Track privacy data protection and sharing method

Abstract

The invention relates to the technical field of track data privacy protection, and discloses a track privacy data protection and sharing method, which comprises the following steps: carrying out sensitivity grading according to track data and historical access information thereof; identifying the behavior of each access user through an event trigger, and determining a risk value of each access user according to an identification result; and dynamically adjusting differential privacy parameters of the track data according to the risk value of each access user and the sensitivity of the access track data. According to the method, sensitivity grading is carried out on the track data, meanwhile, the risk state of the access user is judged, the risk value is obtained, and track data privacy protection in each data sharing access process is dynamically adjusted through sensitivity grading and the risk value of each access user, so that track data privacy protection is realized; and the privacy protection and sharing requirements of the trajectory data are balanced.

Description

Technical Field

[0001] The present application relates to the technical field of trajectory data privacy protection, and in particular to a method for protecting and sharing trajectory privacy data. Background Art

[0002] Trajectory data is data information obtained by sampling the motion process of one or more mobile objects in a spatiotemporal environment, including the sampling point location, sampling time, speed, etc. These sampling point data information constitute trajectory data according to the sampling order. Since trajectory data may contain sensitive information (such as some sensitive areas visited by the visiting user), or other personal information (such as the visiting user's home address, workplace, health status, living habits, etc.) can be inferred from the running trajectory, the privacy protection of trajectory information is crucial.

[0003] Existing privacy protection methods for trajectory data mainly use split privacy methods, which add mathematical noise (such as Laplace noise) to the trajectory data to confuse the real trajectory data, ensuring that attackers cannot restore sensitive information through statistical inference.

[0004] When protecting trajectory data privacy through split privacy, noise injection parameters are set. However, when the noise injection amount is small, the protection of trajectory data privacy is weak. When the noise injection amount is large, although it can effectively protect the privacy of trajectory data, it affects the availability of data during data sharing. Therefore, how to balance the privacy protection and sharing requirements of trajectory data is the fundamental problem to be solved by this invention. Summary of the Invention

[0005] In order to balance the privacy protection and sharing requirements of trajectory data, this application provides a method for protecting and sharing trajectory privacy data.

[0006] The following technical solutions are adopted:

[0007] Methods for protecting and sharing trajectory privacy data include:

[0008] Sensitivity classification based on trajectory data and its historical access information;

[0009] Identify the behavior of each accessing user through event triggers and determine the risk value of each accessing user based on the identification results;

[0010] Dynamically adjust the differential privacy parameters of the trajectory data based on the risk value of each accessing user and the sensitivity of their access trajectory data;

[0011] The process of sensitivity grading trajectory data includes:

[0012] AI-based conditional recognition and spatial dimension recognition of trajectory data to obtain the depth of associated information and spatial accuracy;

[0013] Perform time dimension analysis on trajectory data to obtain time granularity values;

[0014] The sensitivity benchmark index of trajectory data is obtained by weighted summing the correlation information depth, spatial accuracy and time granularity values;

[0015] Obtaining a dynamic risk index of the trajectory data based on historical access information of the trajectory data;

[0016] The sensitivity level of trajectory data is determined based on the sensitivity benchmark index and dynamic risk index of trajectory data.

[0017] By adopting the above technical solution, the sensitivity of trajectory data is judged and the trajectory data is graded based on the judgment results. At the same time, the risk status of the accessing user is judged to obtain a risk value. The privacy protection of trajectory data in each data sharing access process is dynamically adjusted based on the sensitivity classification and the risk value of each accessing user. When the data sensitivity is low and the accessing user risk is low, the differential privacy parameter of the trajectory data is lowered to improve the efficiency of trajectory data sharing. When the data sensitivity is high and the accessing user risk is high, the differential privacy parameter of the trajectory data is increased to achieve privacy protection of trajectory data, thus balancing the privacy protection and sharing requirements of trajectory data. The sensitivity baseline index of the trajectory data is obtained by comprehensively evaluating different factors of trajectory data sensitivity. The risk status of the trajectory data is evaluated based on the historical access information of the trajectory data, and the dynamic risk index of the trajectory data is obtained based on the historical access information of the trajectory data. Therefore, the dynamic risk index reflects the risk status of the trajectory data over a period of time. The sensitivity classification of the trajectory data is determined based on the sensitivity baseline index and the dynamic risk index of the trajectory data. The sensitivity classification of the trajectory data can be adjusted in real time according to the content of the trajectory data and the real-time access data of the trajectory data, thereby providing a relatively accurate reference basis for the dynamic adjustment of its differential privacy parameter.

[0018] Optionally, the process of acquiring the depth of the association information includes:

[0019] pass Calculate the depth of the associated information D, where n is the number of items identified by the condition, i∈[1,n], is the judgment value of the i-th condition identification. When the i-th condition identification result is met, ;otherwise, ; The weight value identified for the i-th condition;

[0020] The spatial accuracy acquisition process includes: determining the accuracy level of the position coordinates in the trajectory information according to the results of AI spatial dimension recognition, and determining the corresponding spatial accuracy according to the accuracy level;

[0021] The process of obtaining the time granularity value includes:

[0022] pass Calculate the time granularity value G, where T is the timestamp precision level. is the average time interval between timestamps, s is the standard deviation of the time interval between consecutive timestamps, is the first control function, is the second control function, is the third comparison function.

[0023] By adopting the above technical solution, the degree of privacy involved in the trajectory data can be judged based on the depth of associated information D; the sensitivity of the trajectory data can be judged based on the spatial accuracy; and the degree of influence of timestamp accuracy, data continuity, and the average time interval between timestamps on the sensitivity of the trajectory data can be judged through the acquisition process of the time granularity value G.

[0024] Optionally, the process of obtaining the dynamic risk index of trajectory data includes:

[0025] Based on the historical access information of the trajectory data in the preset period before the current time point, a curve q(t) of the cumulative number of visits over time is established, and the maximum slope K of q(t) in the preset period is obtained;

[0026] By formula Calculate the dynamic risk index ; Where m is the total number of visits within the preset period, is the slope reference value, is the first adjustment coefficient, is the standard deviation of the interval lengths between consecutive access time points, is the standard deviation reference value.

[0027] By adopting the above technical solution, the dynamic risk index is obtained by , which can judge the dynamic risk of trajectory data and adjust the level of privacy protection accordingly.

[0028] Optionally, the process of determining the sensitivity classification of the trajectory data further includes:

[0029] pass The sensitivity index R is calculated and the corresponding sensitivity grade is determined according to the interval of the sensitivity index R;

[0030] in, is the sensitivity benchmark index, for The normalized value of for The normalized value of .

[0031] By adopting the above technical solution, the sensitivity index R is obtained by combining the sensitivity benchmark index with the dynamic risk index to evaluate the sensitivity state of the trajectory data and classify it.

[0032] Optionally, the process of determining the risk value of each access user includes:

[0033] Determine the baseline risk level of each access user based on account information:

[0034] When the baseline risk level of the access user is low, the baseline risk value of the access user corresponding to the low risk level is used as the access user risk value;

[0035] Otherwise, the access user risk value is determined according to the access user's baseline risk value corresponding to the access user's baseline risk level and the access information.

[0036] By adopting the above technical solution, the efficiency of access data risk judgment is improved through the preliminary screening process of access user risks.

[0037] Optionally, when the access user is not at a low risk level, the process of determining the access user risk value includes:

[0038] Based on the event trigger, it is determined whether the access user behavior triggers an event preset in the event trigger. When the event is triggered, a first increment of the corresponding risk value is determined according to the triggered event;

[0039] Determine the second increment of the risk value based on the access information of the access user. The process of calculating the second increment of the risk value includes obtaining a curve W(t) showing the cumulative number of accesses over time in a preset period before the current time point of the access user, and determining the maximum slope of the curve W(t) in the preset period based on the cumulative number of accesses over time. , through the formula Calculate the second increment of risk value ; Where W is the cumulative number of visits at the current time point, is the visit count threshold, is the second adjustment coefficient, It is the reference value of visit volume slope;

[0040] The access user's baseline risk value, the first incremental risk value, and the second incremental risk value are normalized and accumulated to obtain the access user's risk value.

[0041] By adopting the above technical solution, the security risk of the accessing user is determined by the accessing user risk value. According to the risk of the accessing user and the sensitivity classification of the trajectory data, the differential privacy parameters of the trajectory data can be determined, and the privacy protection and sharing requirements of the trajectory data can be balanced.

[0042] Optionally, the process of dynamically adjusting the differential privacy parameters of trajectory data includes:

[0043] Compare the weighted sum of the sensitivity index R and the access user risk value with the preset threshold interval group, and determine the corresponding differential privacy parameter based on the threshold interval;

[0044] The differential privacy parameters include the noise injection size.

[0045] By adopting the above technical solution, the privacy protection of trajectory data in each data sharing access process can be dynamically adjusted according to the sensitivity classification and the risk level of each accessing user, balancing the privacy protection and sharing needs of trajectory data.

[0046] In summary, this application includes at least one of the following beneficial technical effects:

[0047] The present invention dynamically adjusts the privacy protection of trajectory data during each data sharing access process through sensitivity classification and the risk value of each accessing user. When the data sensitivity is low and the accessing user risk is low, the differential privacy parameter of the trajectory data is lowered to improve the efficiency of trajectory data sharing. When the data sensitivity is high and the accessing user risk is high, the differential privacy parameter of the trajectory data is increased to protect the privacy of the trajectory data, thus balancing the privacy protection and sharing requirements of trajectory data. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a flowchart of the method for protecting and sharing trajectory privacy data in this application. DETAILED DESCRIPTION

[0049] Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings.

[0050] Throughout this specification, reference to the terms "certain embodiments," "one embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples" means that a specific feature, structure, material, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0051] The present application embodiment discloses a method for protecting and sharing trajectory privacy data. Figure 1, including: sensitivity grading according to trajectory data and its historical access information; identifying the behavior of each accessing user through event triggers, and determining the risk value of each accessing user based on the identification results; dynamically adjusting the differential privacy parameters of the trajectory data based on the risk value of each accessing user and the sensitivity of their access trajectory data. This embodiment judges the sensitivity of the trajectory data, grades the trajectory data based on the judgment result, and obtains the risk value by judging the risk status of the accessing user. The privacy protection of the trajectory data of each data sharing access process is dynamically adjusted through the sensitivity grading and the risk value of each accessing user. When the data sensitivity is low and the accessing user risk is low, the differential privacy parameters of the trajectory data are reduced. number, thereby improving the efficiency of trajectory data sharing. When the data sensitivity is high and the risk of access users is high, the privacy of the trajectory data can be protected by increasing the differential privacy parameters of the trajectory data, thereby balancing the privacy protection and sharing needs of the trajectory data; the process of sensitivity grading of trajectory data includes: conditional identification and spatial dimension identification of trajectory data based on AI to obtain the depth of associated information and spatial accuracy. Among them, conditional identification mainly determines whether there is highly sensitive privacy data in the trajectory data, such as the identity information of the person, and spatial dimension identification mainly determines the accuracy of the spatial position appearing in the trajectory data. For example, the precise coordinate position and specific home address are all highly sensitive information, and the above identification processes can be achieved through The existing AI model is used for identification and judgment. By inputting the required commands into the AI ​​model and setting the corresponding rules, it will judge the time data and spatial data involved in the trajectory data one by one. For example, when the position data is accurate to the coordinates and the time data is accurate to the second, it means that the position data and time data are highly sensitive information. The implementation of this process relies on the more basic functions of the AI ​​model as an auxiliary use, which will not be described in detail here; then the trajectory data is analyzed in the time dimension to obtain the time granularity value. The time granularity value reflects the temporal continuity, timestamp accuracy and high frequency of the trajectory data. Therefore, by weighted summing the depth of associated information, spatial accuracy and time granularity value, the depth of associated information, spatial accuracy and time granularity value correspond to The weights are set according to the type of result trajectory data and empirical data. Therefore, different factors of trajectory data sensitivity are comprehensively evaluated by weighted summation to obtain the sensitivity benchmark index of the trajectory data. At the same time, this embodiment also dynamically evaluates the risk status of the trajectory data based on the historical access information of the trajectory data, and obtains the dynamic risk index of the trajectory data based on the historical access information of the trajectory data. Therefore, the dynamic risk index reflects the risk status of the trajectory data over a period of time. The sensitivity classification of the trajectory data is determined based on the sensitivity benchmark index and the dynamic risk index of the trajectory data. The sensitivity classification can be adjusted in real time according to the content of the trajectory data and the real-time access data of the trajectory data, thereby providing a more accurate reference basis for the dynamic adjustment of its differential privacy parameters.

[0052] In one embodiment, a process for obtaining the depth of the associated information is provided, including: Calculate the depth of the associated information D, where n is the number of items identified by the condition, i∈[1,n], is the judgment value of the i-th condition identification. When the i-th condition identification result is met, ;otherwise, ; The weight value identified for the i-th condition, weight value According to the type of setting of the corresponding condition, when the privacy level involved in the corresponding condition is higher, such as the identity ID of the accessing user, the corresponding weight is higher. Therefore, by associating the information depth D, the privacy level involved in the trajectory data can be judged; in addition, the process of obtaining spatial accuracy includes: determining the accuracy level of the location coordinates in the trajectory information according to the results of AI spatial dimension recognition, where the accuracy level includes precise coordinate level, street level and city level, where the precise coordinate level can determine the specific location, the street level can determine the corresponding area, and the city level can determine the specific city. Therefore, the privacy level corresponding to each accuracy level is different. The corresponding spatial accuracy is determined according to the accuracy level, and the sensitivity of the trajectory data is judged through spatial accuracy; in addition, the process of obtaining the time granularity value includes: through Calculate the time granularity value G, where T is the timestamp precision level, including seconds, minutes, hours, and days. is the first control function, which sets different influence coefficients according to the precision level of the timestamp. The size of the influence coefficient is set according to the test data. Therefore, according to the timestamp precision level T, the influence of the timestamp precision level on the time granularity value G is determined. s is the standard deviation of the interval lengths between consecutive timestamps. When the continuity of the trajectory data is higher, its sensitivity is relatively higher. On the basis of the determination of the average interval time of the timestamps, when the consistency of the interval lengths between consecutive timestamps is higher, it means that the data continuity is higher. Among them, is the second control function, which sets different influence coefficients according to the distribution range of the standard deviation s in the test data and its control results, and then determines the time granularity value G through the standard deviation s of the interval length of consecutive adjacent timestamps; is the average time interval of timestamps. When the average time interval of timestamps is smaller, the corresponding sensitivity is higher. The third control function The corresponding influence coefficient is set for the range of the average interval time of different timestamps. The influence coefficient is set according to the test fitting. Therefore, through the acquisition process of the time granularity value G, the influence of timestamp accuracy, data continuity and the average interval time of timestamps on the sensitivity of trajectory data can be judged.

[0053] In one embodiment, a dynamic risk index acquisition process for trajectory data is provided, which includes: establishing a time-varying curve q(t) of the cumulative number of visits based on historical access information of trajectory data in a preset period before the current time point, wherein the preset period is selected based on experience, and obtaining the maximum slope K of q(t) in the preset period; and obtaining the maximum slope K of q(t) in the preset period through the formula Calculate the dynamic risk index ; Where m is the total number of visits within the preset period, is the slope reference value, which is used to normalize the maximum slope value K. Its value is set according to empirical data. is the first adjustment coefficient, which is used to adjust the weights of the factors of the surge rate of access data and the data access volume, where the formula for The simplified result is is the standard deviation of the interval lengths between consecutive access time points, is the standard deviation reference value, which is used to adjust The specific value is set according to the empirical data. Therefore, when the number of user visits is high, the number of user visits surges, and the visit time is regular during the preset period, the greater the risk of the trajectory data. Therefore, the dynamic risk index is obtained by obtaining , which can judge the dynamic risk of trajectory data and adjust the level of privacy protection accordingly.

[0054] In one embodiment, the process of determining the sensitivity classification of the trajectory data further includes: The sensitivity index R is calculated and the corresponding sensitivity level is determined according to the interval of the sensitivity index R; wherein, is the sensitivity benchmark index, for The normalized value of for Therefore, by combining the sensitivity benchmark index with the dynamic risk index, the sensitivity index R is obtained to evaluate the sensitivity status of the trajectory data and classify it.

[0055] In one embodiment, a process for determining the risk value of each accessing user is provided, including: determining a baseline risk level based on the account information of the accessing user: when the baseline risk level of the accessing user is a low risk level, the baseline risk value of the accessing user corresponding to the low risk level is used as the risk value of the accessing user; otherwise, the risk value of the accessing user is determined based on the baseline risk value of the accessing user corresponding to the baseline risk level of the accessing user and its access information, wherein, based on the account ID of the accessing user, it is judged whether the accessing user is an authorized management accessing user or a whitelist accessing user. If so, it means that the accessing user is of a low risk level. The risk level of the accessing user can also be judged based on the registration time of the accessing user, the access IP used, whether a proxy network is used, etc. The specific judgment method is not repeated here. Through the whitelist screening process and the preliminary risk judgment process, the efficiency of access data risk judgment is improved.

[0056] In addition, when the accessing user is not at a low risk level, the process of determining the risk value of the accessing user includes: judging whether the accessing user behavior triggers an event preset in the event trigger based on an event trigger, and when the event is triggered, determining the corresponding first increment of the risk value according to the triggered event; wherein, the event trigger sets a corresponding condition based on historical experience data, and when the accessing user triggers the condition, it indicates that the accessing user has a high risk, and therefore the corresponding first increment of the risk value is determined according to the triggered event, and then the risk level is determined. At the same time, this embodiment also determines the second increment of the risk value based on the access information of the accessing user. The process of calculating the second increment of the risk value includes obtaining a curve W(t) of the cumulative number of visits over time in a preset period before the current time point of the accessing user, and determining the maximum value of its slope in the preset period based on the curve W(t) of the cumulative number of visits over time , through the formula Calculate the second increment of risk value ; Where W is the cumulative number of visits at the current time point, is the visit count threshold, is the second adjustment coefficient, is the reference value of the visit volume slope. Among the above parameters, the visit number threshold and visit volume slope benchmark reference value The second adjustment coefficient is selected and set based on the average level of empirical data, which is used to compare the numerical deviation of the current number of visits and the slope of the visits. It is used to adjust the weight between the number of visits and the speed of visit surge. It is obtained by fitting the risk access data in the empirical data. Therefore, the baseline risk value of the visiting user, the first incremental risk value and the second incremental risk value are normalized and accumulated respectively to obtain the risk value of the visiting user. The security risk of the visiting user is determined by the risk value of the visiting user. According to the risk of the visiting user and the sensitivity classification of the trajectory data, the differential privacy parameters of the trajectory data can be determined to balance the privacy protection and sharing requirements of the trajectory data.

[0057] In one embodiment, the process of dynamically adjusting the differential privacy parameters of trajectory data includes: comparing the weighted sum of the sensitivity index R and the access user risk value with a preset threshold interval group, and determining the corresponding differential privacy parameter according to the threshold interval, wherein the weights corresponding to the sensitivity index R and the access user risk value are set according to their data ranges, so that the weighted sensitivity index R and the access user risk value have the same dimension, and the preset threshold interval group is manually divided and set based on the results calculated after fitting the empirical data. Different preset threshold intervals are set with corresponding differential privacy parameters, and the differential privacy parameters are the size of noise injection. Therefore, when the weighted sum is higher, it means that the privacy security risk of the trajectory data is higher, and the corresponding noise injection is in a higher range. Therefore, through the above process, the privacy protection of the trajectory data in each data sharing access process can be dynamically adjusted according to the sensitivity grade and the risk value of each access user, thereby balancing the privacy protection and sharing needs of the trajectory data.

[0058] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for protecting and sharing trajectory privacy data, characterized in that: include: Sensitivity classification based on trajectory data and its historical access information; Identify the behavior of each accessing user through event triggers and determine the risk value of each accessing user based on the identification results; Dynamically adjust the differential privacy parameters of the trajectory data based on the risk value of each accessing user and the sensitivity of their access trajectory data; The process of sensitivity grading trajectory data includes: AI-based conditional recognition and spatial dimension recognition of trajectory data to obtain the depth of associated information and spatial accuracy; Perform time dimension analysis on trajectory data to obtain time granularity values; The sensitivity benchmark index of trajectory data is obtained by weighted summing the correlation information depth, spatial accuracy and time granularity values; Obtaining a dynamic risk index of the trajectory data based on historical access information of the trajectory data; The sensitivity level of trajectory data is determined based on the sensitivity benchmark index and dynamic risk index of trajectory data.

2. The method for protecting and sharing trajectory privacy data according to claim 1, characterized in that: The process of obtaining the depth of the associated information includes: pass Calculate the depth of the associated information D, where n is the number of items identified by the condition, i∈[1,n], is the judgment value of the i-th condition identification. When the i-th condition identification result is met, ;otherwise, ; The weight value identified for the i-th condition; The spatial accuracy acquisition process includes: determining the accuracy level of the position coordinates in the trajectory information according to the results of AI spatial dimension recognition, and determining the corresponding spatial accuracy according to the accuracy level; The process of obtaining the time granularity value includes: pass Calculate the time granularity value G, where T is the timestamp precision level. is the average time interval between timestamps, s is the standard deviation of the time interval between consecutive timestamps, is the first control function, is the second control function, is the third comparison function.

3. The method for protecting and sharing trajectory privacy data according to claim 2, characterized in that: The process of obtaining the dynamic risk index of trajectory data includes: Based on the historical access information of the trajectory data in the preset period before the current time point, a curve q(t) of the cumulative number of visits over time is established, and the maximum slope K of q(t) in the preset period is obtained; By formula Calculate the dynamic risk index ; Where m is the total number of visits within the preset period, is the slope reference value, is the first adjustment coefficient, is the standard deviation of the interval lengths between consecutive access time points, is the standard deviation reference value.

4. The method for protecting and sharing trajectory privacy data according to claim 3, characterized in that: The process of determining the sensitivity level of trajectory data also includes: pass The sensitivity index R is calculated and the corresponding sensitivity grade is determined according to the interval of the sensitivity index R; in, is the sensitivity benchmark index, for The normalized value of for The normalized value of .

5. The method for protecting and sharing trajectory privacy data according to claim 4, characterized in that: The process of determining the risk value of each access user includes: Determine the baseline risk level of each access user based on account information: When the baseline risk level of the access user is low, the baseline risk value of the access user corresponding to the low risk level is used as the access user risk value; Otherwise, the access user risk value is determined according to the access user's baseline risk value corresponding to the access user's baseline risk level and the access information.

6. The method for protecting and sharing trajectory privacy data according to claim 5, characterized in that: The process of determining the risk value of an access user when the access user's risk level is not low includes: Based on the event trigger, it is determined whether the access user behavior triggers an event preset in the event trigger. When the event is triggered, a first increment of the corresponding risk value is determined according to the triggered event; Determine the second increment of the risk value based on the access information of the access user. The process of calculating the second increment of the risk value includes obtaining a curve W(t) showing the cumulative number of accesses over time in a preset period before the current time point of the access user, and determining the maximum slope of the curve W(t) within the preset period based on the cumulative number of accesses over time. , through the formula Calculate the second increment of risk value ; Where W is the cumulative number of visits at the current time point, is the visit count threshold, is the second adjustment coefficient, It is the reference value of visit volume slope; The access user's baseline risk value, the first incremental risk value, and the second incremental risk value are normalized and accumulated to obtain the access user's risk value.

7. The method for protecting and sharing trajectory privacy data according to claim 6, characterized in that: The process of dynamically adjusting the differential privacy parameters of trajectory data includes: Compare the weighted sum of the sensitivity index R and the access user risk value with the preset threshold interval group, and determine the corresponding differential privacy parameter based on the threshold interval; The differential privacy parameters include the noise injection size.

Images