CN113344052B - Space-time frequent pattern mining method based on Thiessen polygons and K-means clustering - Google Patents
Space-time frequent pattern mining method based on Thiessen polygons and K-means clustering Download PDFInfo
- Publication number
- CN113344052B CN113344052B CN202110591189.6A CN202110591189A CN113344052B CN 113344052 B CN113344052 B CN 113344052B CN 202110591189 A CN202110591189 A CN 202110591189A CN 113344052 B CN113344052 B CN 113344052B
- Authority
- CN
- China
- Prior art keywords
- time
- space
- thiessen
- data
- stop
- 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
- 238000005065 mining Methods 0.000 title claims abstract description 17
- 238000000034 method Methods 0.000 title claims abstract description 13
- 238000003064 k means clustering Methods 0.000 title claims abstract description 12
- 238000011160 research Methods 0.000 claims abstract description 18
- 238000010606 normalization Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 12
- 230000000694 effects Effects 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 9
- 230000001131 transforming effect Effects 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000002776 aggregation Effects 0.000 claims description 3
- 238000004220 aggregation Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000007726 management method Methods 0.000 abstract description 3
- 230000006399 behavior Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
- G06N5/025—Extracting rules from data
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a space-time frequent pattern mining method based on Thiessen polygons and K-means clustering, which comprises the following steps: constructing Thiessen polygons in a research area, and numbering each polygon in the Thiessen polygons by using a Hilbert fractal curve; acquiring position information of each moving object in a research area, and converting longitude and latitude coordinates corresponding to the data set into a coordinate system in which Thiessen polygons are positioned; converting the travel track of the moving object from a curve in the space-time cube to a point in a new space-time space; clustering points in the new space-time space by using K-Means clustering; and taking the spatial characteristic value of each type after clustering to represent the frequent pattern, and restoring the clustered result to a three-dimensional space. According to the invention, the Thiessen polygon is utilized to describe a research area, so that the problem of data redundancy in a sparse place is avoided, and more accurate description is given to a place where the movement behavior is dense; and important reference values are improved for researching group travel demands, travel recommendations, traffic planning and management.
Description
Technical Field
The invention relates to a data mining technology, in particular to a space-time frequent pattern mining method based on Thiessen polygons and K-means clustering.
Background
Along with rapid development of geographic information acquisition equipment, the position information of a mobile object on a road is more and more abundant, and the data volume is also more and more large. The time-space frequent mode is a mode frequently appearing in a data set with time-space attributes, the origin-destination position and the occurrence time of travel directly reflect travel demands and time distribution rules of mobile objects, the application direction of travel mode mining is very wide, and important references are provided for the problems of study group travel demands, travel recommendation, traffic planning and management and the like.
The traditional behavior pattern mining technology is mostly researched based on a regular grid or a traffic cell, and the regular grid has the problems that the scale cannot be well adapted to the distribution of the behaviors of a moving object, a large amount of data redundancy exists, and the position fineness is insufficient; traffic cells are limited by factors such as road network, population density, economic development and the like, so that the mining result is complex.
Disclosure of Invention
The invention aims to: aiming at the problems, the invention aims to provide a space-time frequent pattern mining method based on Thiessen polygons and K-means clustering.
The technical scheme is as follows: the space-time frequent pattern mining method based on Thiessen polygons and K-means clustering comprises the following steps:
(1) According to the distribution condition of interest points in a research area, constructing Thiessen polygons in the research area, wherein the interest points are place names corresponding to the research area, and numbering each polygon in the Thiessen polygons by using a Hilbert fractal curve;
(2) Acquiring the position information of each moving object in a research area, forming the position information into a data set, and converting longitude and latitude coordinates corresponding to effective data in the data set into a coordinate system where Thiessen polygons are located, wherein the coordinate system comprises two dimensions of time and space;
(3) Carrying out data preprocessing on the effective data in two dimensions of time and space, and converting the travel track of the moving object from a curve in a space-time cube into points in a new space-time space;
(4) Clustering the points in the new space-time space in the step (3) by using K-Means clustering;
(5) And taking the spatial characteristic values of each type after clustering to represent the frequent patterns, restoring the clustered results to a three-dimensional space, including position information and time information, and performing visual display in a vector map.
Further, the step (1) numbering includes:
(11) Calculating the area of each polygon in the Thiessen polygons, establishing a square with the minimum area and the like, and dividing the Thiessen polygons into grids;
(12) Numbering each square, and setting the side length of the square as the side length of the Hilbert fractal curve to ensure that each polygon in the Thiessen polygon can be traversed when the fractal curve is conforming to the generation rule;
(13) Each polygon is numbered according to the corresponding number of the covered square, when one square is covered, the number of the covered square is numbered, and when a plurality of squares are covered, the number of the polygon is the number of the square at the middle position.
Further, a coordinate system is established, the longitude and latitude of the research area are taken as x-axis and the y-axis, the movement occurrence Time of the moving object is taken as z-axis, the position information in the step (2) comprises the starting point longitude start_x, the starting point latitude start_y, the ending point longitude stop_x, the ending point latitude stop_y, the starting Time start_time and the ending Time stop_time of the moving object, and each moving object position information corresponds to a unique number; the data set comprises effective data and ineffective data, wherein the ineffective data is that all data in the position information are repeated or the longitude and latitude of the origin and destination are unchanged, and the ineffective data are removed to be the effective data.
Further, the step (3) of data preprocessing includes:
(31) Transforming the space dimension attribute of the effective data, converting the longitude and latitude of each effective data into the located Thiessen polygon number value by using a function, wherein the expression is as follows:
Start_ID＝Thiessen(Start_x,Start_y)
Stop_ID＝Thiessen(Stop_x,Stop_y)
thiessen () is a conversion function, start_ID and stop_ID are respectively a one-dimensional value corresponding to the Start and Stop points, which is obtained by converting Thiessen () into a Start and Stop point coordinate pair, and the start_ID and the stop_ID are used as space characteristic values;
(32) Transforming the effective data time dimension attribute: each valid data duration is [ start_time, stop_time ], a Time period intermediate value act_time is taken as a transformed Time characteristic value, and the expression is as follows:
(33) Respectively carrying out normalization processing on the time dimension data and the space dimension data, wherein the expression is as follows:
normal_value is a new Value obtained after normalization processing, value is a data Value of time or space dimension before normalization processing, and Max and Min are respectively a maximum Value and a minimum Value in a data set under the same dimension as Value;
(34) A New Time variable New_Time is constructed, and the expression is:
New_Time＝σ*Normal_T ime
normal_time is a value obtained after normalization processing of Time variable, sigma is a ratio of Time feature mode describing capacity to space feature mode describing capacity, sigma is a value related to effective data space and Time feature, if the Time attribute of the data presents aggregation feature in a certain Time period, the Time attribute should be properly stretched, and sigma is more than 1; when the spatial attribute of the data presents a spatially scattered characteristic, the spatial attribute of the data is properly compressed, and sigma is smaller than 1;
(35) The travel tracks of all moving objects are converted into points of a New space-Time space, and coordinates are (start_ID, stop_ID, new_Time).
Further, the step (4) clustering includes:
(41) Calculating the profile coefficient of the effective data, and selecting the class number K corresponding to the maximum profile coefficient according to the change relation of the profile coefficient along with the class number;
(42) Dividing the points converted in the step (35) into K groups, and selecting K objects as initial clustering centers;
(43) Calculating the distance between each point and each cluster center, distributing each point to the cluster center closest to the point to form different sub-clusters, and recalculating the cluster center;
(44) Repeating the steps (41) - (43) until all points are distributed to complete the end circulation, and obtaining a clustering result.
Further, the step (5) of restoring to the three-dimensional space comprises two dimensions of time and space, and the specific process is as follows:
(51) According to the clustering result of the step (44), taking a frequent pattern that each class of space characteristic value mean represents the class, reversely mapping the Start-Stop point corresponding to one-dimensional value start_ID and stop_ID of the frequent pattern to the map range of the entity space respectively, and obtaining two areas in the entity space as the transformation result, namely, the position range of the corresponding Thiessen polygon:
Thiessen -1 () As an inverse function of Thiessen (), the function is to reduce the Thiessen polygon number to a corresponding Thiessen polygon range,and->The numbers of Thiessen polygons at the starting and ending points of the type x frequent pattern are respectively corresponding, and the area_Start and area_stop are Thiessen polygons at the starting and ending points of the type frequent pattern, which are obtained by inverse mapping of Thiessen polygon number values;
(52) Taking the mean value of time eigenvalues of cluster xAt offset of +.>The period of intra-interval fluctuation is represented by act_time as a result of the inverse mapping of the Time characteristics, as follows:
d is the total duration of the sample including the time feature mean point for all activity timesMean value between, n is the mean value point of time characteristics of crossing cluster x of meeting the activity timeI represents the ith sample satisfying the condition, T i stop_Time and T i Start_time represents the Time at which the activity of sample i ends and begins, respectively.
The beneficial effects are that: compared with the prior art, the invention has the remarkable advantages that:
1. according to the invention, the Thiessen polygon is utilized to describe a research area, so that the problem of data redundancy in a sparse place is avoided, and more accurate description is given to a place where the movement behavior is dense;
2. numbering Thiessen polygons to reduce the dimension of data and information storage space;
3. the final mining result can be visually presented, so that important reference value is improved for researching group travel requirements, travel recommendation, traffic planning and management.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic diagram of the conversion of valid data from a spatiotemporal cube to a new spatiotemporal space;
FIG. 3 is a schematic diagram of a pattern reduction inverse map;
fig. 4 is a hotspot path mining result display.
Detailed Description
Taking taxi order data of city of Sichuan province as an example, illustrating the space-time frequent pattern mining method based on Thiessen polygons and K-means clustering according to the embodiment, a flowchart is shown in FIG. 1, and specifically includes the following steps:
(1) According to the distribution condition of interest points in a research area, the interest points are specifically each place name in the research area, the equal-speed expansion is carried out by taking each interest point as a base point, a Thiessen polygon is constructed in the research area, and each polygon in the Thiessen polygon is numbered by using a Hilbert fractal curve:
(11) Calculating the area of each polygon in the Thiessen polygons, establishing a square with the minimum area and the like, and dividing the Thiessen polygons into grids;
(12) Numbering each square, and setting the side length of the square as the side length of the Hilbert fractal curve;
(13) Each polygon is numbered according to the corresponding number of the covered square, when one square is covered, the number of the covered square is numbered, and when a plurality of squares are covered, the number of the polygon is the number of the square at the middle position.
(2) Acquiring the position information of each moving object in a research area, forming the position information into a data set, and converting longitude and latitude coordinates corresponding to effective data in the data set into a coordinate system where Thiessen polygons are located, wherein the coordinate system comprises two dimensions of time and space;
establishing a coordinate system, taking longitude and latitude of a research area as an x axis and a y axis respectively, taking movement occurrence Time of a moving object as a z axis, wherein the position information comprises starting point longitude start_x, starting point latitude start_y, ending point longitude stop_x, ending point latitude stop_y, starting Time start_time and ending Time stop_time of the moving object, each moving object position information corresponds to a unique number, and table 1 lists the position information of 10 groups of taxi orders, wherein the Time is expressed by converting real Time into seconds.
The data set comprises effective data and ineffective data, wherein the ineffective data is that all data in the position information are repeated or the longitude and latitude of the origin and destination are unchanged, and the rest of the ineffective data is effective data.
(3) Data preprocessing is carried out on the effective data in two dimensions of time and space, and the travel track of the moving object is converted into points in a new space-time space from a curve in a space-time cube, as shown in figure 2;
(31) Transforming the space dimension attribute of the effective data, converting the longitude and latitude of each effective data into the located Thiessen polygon number value by using a function, wherein the expression is as follows:
Start_ID＝Thiessen(Start_x,Start_y)
Stop_ID＝Thiessen(Stop_x,Stop_y)
thiessen () is a conversion function, start_ID and stop_ID are respectively a one-dimensional value corresponding to the Start and Stop points, which is obtained by converting Thiessen () into a Start and Stop point coordinate pair, and the start_ID and the stop_ID are used as space characteristic values;
(32) Transforming the effective data time dimension attribute: each valid data duration is [ start_time, stop_time ], a Time period intermediate value act_time is taken as a transformed Time characteristic value, and the expression is as follows:
(33) Respectively carrying out normalization processing on the time dimension data and the space dimension data, wherein the expression is as follows:
normal_value is a new Value obtained after normalization processing, value is a data Value of time or space dimension before normalization processing, and Max and Min are respectively a maximum Value and a minimum Value in a data set under the same dimension as Value;
(34) A New Time variable New_Time is constructed, and the expression is:
New_Time＝σ*Normal_T ime
normal_time is a value obtained after normalization processing of Time variable, sigma is a ratio of Time feature mode describing capacity to space feature mode describing capacity, sigma is a value related to effective data space and Time feature, if the Time attribute of the data presents aggregation feature in a certain Time period, the Time attribute should be properly stretched, and sigma is more than 1; when the spatial attribute of the data presents a spatially scattered characteristic, the spatial attribute of the data is properly compressed, and sigma is smaller than 1;
(35) The travel track of all moving objects is converted into points in the New space-Time space, the coordinates are (start_id, stop_id, new_time), and table 2 is the coordinates corresponding to the points in the New space-Time space converted from the 10 sets of taxi order data in table 1.
TABLE 1 position information of taxi orders
TABLE 2 conversion of taxi orders to New spatio-temporal space interior Point coordinates
(4) Clustering the points in the new space-time space in the step (3) by using K-Means clustering:
(41) Calculating the profile coefficient of the effective data, and selecting the class number K corresponding to the maximum profile coefficient according to the change relation of the profile coefficient along with the class number;
(42) Dividing the points converted in the step (35) into K groups, and selecting K objects as initial clustering centers;
(43) Calculating the distance between each point and each cluster center, distributing each point to the cluster center closest to the point to form different sub-clusters, and recalculating the cluster center;
(44) Repeating the steps (41) - (43) until all points are distributed to complete the end circulation, and obtaining a clustering result.
(5) And taking the spatial characteristic values of each type after clustering to represent the frequent patterns, restoring the clustered results to a three-dimensional space, including position information and time information, and performing visual display in a vector map.
(51) According to the clustering result of the step (44), taking a frequent pattern that each class of space characteristic value mean represents the class, reversely mapping the Start-Stop point corresponding to one-dimensional value start_ID and stop_ID of the frequent pattern to the map range of the entity space respectively, and obtaining two areas in the entity space as the transformation result, namely, the position range of the corresponding Thiessen polygon:
Thiessen -1 () As an inverse function of Thiessen (),and->The numbers of Thiessen polygons at the starting and ending points of the type x frequent pattern are respectively corresponding, and the area_Start and area_stop are Thiessen polygons at the starting and ending points of the type frequent pattern, which are obtained by inverse mapping of Thiessen polygon number values;
(52) Taking the mean value of time eigenvalues of cluster xAt offset of +.>The period of intra-interval fluctuation is represented by act_time as a result of Time feature inverse mapping, as shown in fig. 3, as follows:
d is the average value of the total duration of the samples with all the activity time including the time characteristic average value point, and n is the average value point of the time characteristic average value of the cluster x crossing the activity timeI represents the ith sample satisfying the condition, T i stop_Time and T i Start_time represents the Time at which the activity of sample i ends and begins, respectively. Fig. 4 is a view showing the result of hot spot path mining in a certain area.
Claims (5)
1. The space-time frequent pattern mining method based on Thiessen polygons and K-means clustering is characterized by comprising the following steps:
(1) Constructing Thiessen polygons in the research area according to the distribution condition of interest points in the research area, and numbering each polygon in the Thiessen polygons by using a Hilbert fractal curve;
(2) Acquiring the position information of each moving object in a research area, forming the position information into a data set, and converting longitude and latitude coordinates corresponding to effective data in the data set into a coordinate system where Thiessen polygons are located, wherein the coordinate system comprises two dimensions of time and space;
(3) Carrying out data preprocessing on the effective data in two dimensions of time and space, and converting the travel track of the moving object from a curve in a space-time cube into points in a new space-time space;
(4) Clustering the points in the new space-time space in the step (3) by using K-Means clustering;
(5) Taking the clustered spatial feature values of each class to represent the frequent modes, restoring the clustered results to a three-dimensional space, including position information and time information, and performing visual display in a vector map;
the step (3) of data preprocessing comprises the following steps:
(31) Transforming the space dimension attribute of the effective data, converting the longitude and latitude of each effective data into the located Thiessen polygon number value by using a function, wherein the expression is as follows:
Start_ID＝Thiessen(Start_x,Start_y)
Stop_ID＝Thiessen(Stop_x,Stop_y)
thiessen () is a conversion function, start_ID and stop_ID are respectively a one-dimensional value corresponding to the Start and Stop points, which is obtained by converting Thiessen () into a Start and Stop point coordinate pair, and the start_ID and the stop_ID are used as space characteristic values;
(32) Transforming the effective data time dimension attribute: each valid data duration is [ start_time, stop_time ], a Time period intermediate value act_time is taken as a transformed Time characteristic value, and the expression is as follows:
(33) Respectively carrying out normalization processing on the time dimension data and the space dimension data, wherein the expression is as follows:
the normal_value is a new Value obtained after normalization processing, the normal_value comprises normal_time, the Value is a data Value of Time or space dimension before normalization processing, and Max and Min are respectively a maximum Value and a minimum Value in a data set under the same dimension as the Value;
(34) A New Time variable New_Time is constructed, and the expression is:
New_Time＝σ*Normal_Time
normal_time is a value obtained after normalization processing of Time variable, sigma is a ratio of Time feature mode describing capacity to space feature mode describing capacity, sigma is a value related to effective data space and Time feature, if the Time attribute of the data presents aggregation feature in a certain Time period, the Time attribute should be properly stretched, and sigma is more than 1; when the spatial attribute of the data presents a spatially scattered characteristic, the spatial attribute of the data is properly compressed, and sigma is smaller than 1;
(35) The travel tracks of all moving objects are converted into points of a New space-Time space, and coordinates are (start_ID, stop_ID, new_Time).
2. The spatio-temporal frequent pattern mining method of claim 1, wherein step (1) numbering comprises:
(11) Calculating the area of each polygon in the Thiessen polygons, establishing a square with the minimum area and the like, and dividing the Thiessen polygons into grids;
(12) Numbering each square, and setting the side length of the square as the side length of the Hilbert fractal curve;
(13) And numbering each polygon according to the number corresponding to the covered square, wherein when one square is covered, the number of the polygon is the number of the covered square, and when a plurality of squares are covered, the number of the polygon is the number of the square at the middle position.
3. The method according to claim 2, wherein a coordinate system is established, the longitude and latitude of the research area are respectively taken as an x axis and a y axis, the movement occurrence Time of the moving object is taken as a z axis, the position information in the step (2) includes a Start longitude start_x, a Start latitude start_y, a Stop longitude stop_x, a Stop latitude stop_y, a Start Time start_time and a Stop Time stop_time of the moving object, and each moving object position information corresponds to a unique number; the data set comprises effective data and ineffective data, wherein the ineffective data is that all data in the position information are repeated or the longitude and latitude of the origin and destination are unchanged, and the ineffective data are removed to be the effective data.
4. The spatio-temporal frequent pattern mining method of claim 3, wherein step (4) clustering comprises:
(41) Calculating the profile coefficient of the effective data, and selecting the class number K corresponding to the maximum profile coefficient according to the change relation of the profile coefficient along with the class number;
(42) Dividing the points converted in the step (35) into K groups, and selecting K objects as initial clustering centers;
(43) Calculating the distance between each point and each cluster center, distributing each point to the cluster center closest to the point to form different sub-clusters, and recalculating the cluster center;
(44) Repeating the steps (41) - (43) until all points are distributed to complete the end circulation, and obtaining a clustering result.
5. The method of mining space-time frequent patterns according to claim 4, wherein the step (5) is a three-dimensional space, comprising two dimensions of time and space, comprising the following steps:
(51) According to the clustering result of the step (44), taking a frequent pattern of each class of space characteristic value mean value representing the class, reversely mapping a start_ID and a stop_ID corresponding to one-dimensional values of the frequent pattern to a map of the entity space respectively, wherein the transformation result is two areas in the entity space, namely, the position range of the corresponding Thiessen polygon:
Thiessen -1 () As an inverse function of Thiessen (),and->The numbers of Thiessen polygons at the starting and ending points of the type x frequent pattern are respectively corresponding, and the area_Start and area_stop are Thiessen polygons at the starting and ending points of the type frequent pattern, which are obtained by inverse mapping of Thiessen polygon number values;
(52) Taking the mean value of the cluster xAt offset of +.>The period of intra-interval fluctuation is represented by act_time as a result of the inverse mapping of the Time characteristics, as follows:
d is the average value of the total duration of the samples with all the activity time including the time characteristic average value point, and n is the average value point of the time characteristic average value of the cluster x crossing the activity timeI represents the ith sample satisfying the condition, T i stop_Time and T i Start_time represents the Time at which the activity of sample i ends and begins, respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110591189.6A CN113344052B (en) | 2021-05-28 | 2021-05-28 | Space-time frequent pattern mining method based on Thiessen polygons and K-means clustering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110591189.6A CN113344052B (en) | 2021-05-28 | 2021-05-28 | Space-time frequent pattern mining method based on Thiessen polygons and K-means clustering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113344052A CN113344052A (en) | 2021-09-03 |
| CN113344052B true CN113344052B (en) | 2024-04-09 |
Family
ID=77472018
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110591189.6A Active CN113344052B (en) | 2021-05-28 | 2021-05-28 | Space-time frequent pattern mining method based on Thiessen polygons and K-means clustering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113344052B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108230020A (en) * | 2017-12-28 | 2018-06-29 | 西北大学 | A kind of method excavated based on the frequent region of multi-dimensional time granularity space-time |
| CN110909037A (en) * | 2019-10-09 | 2020-03-24 | 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 |
| CN111459997A (en) * | 2020-03-16 | 2020-07-28 | 中国科学院计算技术研究所 | Frequent mode increment mining method of space-time trajectory data and electronic equipment |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7567972B2 (en) * | 2003-05-08 | 2009-07-28 | International Business Machines Corporation | Method and system for data mining in high dimensional data spaces |
-
2021
- 2021-05-28 CN CN202110591189.6A patent/CN113344052B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108230020A (en) * | 2017-12-28 | 2018-06-29 | 西北大学 | A kind of method excavated based on the frequent region of multi-dimensional time granularity space-time |
| CN110909037A (en) * | 2019-10-09 | 2020-03-24 | 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 |
| CN111459997A (en) * | 2020-03-16 | 2020-07-28 | 中国科学院计算技术研究所 | Frequent mode increment mining method of space-time trajectory data and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113344052A (en) | 2021-09-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107247938B (en) | high-resolution remote sensing image urban building function classification method | |
| CN108268597B (en) | Moving target activity probability map construction and behavior intention identification method | |
| CN109359162B (en) | GIS-based school site selection method | |
| CN110321443B (en) | Three-dimensional live-action model database construction method and device and data service system | |
| CN107688906B (en) | Multi-method fused transmission line meteorological element downscaling analysis system and method | |
| CN110232398A (en) | A kind of road network sub-area division and its appraisal procedure based on Canopy+Kmeans cluster | |
| CN108428015B (en) | Wind power prediction method based on historical meteorological data and random simulation | |
| CN106600617A (en) | Method of extracting building contour line from Lidar point cloud data based on curvature | |
| CN111918298B (en) | Clustering-based site planning method and device, electronic equipment and storage medium | |
| CN106162544A (en) | A kind of generation method and apparatus of geography fence | |
| CN113487730A (en) | Urban three-dimensional automatic modeling method based on laser radar point cloud data | |
| CN115512216A (en) | City functional area fine recognition method coupling block space-time characteristics and ensemble learning | |
| CN113137919B (en) | Laser point cloud rasterization method | |
| CN105354882A (en) | Method for constructing big data architecture based three-dimensional panoramic display platform for large-spatial-range electricity transmission | |
| CN114661744B (en) | Terrain database updating method and system based on deep learning | |
| CN111833224A (en) | Urban main and auxiliary center boundary identification method based on population grid data | |
| CN112381644A (en) | Credit scene risk user assessment method based on space variable reasoning | |
| US20140009339A1 (en) | Methods of and systems for extracting patterns of human extent and density from geographically anchored radio signal sources | |
| CN113344052B (en) | Space-time frequent pattern mining method based on Thiessen polygons and K-means clustering | |
| CN110991562B (en) | Animal group geographic division method based on species composition characteristics | |
| Ali et al. | A novel computational paradigm for creating a Triangular Irregular Network (TIN) from LiDAR data | |
| CN113240265A (en) | Urban space division method based on multi-mode traffic data | |
| Abdelguerfi et al. | Representation of 3-D elevation in terrain databases using hierarchical triangulated irregular networks: a comparative analysis | |
| CN111008730B (en) | Crowd concentration prediction model construction method and device based on urban space structure | |
| Kidner et al. | Multiscale terrain and topographic modelling with the implicit TIN |
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 |