position point;

[0056] S102, calculating the error between the target position point and the prediction result corresponding to the target position point, and obtaining the importance evaluation result corresponding to the target position point;

[0057] Furthermore, it includes: predicting the target position point in the trajectory data based on a Kalman filter algorithm.

[0058] The prediction process includes: inputting the state vector of the target position point into the prediction model to obtain an initial prediction result, comparing the error between the initial prediction result and the current target position point to obtain a comparison result, adjusting the state prediction weight based on the comparison result until the stop iteration condition is met, and outputting the prediction result that meets the iteration condition.

[0059] In an embodiment of the present application, a Kalman filter algorithm is used to predict the user's trajectory data. Kalman filter is a recursive algorithm used to estimate the state of a dynamic system in the presence of noise.

[0060] Specifically, Figure 2 As shown, the core of the Kalman filter algorithm lies in its two main steps, including: prediction and update.

[0061] On the one hand, in the prediction stage, based on the current state and its prediction model, the state at the next moment is predicted, where the prediction model can include a state model and a state noise model. In the prediction stage of state prediction, the predicted values ​​of the state vector and covariance matrix can be obtained by the following formula:

[0062]

[0063] In formula (1), k is the round of the current iteration, F k is the state transfer matrix, which is used to describe how the state evolves. is the estimated value of the state vector at the kth iteration, B k is the state noise transfer matrix, which is used to describe how the state noise evolves. k is the state noise vector, P k|k-1 is the delayed estimation error covariance matrix, Q k is the process noise covariance matrix.

[0064] In the embodiment of the present application, the condition for stopping iteration is that the error between the prediction result of the current target position point calculated according to the Kalman filter algorithm and the current target position point is less than a preset error threshold.

[0065] It can be understood that in each iteration, the Kalman filter predicts the target position based on the system's dynamic model and past state information, and the iterative process terminates when the calculated error is less than the error threshold preset by the system.

[0066] On the other hand, in the update phase, when new observations are obtained, the Kalman filter enters the update phase and adjusts the previous state predictions according to the new observations. In this process, the system compares the predictions of the previous round with the actual observations to calculate the error. Based on this error, the system will adjust the weight of the state prediction to make the prediction closer to the observation. At the same time, the system will also update the covariance matrix to reflect the corrected uncertainty. The updated values ​​of the state vector and covariance matrix can be obtained by the following formula:

[0067]

[0068] In formula (2), K k is the intermediate variable, H k is the observation matrix, which is used to describe the relationship between the observation value and the state vector. k is the current actual observation value, R k is the covariance matrix of the observation noise, which represents the uncertainty of the observation model.

[0069] Using the updated value of the previous round as the input for the next round of prediction, the Kalman filter can gradually approach the optimal state estimate of the system through continuous iterations and use it as the key input in the importance assessment stage.

[0070] In the importance evaluation stage, the PID error can be calculated using the following formula to determine the importance of each position point γ[i]:

[0071]

[0072] In formula (3), F i is the error between the predicted point and the true point, and defines x i is the real location point, To predict the location point, it can be calculated by the following formula:

[0073]

[0074] In formula (4), F i is the error between the predicted point and the true point, and defines x i is the real location point, is the predicted location point.

[0075] It can be seen that through the Kalman filter algorithm, the system can effectively predict and update the target position, thereby achieving accurate tracking of the target. The core of the Kalman filter algorithm is that it can combine the system model and the actual measurement value, and continuously optimize the state estimation value and the error covariance matrix through an iterative process. When the error between the predicted result and the actual measurement value is less than the preset error threshold, it means that the current prediction result is accurate enough and no further iteration is required, thereby saving computing resources and improving system efficiency.

[0076] S103: Allocate the privacy budget to the target location point according to the importance evaluation result.

[0077] In the embodiment of the present application, a corresponding privacy budget is allocated to each target location point according to the importance evaluation result of the target location point derived in S102. For those locations with high importance, more privacy budget is allocated to enhance protection measures, while for locations with low importance, the budget can be reduced to ensure the validity of the overall data.

[0078] In order to add different adaptive privacy budgets to each location point, a trajectory-aware randomization mechanism is adopted, which adds random Laplace noise to it based on the importance of the location point. For any given input x, the Laplace mechanism is defined as:

[0079]

[0080] In formula (5), b is the scale parameter of the Laplace mechanism, which is used to determine the shape of the Laplace distribution. We use the following formula to calculate the value of b and associate it with the privacy budget ε:

[0081]

[0082] In formula (6), Δφ is the sensitivity of differential privacy, which is defined as the maximum distance between adjacent position points in the trajectory:

[0083] Δφ=max||x i ,x i-1 ||1………………(7)

[0084] For trajectory sequence data, ω-EP requires that the cumulative privacy budget of consecutive ω events does not exceed ε. Therefore, if a different privacy budget is assigned to each location point, the allocable privacy budget for each location point will not be exactly the same. ′ represents the remaining allocatable privacy budget at each location point, and ε[i] represents the privacy budget actually allocated to each location point, then we have the following formula:

[0085]

[0086] In order to better preserve the trajectory pattern, for points with higher importance, it is necessary to add as little noise as possible, that is, allocate a larger privacy budget. Therefore, we can calculate the privacy budget allocated to each location point by the following formula:

[0087]

[0088] In formula (9), p represents the ratio of the actually allocated privacy budget to the remaining privacy budget, and the exponential function ensures that the value of p ranges from 0 to 1. At the same time, since the input of PID is obtained by Kalman filtering, the problem of continuous large p values ​​leading to rapid consumption of the remaining privacy budget can be avoided.

[0089] It can be seen that through dynamic privacy budget management, that is, dynamically calculating the privacy budget based on the trajectory data generated by the user end and evaluating the importance of different location points, accurate privacy management can be achieved, ensuring that the user's important location information is protected at a higher level and reducing the potential risk of privacy leakage.

[0090] Secondly, by using the Kalman filter algorithm to predict the target location, the prediction error can be effectively reduced and the dynamic tracking capability of the user's trajectory can be improved. This makes the system's real-time response to user movement more accurate and enhances the practicality of the service. By evaluating the importance of the target location point, the corresponding privacy budget is adjusted and allocated. Location points of different importance will receive different degrees of privacy protection, allowing users to enjoy a good privacy protection experience in different scenarios.

[0091] In an embodiment of the present application, after allocating the privacy budget to each target location point according to the weight ratio, it also includes: determining the false location point to be inserted according to the importance evaluation result of the target location point and the privacy budget allocated to the target location point.

[0092] It can be understood that since the target position point is two-dimensional data and Laplace noise is one-dimensional data, the system can convert the result concept of Laplace noise into a preset range of a circle centered on the target position point, and the disturbed false position point falls at a random point within the circle to achieve the purpose of adding noise.

[0093] According to the importance assessment of the target location point and the allocated privacy budget, the fake location points that need to be inserted are determined. The fake location points will add noise around the real target location, making it more difficult for attackers to identify the real user location, thereby improving the privacy protection effect.

[0094] In an embodiment of the present application, allocating the privacy budget to the target location point according to the importance assessment result includes: determining a privacy budget allocation weight value for the target location point according to the importance assessment result for the target location point; and allocating the privacy budget to each target location point according to the weight value.

[0095] In an embodiment of the present application, determining the privacy budget allocation weight of the target location point includes: determining a weight ratio based on the relative size of an importance assessment result of the target location point and importance assessment results of other target location points; and allocating the privacy budget to each target location point according to the weight ratio.

[0096] In the embodiment of the present application, in order to verify the effectiveness of the framework in different scenarios, this embodiment sets three sets of data sets: the first set of Geolife data sets collects a total of 17,621 trajectory data of 182 users, mainly capturing the daily travel and leisure activities of users, covering different sampling rates and 91% of which are dense sampling;

[0097] The second Gowalla dataset collects 6,442,890 check-in data from 196,591 users. Since this dataset is not a trajectory dataset, we treat a group of points with similar check-in times of the same user as a trajectory.

[0098] The third Shanghai dataset collects 100,000 GPS trajectory data of Shanghai taxis in 2007. The spatial span between location points in this dataset is larger than that in other datasets.

[0099] For each data set, 5 trajectories are randomly selected for calculation, and the average value is calculated after 5000 repetitions to make the experimental results as close to the characteristics of the data set as possible.

[0100] The performance of the algorithm is evaluated by setting two indicators: L1 distance and time series similarity (DTW, Dynamic Time Warping). L1 distance is used to measure the absolute difference in spatial distance between location points:

[0101]

[0102] In formula (10), x i ,y i and Represent the horizontal and vertical coordinates of the real position point and the position point after disturbance respectively. The time series similarity is used to measure the similarity of two trajectories in time series:

[0103]

[0104] In formula (11), T i and They represent the i-th true trajectory and the j-th perturbed trajectory respectively.

[0105] The parameter settings in the experiment include:

[0106] K p :0.1;K i :0.15;K d : 0.1

[0107] ε: 1, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8

[0108] Comparison methods: TALDP, LSRR, GRR, PLDP

[0109] The relationship between privacy budget ε and L1 distance: Figure 3 , 4 , 5 show that TALDP is generally better than other methods in L1 distance on the three datasets, indicating that TALDP can achieve the same degree of privacy protection by adding less noise, which can effectively improve data availability. However, TALDP tends to decline rapidly when the privacy budget increases. This is because TALDP allocates the privacy budget in a certain proportion according to the importance of the location point. When the privacy budget increases, the privacy budget allocated to the location point with a greater impact on the L1 distance increases more, which makes the L1 distance change significantly when the privacy budget is small.

[0110] The relationship between privacy budget ε and DTW: Figure 4 The DTW of TALDP on the three datasets is generally better than other methods. DTW reflects the similarity between trajectories, which shows that TALDP can effectively preserve trajectory patterns, maintain trajectory characteristics, and improve data availability. At the same time, it can be found that when the privacy budget is small, LDPTPM and TALDP have similar DTW values, but TALDP is still ahead. As the privacy budget increases, this lead is further expanded. This is because TALDP considers the integrity of the trajectory more than LDPTPM, making the overall effect better.

[0111] It can be seen that by comparing TALDP with the pure random method, Figures 3 to 5It shows that TALDP has a significant improvement on the three data sets compared to the pure random method, and it can be noticed that when the privacy budget is small, the advantage of TALDP is very obvious, indicating that TALDP can effectively improve data availability. As the privacy budget increases, the advantage of TALDP gradually decreases, because when the privacy budget is sufficient, each location point can be allocated a larger privacy budget, and the advantage of the specific algorithm is relatively small at this time.

[0112] like Figure 6 As shown, a user track protection device in a crowdsourcing service includes:

[0113] An acquisition module 61 is used to acquire the trajectory data generated by the user terminal and calculate the privacy budget corresponding to the trajectory data;

[0114] A prediction module 62, used to predict the target position point in the trajectory data to obtain a prediction result corresponding to the target position point;

[0115] A calculation module 63, used to calculate the error between the target position point and the prediction result corresponding to the target position point, and obtain the importance evaluation result corresponding to the target position point;

[0116] The allocation module 64 is used to allocate the privacy budget to the target location point according to the importance evaluation result.

[0117] The prediction module 62 is used to predict the target position point in the trajectory data based on the Kalman filter algorithm.

[0118] The prediction module 62 is also used to:

[0119] Inputting the state vector of the target position point into the prediction model to obtain an initial prediction result;

[0120] Comparing the error between the initial prediction result and the current target position point to obtain a comparison result;

[0121] The state prediction weight is adjusted based on the comparison result until the iteration stop condition is met, and the prediction result that meets the iteration condition is output.

[0122] The prediction module 62 is also used to:

[0123] The condition for stopping iteration is that the error between the prediction result of the current target position point calculated according to the Kalman filter algorithm and the current target position point is less than a preset error threshold.

[0124] The allocation module 64 is used to:

[0125] Determine a privacy budget allocation weight value for the target location point according to an importance evaluation result of the target location point;

[0126] The privacy budget is allocated to each target location point according to the weight value.

[0127] The allocation module is also used to:

[0128] Determine the weight ratio according to the relative size of the importance evaluation result of the target location point and the importance evaluation results of other target location points;

[0129] The privacy budget is allocated to each target location point according to the weight ratio.

[0130] Optionally, after allocating the privacy budget to each target location point according to the weight ratio, the method further includes:

[0131] According to the importance evaluation result of the target location point and the privacy budget allocated to the target location point, the false location point to be inserted is determined.

[0132] The present application provides a user trajectory protection device in a crowdsourcing service, which first obtains the trajectory data generated by the user, calculates the privacy budget corresponding to the trajectory data, predicts the target location point in the trajectory data, obtains the prediction result corresponding to the target location point, and then calculates the error between the target location point and the prediction result corresponding to the target location point to obtain the importance evaluation result corresponding to the target location point, and finally allocates the privacy budget to the target location point according to the importance evaluation result. It can effectively protect the privacy of the user and ensure the maximization of the data value, and establish a safe and reliable interactive relationship between the user and the service provider. By applying Kalman filtering technology and comparing the predicted location point with the actual location point to evaluate its importance, better privacy protection effect and higher data availability are achieved.

[0133] See also Figure 7 , Figure 7 It is a schematic diagram of the structure of an electronic device shown in an embodiment of the present application. The electronic device 700 includes a memory 710 and a processor 720.

[0134] The processor 720 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 gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0135] The memory 710 may include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. Among them, ROM can store static data or instructions required by the processor 720 or other modules of the computer. The permanent storage device may be a readable and writable storage device. The permanent storage device may be a non-volatile storage device that does not lose the stored instructions and data even after the computer is powered off. In some embodiments, the permanent storage device uses a large-capacity storage device (such as a magnetic or optical disk, flash memory) as a permanent storage device. In some other embodiments, the permanent storage device may be a removable storage device (such as a floppy disk, optical drive). The system memory may be a readable and writable storage device or a volatile readable and writable storage device, such as a dynamic random access memory. The system memory may store some or all instructions and data required by the processor at run time. In addition, the memory 710 may include any combination of computer-readable storage media, including various types of semiconductor memory chips (DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and disks and / or optical disks may also be used. In some embodiments, the memory 710 may include a readable and / or writable removable storage device, such as a laser disc (CD), a read-only digital versatile disc (e.g., DVD-ROM, double-layer DVD-ROM), a read-only Blu-ray disc, an ultra-density optical disc, a flash memory card (e.g., SD card, mini SD card, Micro-SD card, etc.), a magnetic floppy disk, etc. The computer-readable storage medium does not include carrier waves and transient electronic signals transmitted wirelessly or wired.

[0136] The memory 710 stores executable codes, and when the executable codes are processed by the processor 720 , the processor 720 can execute part or all of the methods described above.

[0137] The scheme of the present application has been described in detail above with reference to the accompanying drawings. In the above embodiments, the description of each embodiment has its own emphasis. For the part that is not described in detail in a certain embodiment, refer to the relevant description of other embodiments. Those skilled in the art should also know that the actions and modules involved in the description are not necessarily required for the present application. In addition, it can be understood that the steps in the method of the embodiment of the present application can be adjusted in order, merged and deleted according to actual needs, and the modules in the device of the embodiment of the present application can be merged, divided and deleted according to actual needs.

[0138] In addition, the method according to the present application may also be implemented as a computer program or a computer program product, which includes computer program code instructions for executing some or all of the steps in the above method of the present application.

[0139] Alternatively, the present application can also be implemented as a non-temporary machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) on which executable code (or computer program, or computer instruction code) is stored. When the executable code (or computer program, or computer instruction code) is executed by a processor of an electronic device (or electronic device, server, etc.), the processor executes part or all of the steps of the above-mentioned method according to the present application.

[0140] Those skilled in the art will further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the application herein may be implemented as electronic hardware, computer software, or combinations of both.

[0141] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system and method according to multiple embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a part of a module, a program segment or a code, and the part of the module, the program segment or the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or the flow chart, and the combination of the square boxes in the block diagram and / or the flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0142] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for protecting user tracks in crowdsourcing services, characterized in that: include: Obtaining trajectory data generated by a user and calculating a privacy budget corresponding to the trajectory data; Predicting the target location point in the trajectory data to obtain a prediction result corresponding to the target location point; Calculating the error between the target position point and the prediction result corresponding to the target position point to obtain the importance evaluation result corresponding to the target position point; The privacy budget is allocated to the target location point according to the importance evaluation result.

2. The method according to claim 1, characterized in that include: The target position point in the trajectory data is predicted based on the Kalman filter algorithm.

3. The method according to claim 2, characterized in that The prediction process includes: Inputting the state vector of the target position point into the prediction model to obtain an initial prediction result; Comparing the error between the initial prediction result and the current target position point to obtain a comparison result; The state prediction weight is adjusted based on the comparison result until the iteration stop condition is met, and the prediction result that meets the iteration condition is output.

4. The method according to claim 3, characterized in that: The condition for stopping iteration is that the error between the prediction result of the current target position point calculated according to the Kalman filter algorithm and the current target position point is less than a preset error threshold.

5. The method according to claim 1, characterized in that The allocating the privacy budget to the target location point according to the importance evaluation result includes: Determine a privacy budget allocation weight value for the target location point according to an importance evaluation result of the target location point; The privacy budget is allocated to each target location point according to the weight value.

6. The method according to claim 5, characterized in that The determining of the privacy budget allocation weight of the target location point includes: Determine the weight ratio according to the relative size of the importance evaluation result of the target location point and the importance evaluation results of other target location points; The privacy budget is allocated to each target location point according to the weight ratio.

7. The method according to claim 6, characterized in that After allocating the privacy budget to each target location point according to the weight ratio, the method further includes: According to the importance evaluation result of the target location point and the privacy budget allocated to the target location point, the false location point to be inserted is determined.

8. A user track protection device in crowdsourcing service, characterized in that: include: An acquisition module, used to acquire trajectory data generated by a user and calculate a privacy budget corresponding to the trajectory data; A prediction module, used to predict the target position point in the trajectory data to obtain a prediction result corresponding to the target position point; A calculation module, used to calculate the error between the target position point and the prediction result corresponding to the target position point, and obtain the importance evaluation result corresponding to the target position point; An allocation module is used to allocate the privacy budget to the target location point according to the importance evaluation result.

9. An electronic device, characterized in that: include: processor; as well as A memory having executable codes stored thereon, which, when executed by the processor, causes the processor to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having executable code stored thereon, characterized in that: When the executable code is executed by a processor of an electronic device, the processor is caused to execute the method as claimed in any one of claims 1 to 7.

Images