CN111723166B - Track data processing method and system - Google Patents

Abstract

The embodiment of the invention provides a track data processing method and a track data processing system, wherein the provided method comprises the following steps: acquiring track processing parameters according to the privacy protection degree; performing offset processing on a starting point and an ending point in original track data according to the track processing parameters to obtain a starting point offset point and an ending point offset point, and performing extension processing on the starting point offset point and the ending point offset point to obtain a binding track segment; in the original track data set, all paths between the internal starting point and the internal ending point are selected, and the path in which the PoI score meets the preset condition is used as the innermost track segment; and connecting the innermost track section with the binding track section to form an inner track section, and extending the inner track section to obtain a track after privacy protection. The method and the system provided by the invention promote the safety of the track data, simultaneously reserve and protect the information of the real starting point and the real stopping point, and keep the usability of the released track data set in urban traffic planning.

Description

A trajectory data processing method and system

technical field

The embodiments of the present invention relate to the field of computer technology, and in particular to a trajectory data processing method and system.

Background technique

Location-based services based on trajectory data have been widely used in people's daily life. Valuable trajectory data can be used to query navigation information or traffic report, which can effectively improve the quality of urban transfers and improve travel efficiency. However, the data directly submitted by the user may contain a lot of personal sensitive information, such as identity information. If the real trajectory information is released directly, it is vulnerable to malicious attacks. The attacker will analyze and mine the trajectory data user or its published location, semantics, etc., and infer sensitive data, such as home address and work location. , health status, and even social relationships.

In the existing track release privacy protection research work, many privacy protection methods have been proposed. Most of these methods protect trajectory data by adopting techniques such as pseudonym replacement and trajectory clustering independently or in combination. Among the existing trajectory publishing privacy protection methods, the trajectory clustering method is widely used. The trajectory clustering method is to classify the trajectory according to certain rules, and only publish some fragments of the real trajectory according to the rules.

However, in the prior art, compared with the real unprocessed trajectory data, the trajectory data processed by the trajectory clustering method is likely to be composed of multi-segment trajectories (may not be connected), and the obtained Trajectory data sets cannot play a guiding role in urban traffic planning, and at the same time, they cannot better protect users' private data.

Contents of the invention

Embodiments of the present invention provide a trajectory data processing method and system to solve the problem in the prior art that the privacy protection of user trajectory data is not high, and the processed trajectory data cannot meet the usability of urban traffic planning.

In a first aspect, an embodiment of the present invention provides a trajectory data processing method, including:

Obtain trajectory processing parameters according to the degree of privacy protection;

According to the trajectory processing parameters, offset processing is performed on the starting point and the end point in the original trajectory data to obtain a starting point offset point and an end point offset point, and an extension process is performed on the starting point offset point and an end point offset point to obtain a binding Fixed track segment;

In the original trajectory data set, for all the paths between the inner starting point and the inner end point, select the path whose PoI score meets the preset conditions and the semantic type of the passing point meets the preset conditions, as the innermost trajectory segment;

connecting the innermost trajectory segment with the bound trajectory segment to form an internal trajectory segment, and extending the internal trajectory segment to obtain a privacy-protected trajectory;

Wherein, the inner starting point is the end point of the bound track segment formed after the starting point offset processing in the original track data;

Wherein, the inner end point is the first end point of the bound track segment formed after the end point offset processing in the original track data.

In a second aspect, an embodiment of the present invention provides a trajectory data processing system, including:

A parameter acquisition module, configured to acquire trajectory processing parameters according to the degree of privacy protection;

Binding trajectory segment acquisition module, used to offset the starting point and end point in the original trajectory data according to the trajectory processing parameters, obtain the starting point offset point and the end point offset point, and offset the starting point and the end point The offset point is extended to obtain the bound trajectory segment;

The innermost track segment acquisition module is used to select the path between all inner start points and inner end points in the original track data set, where the PoI score meets the preset conditions and the semantic type of the passing point meets the preset conditions, as the innermost track part;

A trajectory generation module, configured to connect the innermost trajectory segment with the bound trajectory segment to form an internal trajectory segment, and extend the internal trajectory segment to obtain a privacy-protected trajectory;

Wherein, the inner starting point is the end point of the bound track segment formed after the starting point offset processing in the original track data;

Wherein, the inner end point is the first end point of the bound track segment formed after the end point offset processing in the original track data.

In a third aspect, an embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and operable on the processor, and the processor implements the above-mentioned first aspect when executing the program. Steps of the provided trajectory data processing method.

In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the trajectory data processing method provided in the above-mentioned first aspect are implemented. .

The method provided by the embodiment of the present invention protects the user's trajectory information by randomly selecting trajectory segments with similar popularity between the real start and end points, which can resist long-term observation attacks. The information of real starting and ending points keeps the availability of published trajectory datasets in urban traffic planning.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present invention. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is a schematic flow chart of a trajectory data processing method provided by an embodiment of the present invention;

Fig. 2 is a schematic structural diagram of a trajectory data processing system provided by an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

Referring to FIG. 1, FIG. 1 is a schematic flow diagram of a trajectory data processing method provided by an embodiment of the present invention, and the provided method includes:

S1, according to the degree of privacy protection, obtain trajectory processing parameters;

S2. Perform offset processing on the start point and end point in the original track data according to the track processing parameters to obtain a start point offset point and an end point offset point, and perform extension processing on the start point offset point and end point offset point, Get the bound track segment;

S3, in the original trajectory data set, for all the paths between the inner starting point and the inner end point, select the path whose PoI score meets the preset conditions and the semantic type of the passing point meets the preset conditions, as the innermost trajectory segment;

S4. Connect the innermost trajectory segment with the bound trajectory segment to form an internal trajectory segment, and extend the internal trajectory segment to obtain a privacy-protected trajectory;

Wherein, the inner starting point is the end point of the bound track segment formed after the starting point offset processing in the original track data;

Wherein, the inner end point is the first end point of the bound track segment formed after the end point offset processing in the original track data;

Wherein, the PoI score specifically adopts the form of quantitative scoring to evaluate the PoI situation within a certain distance range around all passing points in a route.

Specifically, in trajectory data processing, in order to protect user privacy, it is necessary to build certain background knowledge, so as to design a more effective and reasonable privacy protection scheme. In the embodiment of the present invention, the background knowledge is mainly composed of three parts: the first part is to utilize the real, unprocessed track data set, the transition probability of each point (i.e. crossing) obtained (entering this point or departing from this point two times) Direction, that is, in-degree and out-degree transition probability); the second part is the PoI (Point of Interest) situation around each point, and the PoI score of each point is used to evaluate its popularity; the third part is the reference semantic division The standard determines the semantics of each point, and each point has one and only one semantics.

In this embodiment, the step of processing trajectory data includes inputting parameters. The steps of true start-stop offset and bound trajectory segment generation and new trajectory generation. First of all, for the input of parameters, the owner of the trajectory data determines the range of the offset of the real start and end points, the length of the bound trajectory segment of the real start and stop points, and the true The starting and ending points are bound to parameters such as the length of the outward extension of the track segment.

After the trajectory processing parameters are obtained, the steps of real starting and ending point offset and bound trajectory segment generation are carried out. This step first selects the offset point in the original trajectory data whose starting point and ending point are within a certain range. Then, on the basis of the offset point, extend forward and backward to obtain the bound trajectory segment after the offset of the real start point and end point. During the extension process, the transition probability of each point is considered, the PoI score of the extension point and the real point is as close as possible, and the semantics of the real point or the selected extension point are not repeated as much as possible.

Then, in the step of generating new trajectories, firstly, in the real and unprocessed trajectories set, traverse the path between the inner start point and the inner end point before and after, and then randomly select the PoI score close to the PoI score of the bound trajectory segment, and cover more Paths with different semantics, become innermost trajectory segments. Then connect the innermost track segment with the two bound track segments to form the inner track segment. Then, the internal trajectory segment is extended forward and backward. This extension process is similar to the extension process of the second step, and finally forms a privacy-protected trajectory corresponding to the original trajectory.

Wherein, the forward or backward is compared with the direction of the real start-stop vector, and the direction opposite to the start-stop vector is called forward; otherwise, it is called backward. The PoI score represents the evaluation of the PoI (Point of Interest) conditions around a location or all passing points of a path in the form of quantitative scoring. The inner starting point refers to the end point of the bound trajectory segment formed after the real starting point is offset. The inner end point refers to the first end point of the bound trajectory segment formed after the real end point is offset.

Through this method, the user's trajectory information can be protected by randomly selecting trajectory segments with similar popularity between the real start and end points, which can resist long-term observation attacks. information to maintain the availability of published trajectory datasets for urban mobility planning.

On the basis of the above-mentioned embodiments, the step of obtaining trajectory processing parameters according to the degree of privacy protection specifically includes: determining the trajectory processing parameters according to the degree of privacy protection; wherein, the trajectory processing parameters include but are not limited to the original trajectory data The area of the range where the start and end offset points of , the length of the bound track segment and the outward extension length of the bound track segment.

Specifically, the user determines the appropriate parameters according to the subjective and objective conditions and the degree of privacy protection required by the environment. These parameters will affect the trajectory data privacy, required cost and trajectory data availability of the scheme.

The privacy of trajectory data requires two dimensions: one is to measure the short-term privacy protection effect after release; the other is to measure the long-term privacy protection effect after data release. The short-term privacy protection effect after the release refers to only considering the adjustment of the currently released trajectory data, the real start and end points of the release trajectory, and all background information; the long-term privacy protection effect after the release of the data is the short-term protection effect. On the basis of the same real starting and ending points in history, the trajectory adjustment of the same real starting and ending points should also be considered.

The required cost mainly refers to computing overhead and historical data storage overhead. Trajectory data availability refers to the fact that the trajectory collection users use the published trajectories, and information such as the user's travel habits and travel needs can still be mined.

The parameters include: the area size r of the range where the offset point can be selected at the real start-stop point, the length of the real start-stop point bound trajectory segment, that is, the total number of transfer nodes (jumps) forward or backward: k _i ∈ [2 , 4]; when a new trajectory is generated, the length of the trajectory segment bounded by the real starting and ending points extends forward or backward, that is, the number of forward or backward transfer nodes (hops): k _f ∈ [2, 8] and k _b ∈[2,8]; when selecting the middle trajectory segment, the acceptable PoI score deviation α (that is, the PoI score L _A of the trajectory segment after the starting point is offset, and the PoI score L _B of the trajectory segment after the end point is offset), and the innermost trajectory During segment selection, the optional trajectory segment PoI score interval is ( _LA -α, L _B +α) or ( _LB -α, L _A +α).

With this approach, various parameters are set to support varying trajectory data privacy, required cost, and trajectory data availability. Therefore, the trajectory publisher can change the parameters at any time according to the subjective and objective environment including the needs of the service object (trajectory data user), so as to achieve a dynamic balance between the privacy of trajectory data, the required cost and the availability of trajectory data.

On the basis of the above-mentioned embodiments, according to the estimated processing parameters, the starting point and the ending point in the original trajectory data are offset, and the starting point and the ending point are obtained, and the starting point and the ending point are obtained. The step of extending the end point offset point to obtain the bound track segment specifically includes: offsetting the start point and end point in the original track data according to the track processing parameters to obtain the start point offset point and the end point offset point; The start offset point and the end offset point are used as the starting point, and a preset number of nodes are extended forward and/or backward to generate a bound trajectory segment of the start offset point and a bound trajectory segment of the terminal offset point.

Specifically, firstly, according to the parameter setting, the real start and end points are offset within a set small range to obtain the offset point of the real start and end point, and the offset point is used as the starting point to extend forward/backward to multiple nodes , to meet the setting of the bound track segment length. Each node selected into the bound trajectory segment should be as semantically different as possible from the node selected into the trajectory segment, and keep the PoI score close. The bound track segment means that in the case of the same parameter setting, the bound track segment will not change, and the points on the bound track segment have corresponding bound track segments.

On the basis of the above-described embodiments, the step of selecting a path whose PoI score satisfies a preset condition as the innermost trajectory segment according to the path between all inner starting points and inner end points in the original trajectory data set specifically includes: traversing all In the original trajectory data of the original trajectory data, the trajectory segments passing through the inner starting point and the inner destination point are calculated, and the PoI scores of all trajectory segments are calculated; the path whose PoI score is the closest to the bound trajectory segment and has more semantic types of passing points is selected as the innermost track segment.

The passing point is the place where the trajectory passes by; the semantics indicates the industrial function to which a place belongs, conveys the location information of a place, and each place has one and only one semantics.

The step of connecting the innermost track segment with the bound track segment to form an inner track segment, and extending the inner track segment to obtain a privacy-protected track specifically includes: connecting the innermost track segment The track segment is connected to the bound track segment to form an internal track segment; according to preset conditions, the internal track segment is extended to form a privacy-protected track; wherein, the preset conditions include but are not limited to: extension length, A combination of one or more of transition probabilities and PoI scores.

Specifically, in the post-extension generation step of the new trajectory, after obtaining the parameters and the offset trajectory segment of the real starting and ending points, the adjusted trajectory data is output according to the global background knowledge, PoI score and semantic situation.

The first choice is to select the innermost trajectory segment, traverse all the real trajectory data, and obtain all the trajectory segments that pass through the inner starting point and the inner end point in turn to form a set of candidate innermost trajectory segments. For each trajectory segment in the candidate set, the PoI score and the number of coverage semantics are counted. Among the trajectory segments that meet the requirements of the PoI score parameters, a trajectory segment with a trajectory PoI score that is relatively close to the two bound trajectory segments and that has a relatively large number of passing point coverage semantic types is randomly selected as the innermost trajectory segment. Concatenates the innermost track segment with the two bound track segments to form the inner track segment. Generates a release-ready track based on internal track segments. Take the starting point (end point) of the internal trajectory segment as the starting point, and extend forward (backward) to meet the set parameters. Factors such as the PoI score situation, coverage semantics situation, and transition probability situation of the selected point are considered. Outputs a release-ready trajectory containing information about true origins and stops.

Through this method, by traversing all the trajectories of the real data set from the real starting point to the real end point, select the trajectories that cover as much semantics as possible and have similar PoI scores at each point of the path, so as to resist the reconstruction of the trajectories. Considering the background knowledge of the attacker, it can resist a variety of inference attacks. The background knowledge mainly consists of two aspects: road background knowledge, for example, the semantics and PoI score of each point; travel habit background knowledge, for example, each point The out-degree transition probability and in-degree transition probability of .

With reference to Fig. 2, Fig. 2 is a schematic structural diagram of a trajectory data processing system provided by an embodiment of the present invention, the provided system includes: a parameter acquisition module 21, a bound trajectory segment acquisition module 22, an innermost trajectory segment acquisition module 23 and a trajectory Build module 24.

Wherein, the parameter acquisition module 21 is used to acquire trajectory processing parameters according to the degree of privacy protection.

The binding trajectory segment acquisition module 22 is used to perform offset processing on the starting point and the ending point in the original trajectory data according to the trajectory processing parameters, obtain the starting point offset point and the ending point offset point, and perform the offset processing on the starting point and the ending point The offset point is extended to obtain the bound trajectory segment.

The innermost track segment acquisition module 23 is used to select the path between all inner start points and inner end points according to the original track data set, and select the path whose PoI score satisfies the preset condition and the semantic type of the passing point satisfies the preset condition, as the innermost track part.

The trajectory generating module 24 is used to connect the innermost trajectory segment and the binding trajectory segment to form an internal trajectory segment, and extend the internal trajectory segment to obtain a privacy-protected trajectory.

Wherein, the inner starting point is the end point of the bound track segment formed after the starting point offset processing in the original track data; wherein, the inner end point is after the end point offset processing in the original track data, The first end point of the formed bound trajectory segment; wherein, the PoI score is specifically in the form of quantitative scoring to evaluate all the way points in a path.

Specifically, for the input of parameters, the trajectory data owner determines the range of the real start-stop offset, the length of the real start-stop bound trajectory segment, and the time when the new trajectory is generated according to the subjective and objective conditions and the degree of privacy protection required by the environment. The real starting and ending points are bound to parameters such as the length of the trajectory segment extending forward or backward.

After the trajectory processing parameters are obtained, the steps of real starting and ending point offset and bound trajectory segment generation are carried out. This step first selects the offset point in the original trajectory data whose starting point and ending point are within a certain range. Then, on the basis of the offset point, extend forward and backward to obtain the bound trajectory segment after the offset of the real start point and end point. During the extension process, the transition probability of each point is considered, the PoI score of the extension point and the real point is as close as possible, and the semantics of the real point or the selected extension point are not repeated as much as possible.

Then, in the step of generating new trajectories, firstly, in the real and unprocessed trajectories set, traverse the path between the inner start point and the inner end point before and after, and then randomly select the PoI score close to the PoI score of the bound trajectory segment, and cover more Paths with different semantics, become innermost trajectory segments. Then connect the innermost track segment with the two bound track segments to form the inner track segment. Then, the internal trajectory segment is extended forward and backward. This extension process is similar to the extension process of the second step, and finally forms a privacy-protected trajectory corresponding to the original trajectory.

Wherein, the forward or backward is compared with the direction of the real start-stop vector, and the direction opposite to the start-stop vector is called forward; otherwise, it is called backward. The PoI score represents the evaluation of the PoI (Point of Interest) conditions around a location or all passing points passing by a path in the form of quantitative scoring. The inner starting point refers to the end point of the bound trajectory segment formed after the real starting point is offset. The inner end point refers to the first end point of the bound trajectory segment formed after the real end point is offset.

Through this system, the user's trajectory information can be protected by randomly selecting trajectory segments with similar popularity between the real starting and ending points, which can resist long-term observation attacks, and at the same time, by adding the trajectory before and after the real starting and ending points, the real starting and ending points can be preserved and protected. information to maintain the availability of published trajectory datasets for urban mobility planning.

Fig. 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. As shown in Fig. 3, the electronic device includes: a processor (processor) 301, a communication interface (Communications Interface) 302, a memory (memory) 303 and a bus 304, Wherein, the processor 301 , the communication interface 302 , and the memory 303 communicate with each other through the bus 304 . The processor 301 can call the logic instructions in the memory 303 to perform the following methods, for example, including: obtaining trajectory processing parameters according to the degree of privacy protection; and performing offset processing on the starting point and the ending point in the original trajectory data according to the trajectory processing parameters , to obtain the starting point offset point and the end point offset point, extend the starting point offset point and the end point offset point, and obtain the bound trajectory segment; according to the original trajectory data set, all paths between the inner starting point and the inner ending point , select the path whose PoI score satisfies the preset condition and the semantic type of the passing point satisfies the preset condition, as the innermost track segment; connect the innermost track segment with the bound track segment to form an inner track segment, and The internal trajectory segment is extended to obtain a privacy-protected trajectory.

The embodiment of the present invention discloses a computer program product. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by the computer, the computer can execute the above methods. The method provided by the embodiment, for example, includes: obtaining the trajectory processing parameters according to the degree of privacy protection; according to the trajectory processing parameters, performing offset processing on the starting point and the ending point in the original trajectory data to obtain the starting point offset point and the ending point offset point, extend the start offset point and the end offset point to obtain the bound trajectory segment; according to the original trajectory data set, all paths between the inner start point and the inner end point are selected, wherein the PoI score satisfies the preset condition and The path whose semantic type of the passing point satisfies the preset condition is used as the innermost track segment; the innermost track segment is connected with the bound track segment to form an inner track segment, and the inner track segment is extended to obtain privacy trajectory after protection.

This embodiment provides a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium stores computer instructions. The computer instructions cause the computer to execute the methods provided in the above-mentioned method embodiments, for example, including: according to the degree of privacy protection, Obtain trajectory processing parameters; according to the trajectory processing parameters, perform offset processing on the starting point and the end point in the original trajectory data, obtain the starting point offset point and the end point offset point, and perform the offset processing on the starting point offset point and the end point offset point Extend processing to obtain bound trajectory segments; according to the original trajectory data set, all the paths between the inner start point and the inner end point, select the path whose PoI score meets the preset conditions and the semantic type of the passing point meets the preset conditions, as the innermost trajectory segment; connecting the innermost track segment with the bound track segment to form an inner track segment, and extending the inner track segment to obtain a privacy-protected track.

The device embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed to multiple network elements. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. It can be understood and implemented by those skilled in the art without any creative efforts.

Through the above description of the implementations, those skilled in the art can clearly understand that each implementation can be implemented by means of software plus a necessary general hardware platform, and of course also by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic discs, optical discs, etc., including several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) execute the methods described in various embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still be Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims (8)

1. A trajectory data processing method, characterized in that, comprising: Obtain trajectory processing parameters according to the degree of privacy protection; According to the trajectory processing parameters, offset processing is performed on the starting point and the end point in the original trajectory data to obtain a starting point offset point and an end point offset point, and an extension process is performed on the starting point offset point and an end point offset point to obtain a binding Fixed track segment; In the original trajectory data set, for all the paths between the inner starting point and the inner end point, select the path whose PoI score meets the preset conditions and the semantic type of the passing point meets the preset conditions, as the innermost trajectory segment; connecting the innermost track segment with the bound track segment to form an inner track segment, and extending the inner track segment to obtain a privacy-protected complete track; Wherein, the inner starting point is the end point of the bound track segment formed after the starting point offset processing in the original track data; Wherein, the inner end point is the first end point of the bound track segment formed after the end point offset processing in the original track data. 2. The method according to claim 1, wherein the step of obtaining trajectory processing parameters according to the degree of privacy protection includes: Determine the trajectory processing parameters according to the degree of privacy protection; Wherein, the trajectory processing parameters include, but are not limited to, the area of the start and end offset ranges in the original trajectory data, the length of the bound trajectory segment, and the outward extension length of the bound trajectory segment. 3. The method according to claim 1 or 2, wherein, according to the trajectory processing parameters, the starting point and the ending point in the original trajectory data are offset to obtain a starting point offset point and an end point offset point , performing extension processing on the starting point offset point and the ending point offset point, and obtaining a bound track segment, which specifically includes: According to the trajectory processing parameters, the starting point and the ending point in the original trajectory data are offset to obtain the starting point offset point and the end point offset point; Taking the start offset point and the end offset point as the starting point respectively, extending a preset number of nodes in the forward and backward directions to generate a bound trajectory segment of the starting offset point and a bound trajectory segment of the terminal offset point . 4. The method according to claim 1, characterized in that, in the original trajectory data set, all paths between the inner starting point and the inner end point are selected wherein the PoI score satisfies the preset condition and the semantic category of the points passed through satisfies the preset condition. The steps for setting the conditional path as the innermost trajectory segment specifically include: Traverse all the original trajectory data through the trajectory segment of the inner starting point and the inner end point, and calculate the PoI score of all trajectory segments; The path whose PoI score is the closest to the bound trajectory segment and which passes through the most semantic types is selected as the innermost trajectory segment. 5. The method according to claim 4, wherein the innermost track segment is connected with the binding track segment to form an inner track segment, and the inner track segment is extended outward, Steps to obtain a privacy-preserving trajectory, including: connecting the innermost track segment with the binding track segment to form an inner track segment; According to the preset rules, the internal trajectory segment is extended to form a privacy-protected trajectory; Wherein, the preset rules include but are not limited to: a combination of one or more of extension length, transition probability and PoI score. 6. A trajectory data processing system, characterized in that, comprising: A parameter acquisition module, configured to acquire trajectory processing parameters according to the degree of privacy protection; Binding trajectory segment acquisition module, used to offset the starting point and end point in the original trajectory data according to the trajectory processing parameters, obtain the starting point offset point and the end point offset point, and offset the starting point and the end point The offset point is extended to obtain the bound trajectory segment; The innermost track segment acquisition module is used to select the path between all inner start points and inner end points in the original track data set, where the PoI score meets the preset conditions and the semantic type of the passing point meets the preset conditions, as the innermost track part; A trajectory generation module, configured to connect the innermost trajectory segment with the bound trajectory segment to form an internal trajectory segment, and extend the internal trajectory segment to obtain a privacy-protected trajectory; Wherein, the inner starting point is the end point of the bound track segment formed after the starting point offset processing in the original track data; Wherein, the inner end point is the first end point of the bound track segment formed after the end point offset processing in the original track data. 7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, wherein the processor implements any of claims 1 to 5 when executing the program. The steps of the trajectory data processing method described in the item. 8. A non-transitory computer-readable storage medium, on which a computer program is stored, characterized in that, when the computer program is executed by a processor, the steps of the trajectory data processing method according to any one of claims 1 to 5 are realized .

Images