A trajectory visualization method based on embedding timeline with hierarchical location information

Abstract

The application relates to a trajectory visualization method based on a timeline with hierarchical place information, which extracts hierarchical structure descriptive trajectory data from a text description to be analyzed, uses two position dimensions of horizontal and vertical directions for coding according to two information elements of time and space, maps each trajectory as a curve changing in space along time, adopts a layout optimization algorithm for curve layout optimization, allows trajectories under the same parent place to share the vertical position interval without interfering with each other in time, and further obtains visualized trajectories embedded with timelines. The method disclosed in the application shows the spatial proximity of different individuals at different times to users in a manner similar to a story timeline, adapts to the condition that the spatial information in data has different description granularities by using an abstract representation of a hierarchical structure, can objectively and effectively reveal trajectory characteristics, and lays a reliable foundation for the development of spatial information visualization.

Description

Technical Field

[0001] This invention belongs to the field of visualization, specifically relating to a trajectory visualization method based on embedding timelines with hierarchical location information. Background Technology

[0002] Trajectory data is an important data analysis object in many fields, and analyzing the co-occurrence of multiple individual trajectories is an important research direction. However, when trajectories have high uncertainty and co-occurrence data cannot be obtained directly, it is necessary to rely on external automated methods or the participation of analysts to make fuzzy inferences. For example, in epidemiological surveys of individual patients' activity trajectories, the survey report can only provide hierarchical regional descriptions of the locations the patient passed through, the shape of the data points in the trajectory, etc. In the following description of this invention, for the purpose of geographic information description, trajectory data with such characteristics are called hierarchical descriptive trajectories.

[0003] Existing trajectory visualization methods are not well suited for analysis tasks on hierarchical descriptive trajectories. Traditional trajectory visualization often relies on maps, but maps are not suitable for displaying location descriptions with different degrees of ambiguity in hierarchical descriptive trajectories. In addition, there are many trajectory visualizations that use abstract location representations, such as flow maps and spatiotemporal networks, but these methods cannot fully adapt to the people and data that are analyzing the co-occurrence of hierarchical descriptive trajectories, so they cannot efficiently and objectively extract trajectory features from the co-occurrence of hierarchical descriptive trajectories. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a trajectory visualization method based on a timeline with hierarchical location information embedded in it. This method analyzes the co-occurrence of hierarchical descriptive trajectories and, through visual encoding and layout calculation, presents the spatial proximity of different individuals at different times in a story-like timeline manner. By using an abstract representation of hierarchical structure in the display of spatial information, it adapts to situations where spatial information in the data has different descriptive granularities, objectively and effectively revealing trajectory characteristics and laying a reliable foundation for the development of spatial information visualization.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A trajectory visualization method based on timelines with hierarchical location information embedded, the method comprising the following steps:

[0007] S1. Extract hierarchical descriptive trajectory data from the text description to be analyzed. Encode the two information elements of time and space using the horizontal and vertical positional dimensions respectively. Each trajectory will be mapped as a curve that changes in space along time.

[0008] S2. Set an energy function for the optimization objective of the curve, and use a layout optimization algorithm to optimize the curve layout;

[0009] S3. Allowing locations with independent time signatures to share longitudinal position intervals under the same parent location to improve the encoding rules, thereby obtaining a visualized trajectory embedded in the timeline.

[0010] Furthermore, the encoding in step S1 using both horizontal and vertical positional dimensions includes the following sub-steps:

[0011] S11. Use the horizontal position to encode time information, and the horizontal position is linearly mapped from left to right to the time range involved in the hierarchical descriptive trajectory data;

[0012] S12. Location information is encoded using vertical location codes. Locations in the hierarchical descriptive trajectory data are organized into a location hierarchy tree structure according to administrative regions. Each location is mapped to an interval or a node in the vertical location.

[0013] Furthermore, the hierarchical relationship between locations in step S1 is reflected through the relationship between corresponding intervals.

[0014] Furthermore, if location A includes location B, then the interval in the vertical direction corresponding to location A also includes the interval corresponding to location B. If there is no inclusion relationship between the two locations, then their corresponding intervals do not overlap.

[0015] Furthermore, the optimization objective in step S2 includes at least one of minimizing the trajectory curve length, minimizing the number of trajectory curve intersections, and minimizing the total visualization height.

[0016] Furthermore, the final energy function used in the simulated annealing in step S2 is a weighted sum of the energy functions corresponding to the optimization objective:

[0017] Minimize E(x) = β1E length +β2E crossing +β3E white-space ,

[0018] Where E length To minimize the energy function corresponding to the length of the trajectory curve; E crossing To minimize the energy function corresponding to the number of intersections of trajectory curves; E white-space Minimize the energy function corresponding to the total visual height.

[0019] Furthermore, the energy function corresponding to minimizing the trajectory curve length is calculated using the following function:

[0020]

[0021] Furthermore, the energy function corresponding to minimizing the number of intersections of trajectory curves is calculated using the following function:

[0022] E crossing =#(trajectory crossing).

[0023] Furthermore, the energy function that minimizes the total visual height is calculated using the following function:

[0024] E white-space =Y max -Y min .

[0025] The beneficial technical effects of this invention are as follows: the trajectory visualization method based on timeline embedding with hierarchical location information disclosed in this invention can accurately and effectively present the information in hierarchical descriptive trajectory data and help users understand the co-occurrence of these trajectories at different granular locations. Through the combination of visual encoding methods, layout optimization algorithms and encoding improvements, it can produce visualizations with high space utilization and good readability, and can objectively and effectively reveal trajectory characteristics, laying a reliable foundation for the development of spatial information visualization. Attached Figure Description

[0026] Figure 1 This is a schematic diagram illustrating the process of encoding location information in the vertical direction using a trajectory visualization method based on a timeline with hierarchical location information, as described in Embodiment 1 of the present invention.

[0027] Figure 2 This is a schematic diagram of a trajectory visualization method based on embedding timeline with hierarchical location information, as described in Embodiment 1 of the present invention, which generates a curve that varies spatially along time.

[0028] Figure 3 This is a schematic diagram illustrating the curve layout optimization process using a layout optimization algorithm in a trajectory visualization method based on timelines with hierarchical location information, as described in Embodiment 1 of the present invention.

[0029] Figure 4 This is a schematic diagram illustrating the process of improving the encoding rules in a trajectory visualization method based on embedding timelines with hierarchical location information, as described in Embodiment 1 of the present invention.

[0030] Figure 5 This is a visualized trajectory map generated using a trajectory visualization method based on a timeline with hierarchical location information, as described in Embodiment 1 of the present invention. Detailed Implementation

[0031] The present invention will now be further described with reference to the accompanying drawings and specific embodiments.

[0032] Example 1

[0033] This invention provides a trajectory visualization method based on timelines embedded with hierarchical location information. The method includes the following steps:

[0034] S1. Extract hierarchical descriptive trajectory data from the text description to be analyzed. Encode the two information elements of time and space using the horizontal and vertical positional dimensions respectively. Each trajectory will be mapped as a curve that changes spatially along time.

[0035] The specific encoding rules are as follows:

[0036] S11. Encode time information using horizontal positions (x-axis), with the horizontal positions linearly mapped from left to right to the time range covered by the data.

[0037] S12. Location information is encoded using the vertical position (y-axis). The locations in the data are organized into a location hierarchy structure tree according to administrative regions. Each location is mapped to an interval or a node on the vertical position (y-axis).

[0038] like Figure 1 As shown, hierarchical descriptive trajectory data is extracted from the text description to be analyzed. The geographical locations in the data have different levels of uncertainty, and the geographical location of each record is a node on the location hierarchy tree.

[0039] like Figure 2 As shown, time information is linearly encoded on the horizontal axis (x-axis), while the location hierarchy structure tree representing location information is encoded on the vertical axis (y-axis). Each location is mapped to an interval, and the parent-child relationships in the tree are reflected in the inclusion relationships of the intervals. Under this encoding rule, the trajectory is mapped as a curve.

[0040] S2. Set an energy function for the optimization objective of the curve, and use a layout optimization algorithm to optimize the curve layout.

[0041] The hierarchical relationship between locations is reflected in the relationship between corresponding intervals. If location A contains location B, then the interval on the vertical position (y-axis) corresponding to location A also contains the interval corresponding to location B. If there is no inclusion relationship between two locations, then their corresponding intervals do not overlap.

[0042] Under this encoding rule, there is a degree of freedom in the allocation of locations along the vertical axis (y-axis) and the plotting of the trajectory curve along the vertical axis (y-axis). Intervals of sibling locations under the same parent location can be interchanged, and the size of the intervals can be adjusted. Therefore, the specific plotting position of the trajectory curve can also be adjusted within the corresponding interval. Under this degree of freedom, different plotting methods, while representing the same semantics, will give different specific presentation forms.

[0043] Therefore, in this embodiment of the invention, a layout optimization algorithm is employed to improve space utilization and reduce visual interference. The layout optimization algorithm proposes several optimization objectives, sets energy functions for these objectives, and uses a simulated annealing algorithm for optimization. Specifically, the optimization objectives include:

[0044] (1) Minimize the length of the trajectory curve: The curve of a trajectory should not have excessive curvature and should be kept as straight as possible. In this embodiment of the invention, the curve length is used to measure the degree of curvature, and the length of each trajectory curve is minimized as much as possible. The energy function corresponding to minimizing the length of the trajectory curve is:

[0045]

[0046] (2) Minimize the number of intersections between trajectory curves: Intersections between trajectory curves can interfere with the user's ability to read and understand information from different trajectory curves. Therefore, optimization aims to minimize the number of intersections between trajectory curves. The energy function corresponding to minimizing the number of intersections is:

[0047] E crossing =#(trajectory crossing)

[0048] (3) Minimize the total visualization height: The goal is to maximize the space utilization of the final visualization and present more information compactly. Since the horizontal position (x-axis) is mapped to a fixed time, the optimization objective only needs to minimize the total visualization height. The energy function corresponding to minimizing the total visualization height is:

[0049] E white-space =Y max -Y min

[0050] The final energy function used in simulated annealing is a weighted sum of the above three factors:

[0051] minimize E(x)=β1E length +β2E crossing +β3E white-space

[0052] like Figure 3As shown, the layout optimization algorithm uses simulated annealing, and the three optimization objectives are represented by three energy functions. The optimization objectives include minimizing the trajectory curve length, minimizing the trajectory curve intersection, and minimizing the total visual height.

[0053] S3. Allow trajectories under the same parent location to share the longitudinal position (y-axis) range where the time does not interfere with each other, in order to improve the encoding rules and obtain a visualized trajectory embedded in the timeline.

[0054] In real-world data, many locations are accessed only a few times by a limited number of trajectories, resulting in low spatial utilization. Allowing these infrequently used locations to share the y-axis interval can significantly improve the spatial utilization of visualizations with minimal impact on user comprehension. Therefore, a location-packaging optimization strategy is proposed, which allows locations whose trajectories under the same parent location do not interfere with each other in terms of time to share the vertical position (y-axis) interval.

[0055] like Figure 4 As shown, under the original encoding rules, many locations only have trajectories appearing in a limited number of time periods, resulting in significant space wastage. Therefore, the original encoding rules are improved to allow locations with non-overlapping trajectories under the same parent location to share a vertical position (y-axis) interval.

[0056] like Figure 5 As shown, a trajectory visualization method based on timeline embedding with hierarchical location information, disclosed in this invention, was used to generate a visualization trajectory map containing 8 cases. Points A and B in the figure show the possible contact of two cases in a village and the possible contact of three cases in a hotel, respectively.

[0057] As can be seen from the above embodiments, the trajectory visualization method based on timeline with hierarchical location information disclosed in this invention can accurately and effectively present the information in hierarchical descriptive trajectory data and help users understand the co-occurrence of these trajectories at different granular locations. Through the combination of visual encoding methods, layout optimization algorithms and encoding improvements, it can produce visualizations with high space utilization and good readability, and can objectively and effectively reveal trajectory characteristics, laying a reliable foundation for the development of spatial information visualization.

[0058] The method described in this invention is not limited to the embodiments described in the specific implementation. Other implementation methods derived by those skilled in the art based on the technical solution of this invention also fall within the scope of technical innovation of this invention.

Claims

1. A trajectory visualization method based on timelines embedded with hierarchical location information, the method comprising the following steps: S1. Extract hierarchical descriptive trajectory data from the text description to be analyzed. Encode the two information elements of time and space using the horizontal and vertical positional dimensions respectively. Each trajectory will be mapped as a curve that changes in space along time. S2. Set an energy function for the optimization objective of the curve, and use a layout optimization algorithm to optimize the curve layout; S3. Allowing locations with independent time signatures to share longitudinal position intervals under the same parent location to improve the encoding rules, thereby obtaining a visualized trajectory embedded in the timeline; Step S1, which uses both horizontal and vertical positional dimensions for encoding, includes the following sub-steps: S11. Use the horizontal position to encode time information, and the horizontal position is linearly mapped from left to right to the time range involved in the hierarchical descriptive trajectory data; S12. Location information is encoded using vertical location codes. Locations in the hierarchical descriptive trajectory data are organized into a location hierarchy structure tree according to administrative regions. Each location is mapped to an interval or a node in the vertical location. The hierarchical relationship between locations in step S1 is reflected through the relationship between corresponding intervals.

2. The trajectory visualization method based on timeline embedding with hierarchical location information as described in claim 1, characterized in that: If location A contains location B, then the interval in the vertical direction corresponding to location A also contains the interval corresponding to location B. If there is no inclusion relationship between the two locations, then their corresponding intervals do not overlap.

3. The trajectory visualization method based on timeline embedding with hierarchical location information as described in claim 1, characterized in that: The optimization objective in step S2 includes at least one of minimizing the trajectory curve length, minimizing the number of trajectory curve intersections, and minimizing the total visualization height.

4. The trajectory visualization method based on timeline embedding with hierarchical location information as described in claim 3, characterized in that, The final energy function used in the simulated annealing in step S2 is a weighted sum of the energy functions corresponding to the optimization objective: ， in To minimize the energy function corresponding to the length of the trajectory curve; To minimize the energy function corresponding to the number of intersections of trajectory curves; Minimize the energy function corresponding to the total visual height.

5. The trajectory visualization method based on timeline embedding with hierarchical location information as described in claim 4, characterized in that, The energy function corresponding to minimizing the trajectory curve length is calculated using the following function: 。 6. The trajectory visualization method based on timeline embedding with hierarchical location information as described in claim 4, characterized in that, The energy function corresponding to minimizing the number of intersections of trajectory curves is calculated using the following function: 。 7. The trajectory visualization method based on timeline embedding with hierarchical location information as described in claim 4, characterized in that, The energy function that minimizes the total visual height is calculated using the following function: 。

Images