CN109213941B - Indoor track frequent pattern mining method based on fuzzy grid sequence - Google Patents
Indoor track frequent pattern mining method based on fuzzy grid sequence Download PDFInfo
- Publication number
- CN109213941B CN109213941B CN201810838419.2A CN201810838419A CN109213941B CN 109213941 B CN109213941 B CN 109213941B CN 201810838419 A CN201810838419 A CN 201810838419A CN 109213941 B CN109213941 B CN 109213941B
- Authority
- CN
- China
- Prior art keywords
- grid
- track
- fuzzy
- support degree
- grids
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
Abstract
The invention discloses an indoor track frequent pattern mining method based on a fuzzy grid sequence, which comprises the following steps: dividing the map by adopting a regular hexagonal grid; dividing the grid into an accurate area and a fuzzy area according to the projection distance of the vertical line; traversing each track in the track database, and converting track data into a fuzzy grid sequence according to grids of sample points in the tracks; recording the index position of each grid in the track, calculating the support degree of the index position, and adding the grids with the support degree exceeding the minimum support degree into the candidate set; establishing a projection database for each grid in the candidate set, and calculating the support degree of the newly generated candidate track mode; these steps are repeated until all track frequent patterns are mined. The invention has the advantages that: the distances from the centers of the regular hexagonal grids to the centers of the adjacent grids are equal, so that the regular hexagonal grids are adopted to divide the map, and the equality among all grids is ensured; dividing the grid into an accurate area and a fuzzy area according to the projection distance of the vertical line, and effectively solving the grid boundary problem faced by the traditional method; under any support, the number of track frequent modes excavated by the method is considerable, and the excavation efficiency is higher. The method is suitable for excavating the frequent track mode of the moving object in the indoor environment, and has universality.
Description
Technical Field
The invention relates to an indoor track frequent pattern mining method based on a fuzzy grid sequence, which is a mining method aiming at a track of a moving object in an indoor environment and belongs to the crossing field of engineering application and information science.
Background
With the popularity of a variety of mobile location devices, location Based Services (LBS) have been widely available in everyday life. According to the data provided by Nokia, 87% to 90% of people spend in indoor space in life, so that people see huge business opportunities brought by indoor positioning technology, and related companies in all countries of the world are actively approaching the indoor location service industry.
Although some research results on the trajectories of moving objects have been carried out at present, most of them are focused on the outdoor environment, and there are few studies on the trajectories of moving objects in the indoor environment. The indoor space has the characteristics of small movable space and large traffic flow, and the constraint is between European space and road network space, so that the research scheme in the outdoor environment cannot be directly used for tracks in the indoor environment, and a specific mining method aiming at a track mode of a moving object in the indoor environment needs to be further researched and established. Therefore, the track frequent pattern mining of indoor moving objects is particularly important and urgent.
In addition, the conventional study of the track frequent pattern does not consider the problem of the grid boundary. In order to enable indoor location service to be developed faster and better, the invention designs an effective and feasible solution, and the indoor track frequent pattern mining method based on the fuzzy grid sequence has the advantages of considerable mining quantity, high mining efficiency and good application prospect, and can be applied to large markets, hospitals and the like.
Disclosure of Invention
The invention aims to: the invention aims to provide an indoor track frequent pattern mining method based on a fuzzy grid sequence. The method fully utilizes the track data of the indoor mobile object, solves the grid boundary problem which is not considered by the traditional method, carries out multi-parameter autonomous selection and mining of an indoor track frequent mode, can better promote the development of the industry of Location Based Services (LBS), reduces the risk of track privacy disclosure of a user and even protects the track privacy of the user from being infringed.
The technical scheme is as follows: in order to achieve the above purpose, the invention provides an indoor track frequent pattern mining method based on a fuzzy grid sequence, which mainly utilizes the advantage that the distances from the center of a regular hexagonal grid to the center of an adjacent grid are equal, solves the grid boundary problem which is not considered by the existing research method, and makes the research scheme more rigorous and feasible; by utilizing a multi-parameter autonomous selection method, the number of the trace frequent modes with considerable quantity is excavated, and the efficiency is higher. The specific technical scheme comprises the following steps:
step one: and preprocessing indoor track data.
(1) The indoor map is divided by utilizing the regular hexagonal grids, and the distances from the centers of the regular hexagonal grids to the centers of the adjacent grids are equal, so that the equality of all the grids is ensured;
(2) The projection distance of the vertical line is defined, and the projection distance of a line segment from a certain point to the center point of the grid on the vertical line with the minimum included angle with the line segment in the grid is defined.
The calculation formula is as follows:
p_dist(p，g)＝dist*cos(30°-θ) 0°≤θ≤30°
p_dist(p，g)＝dist*cos(θ-30°) 30°＜θ≤60°
p_dist(p，g)＝dist*cos(90°-θ) 60°＜θ≤90°
wherein p_dist is the projection distance of a vertical line, p is a certain point, g is a grid, dist is the distance between the point p and the central point of the grid g, θ is the included angle between the connecting line of the point p and the central point of the grid g and the X axis, and the value range is 0 DEG to 90 deg.
(3) According to the perpendicular projection distance in (2), an area in the grid where the perpendicular projection distance is smaller than the given threshold r is regarded as an accurate area of the grid, which is an area where each grid contains and does not overlap each other. Points within the exact area of a grid belong to and only belong to the grid;
(4) According to the perpendicular projection distance in (2), the perpendicular projection distance in the grid is in the range of r toIs considered as a blurred region of the mesh (R is a given perpendicular projection distance threshold, R is the side length of a regular hexagonal mesh), which is a region that each mesh contains and overlaps with each other. Points within a blurred region of a mesh may belong to the mesh or may belong to its neighbor mesh.
Step two: the trajectory data is converted into a blurred grid sequence.
(1) Traversing each track in a track database, judging according to the position of a sample point in the track, and if the sample point is certain to belong to the grid, namely is in an accurate area of the grid, representing the sample point by using the id of the grid; if the sample point possibly belongs to the grid, namely is in the fuzzy area of the grid, the sample point is expressed by the opposite number of the grid id (id is the identifier of the grid);
(2) For each track, integrating the representation methods of all sample points in (1) to form a fuzzy grid sequence of the track.
Step three: the grid is processed.
(1) Recording index positions of the grid: for each track in the track database, each grid through which the track passes is recorded in the form of an information pair (i, j), wherein i represents that the index of the track where the grid is positioned in the track database is i, and j represents that the index of the grid on the track is j;
(2) Calculating the support degree of the grid: and calculating the support degree of each grid in each track, and adding the grids with the support degree exceeding the minimum support degree into the candidate set.
Step four: and establishing a projection database for each grid in the candidate set, calculating the support degree of the newly generated candidate track pattern, and repeating the steps until all the track frequent patterns are mined by the candidate track pattern with the support degree exceeding the minimum support degree.
The beneficial effects are that: aiming at the track frequent pattern of the indoor moving object, the invention provides a novel indoor track frequent pattern mining method based on a fuzzy grid sequence, and a series of problems of grid boundaries and the like which are not considered by the traditional method are effectively solved. The method can provide reference for the location-based service (LBS) industry through the frequent pattern of the excavated track, and promote the vigorous development of the industry. Meanwhile, the track privacy revealing method and the track privacy revealing device can reduce the risk of track privacy revealing of the user and even can protect the track privacy of the user from being violated.
Drawings
Fig. 1 is a general flow chart of the method of the present invention.
Fig. 2 is a schematic view of a regular hexagonal grid.
Fig. 3 is a schematic diagram of calculating the projected distance of the perpendicular.
Fig. 4 is a flowchart of data preprocessing based on a regular hexagonal grid.
Fig. 5 is a schematic diagram of grid area division.
Fig. 6 is a schematic diagram of a trace.
Fig. 7 is a flow chart of processing a grid.
Fig. 8 is a flow chart of processing a candidate set.
Detailed Description
The invention will be further described with reference to the accompanying drawings.
The general flow of the present invention is shown in figure 1. The flow of the submodules included in the method is shown in fig. 2, 3, 4, 5, 6, 7 and 8, and is described in detail below with reference to the drawings.
The invention adopts regular hexagonal grids to divide the map; dividing the grid into an accurate area and a fuzzy area according to the projection distance of the vertical line; traversing each track in the track database, and converting track data into a fuzzy grid sequence according to grids of sample points in the tracks; recording the index position of each grid in the track, calculating the support degree of the index position, and adding the grids with the support degree exceeding the minimum support degree into the candidate set; establishing a projection database for each grid in the candidate set, and calculating the support degree of the newly generated candidate track mode; these steps are repeated until all track frequent patterns are mined. The specific implementation steps are as follows, and the general flow is shown in figure 1.
1. Data preprocessing based on regular hexagonal grid
Since the distances from the center of the regular hexagonal grid to the centers of the adjacent grids are equal, the indoor map is divided by adopting the regular hexagonal grids, so that the equality among all grids can be ensured, the specific processing steps are as follows, and the detailed details are shown in the accompanying drawings
2. Fig. 3, fig. 4 and fig. 5.
(1): the projection distance of the vertical line is defined, and the projection distance of a line segment from a certain point to the center point of the grid on the vertical line with the minimum included angle with the line segment in the grid is defined.
The calculation formula is as follows:
p_dist(p，g)＝dist*cos(30°-θ) 0°≤θ≤30°
p_dist(p，g)＝dist*cos(θ-30°) 30°＜θ≤60°
p_dist(p，g)＝dist*cos(90°-θ) 60°＜θ≤90°
wherein p_dist is the projection distance of a vertical line, p is a certain point, g is a grid, dist is the distance between the point p and the central point of the grid g, θ is the included angle between the connecting line of the point p and the central point of the grid g and the X axis, and the value range is 0 DEG to 90 DEG;
(2): according to the perpendicular projection distance in step (1), an area in the grid where the perpendicular projection distance is smaller than a given threshold r is regarded as an accurate area of the grid, which is an area where each grid contains and does not overlap each other. Points within the exact area of a grid belong to and only belong to the grid;
(3): according to the perpendicular projection distance in the step (1), the perpendicular projection distance in the grid is in the range of r toIs considered as a blurred region of the mesh (R is a given perpendicular projection distance threshold, R is the side length of a regular hexagonal mesh), which is a region that each mesh contains and overlaps with each other. Points within a blurred region of a mesh may belong to the mesh or may belong to its neighbor mesh.
2. Converting trajectory data into a fuzzy grid sequence
The original trajectory data is converted into a blurred grid sequence based on the trajectory of the blurred grid sequence, as the name implies. The specific process is shown below, and the schematic diagram is shown in fig. 6.
(1): traversing each track in a track database, judging according to the position of a sample point in the track, and if the sample point is certain to belong to the grid, namely is in an accurate area of the grid, representing the sample point by using the id of the grid; if the sample point possibly belongs to the grid, namely is in the fuzzy area of the grid, the sample point is expressed by the opposite number of the grid id (id is the identifier of the grid);
(2): for each track, integrating the representation methods of all sample points in (1) to form a fuzzy grid sequence of the track.
3. Processing grids
Processing the grid in the track is the core step of the invention. The specific process is as follows, and the flow chart of the process is shown in fig. 7.
(1): recording index positions of the grid: for each track in the track database, each grid through which the track passes is recorded in the form of an information pair (i, j), wherein i represents that the index of the track where the grid is positioned in the track database is i, and j represents that the index of the grid on the track is j;
(2): calculating the support degree of the grid: and calculating the support degree of each grid in each track, and adding the grids with the support degree exceeding the minimum support degree into the candidate set.
4. Processing candidate set
The frequent pattern processing of the tracks in the candidate set in the step 3 is proper, and is an important link of success of the invention and a last step. The specific process is as follows, and the detailed process flow is shown in fig. 8.
(1): and establishing a projection database for each grid in the candidate set, calculating the support degree of the newly generated candidate track pattern, and repeating the steps until all the track frequent patterns are mined by the candidate track pattern with the support degree exceeding the minimum support degree.
The excavation method proposed by the invention is generally described as follows:
input:
track database D T ；
Minimum Support min_support;
track frequent pattern s;
length k of track frequent pattern;
a grid index Set;
and (3) outputting:
a frequent track set FreTrajSet;
Claims (5)
1. an indoor track frequent pattern mining method based on a fuzzy grid sequence comprises the following steps:
(1) Indoor track data preprocessing: firstly, dividing an indoor map by utilizing a regular hexagonal grid, and then dividing the regular hexagonal grid into an accurate area and a fuzzy area according to the projection distance of a perpendicular;
(2) The trajectory data is converted into a fuzzy grid sequence: traversing each track in the track database, and converting track data into a fuzzy grid sequence according to grids passed by sample points in the track database;
(3) Processing the grid: recording the index position of each grid in the track, calculating the support degree of each grid, adding the grids with the support degree exceeding the minimum support degree into the candidate set, and discarding the rest grids;
(4) Processing the candidate set: and establishing a projection database for each grid in the candidate set, calculating the support degree of the newly generated candidate track pattern, adding the candidate track pattern with the support degree exceeding the minimum support degree into the candidate set, and repeating the steps until all track frequent patterns are mined.
2. The indoor track frequent pattern mining method based on a fuzzy grid sequence of claim 1, wherein the step (1) is a preprocessing of indoor track data, and the implementation method comprises:
(2-1) dividing the indoor map by using regular hexagonal grids, wherein the distances from the center of each regular hexagonal grid to the centers of adjacent grids are equal, so that the equality of all grids is ensured;
(2-2) defining a vertical projection distance, which is the projection distance of a line segment from a certain point to the center point of the grid on a vertical line with the minimum included angle with the line segment in the grid;
the calculation formula is as follows:
p_dist(p，g)＝dist*cos(30°-θ) 0°≤θ≤30°
p_dist(p，g)＝dist*cos(θ-30°) 30°＜θ≤60°
p_dist(p，g)＝dist*cos(90°-θ) 60°＜θ≤90°
wherein p_dist is the projection distance of a vertical line, p is a certain point, g is a grid, dist is the distance between the point p and the central point of the grid g, θ is the included angle between the connecting line of the point p and the central point of the grid g and the X axis, and the value range is 0 DEG to 90 DEG;
(2-3) regarding the area in which the perpendicular projection distance is smaller than the given threshold r as an accurate area of the mesh, which is an area in which each mesh contains and does not overlap with each other, according to the perpendicular projection distance in (2-2), points in the accurate area of a certain mesh belonging to and only belonging to the mesh;
(2-4) based on the perpendicular projection distance in (2-2), setting the perpendicular projection distance in the grid to be in the range of r toIs considered as a fuzzy area of a grid, R is a given vertical projection distance threshold, R is the side length of a regular hexagonal grid, and is the area that each grid contains and overlaps with each other, and points in the fuzzy area of a grid may belong to the grid or may belong to a neighbor grid thereof.
3. The method for mining indoor track frequent pattern based on fuzzy grid sequence as defined in claim 1, wherein the step (2) is to convert track data into fuzzy grid sequence, and the specific implementation method comprises:
(3-1) traversing each track in the track database, judging according to the position of a sample point in the track, and if the sample point is certain to belong to the grid, namely is in the accurate area of the grid, representing the sample point by using the id of the grid; if the sample point possibly belongs to the grid, namely is in a fuzzy area of the grid, the sample point is represented by the opposite number of the grid id, and the id is an identifier of the grid;
(3-2) for each track, integrating the representation of all sample points in (3-1) to form a fuzzy grid sequence of tracks.
4. The method for mining indoor track frequent patterns based on fuzzy grid sequences of claim 1, wherein step (3) is to process the grid, and the implementation method is as follows:
(4-1) recording index positions of the grid: for each track in the track database, each grid through which the track passes is recorded in the form of an information pair (i, j), wherein i represents that the index of the track where the grid is positioned in the track database is i, and j represents that the index of the grid on the track is j;
(4-2) calculating the support degree of the mesh: and calculating the support degree of each grid in each track, and adding the grids with the support degree exceeding the minimum support degree into the candidate set.
5. The method for mining indoor track frequent patterns based on fuzzy grid sequences of claim 1, wherein step (4) is to process the candidate set, and the implementation method is as follows:
and (5-1) establishing a projection database for each grid in the candidate set, calculating the support degree of the newly generated candidate track pattern, and repeating the steps until all track frequent patterns are mined by the candidate track pattern with the support degree exceeding the minimum support degree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810838419.2A CN109213941B (en) | 2018-07-24 | 2018-07-24 | Indoor track frequent pattern mining method based on fuzzy grid sequence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810838419.2A CN109213941B (en) | 2018-07-24 | 2018-07-24 | Indoor track frequent pattern mining method based on fuzzy grid sequence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109213941A CN109213941A (en) | 2019-01-15 |
| CN109213941B true CN109213941B (en) | 2023-07-18 |
Family
ID=64990321
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810838419.2A Active CN109213941B (en) | 2018-07-24 | 2018-07-24 | Indoor track frequent pattern mining method based on fuzzy grid sequence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109213941B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110909037B (en) * | 2019-10-09 | 2024-02-13 | 中国人民解放军战略支援部队信息工程大学 | Frequent track mode mining method and device |
| CN111143500B (en) * | 2019-12-27 | 2023-07-18 | 中国联合网络通信集团有限公司 | Visual area calculation method, terminal, control device and storage medium |
| CN112288029A (en) * | 2020-11-06 | 2021-01-29 | 电子科技大学 | Method for classifying vehicle tracks in urban road network |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10387457B2 (en) * | 2014-06-17 | 2019-08-20 | Sap Se | Grid-based analysis of geospatial trajectories |
| CN104331466B (en) * | 2014-10-31 | 2018-01-19 | 南京邮电大学 | Motion track sequence pattern Fast Mining Algorithm based on space-time proximity search |
| JP6543180B2 (en) * | 2015-12-08 | 2019-07-10 | 日本電信電話株式会社 | Destination prediction apparatus, method, and program |
| CN107016126A (en) * | 2017-05-12 | 2017-08-04 | 西南交通大学 | A kind of multi-user's model movement pattern method based on sequential mode mining |
-
2018
- 2018-07-24 CN CN201810838419.2A patent/CN109213941B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109213941A (en) | 2019-01-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109213941B (en) | Indoor track frequent pattern mining method based on fuzzy grid sequence | |
| CN107301254B (en) | Road network hot spot area mining method | |
| CN110418287B (en) | Population residence migration identification method based on mobile phone signaling | |
| CN102136192B (en) | Method for identifying trip mode based on mobile phone signal data | |
| CN101996515B (en) | Urban vector road network registration method based on local control in GIS-T | |
| Chen et al. | An indoor trajectory frequent pattern mining algorithm based on vague grid sequence | |
| CN104239556A (en) | Density clustering-based self-adaptive trajectory prediction method | |
| CN108091134B (en) | Universal data set generation method based on mobile phone signaling position track data | |
| CN103634902B (en) | Novel indoor positioning method based on fingerprint cluster | |
| CN109189949B (en) | A kind of population distribution calculation method | |
| CN102289466A (en) | K-nearest neighbor searching method based on regional coverage | |
| CN110334171A (en) | It is a kind of based on the space-time of Geohash with object method for digging | |
| CN110070121A (en) | A kind of quick approximate k nearest neighbor method based on tree strategy with balance K mean cluster | |
| CN105188030A (en) | Geographic grid mapping method of mobile network data | |
| CN105338540A (en) | Base station data modeling method and terminal | |
| CN108510008B (en) | Road network extraction method based on floating car track point spatial relationship and distribution | |
| CN104270785A (en) | Wireless network region problem positioning method based on geography grid aggregation | |
| CN107463585A (en) | Finger print data is put in storage processing method and processing device | |
| CN105335478B (en) | The method and apparatus for building urban land space multistory survey data semantic association | |
| KR101846294B1 (en) | Rainfall center tracking method based on weather radar | |
| CN103337084B (en) | A kind of line map spot automatic generation method based on man-made features feature | |
| CN114003623A (en) | Similar path typhoon retrieval method | |
| CN104899328A (en) | Method for rapidly seeking boundaries of land parcel in four directions | |
| CN103365911B (en) | Map space indexing method and system based on two dimension partitioned structure | |
| Zhou et al. | Complexity of functional urban spaces evolution in different aspects: Based on urban land use conversion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |