CN109508750A - The clustering method of user origin and destination, device and storage medium - Google Patents
The clustering method of user origin and destination, device and storage medium Download PDFInfo
- Publication number
- CN109508750A CN109508750A CN201811465250.7A CN201811465250A CN109508750A CN 109508750 A CN109508750 A CN 109508750A CN 201811465250 A CN201811465250 A CN 201811465250A CN 109508750 A CN109508750 A CN 109508750A
- Authority
- CN
- China
- Prior art keywords
- cluster
- data
- clustering
- density
- local density
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
Abstract
The present invention provides a kind of clustering method of user origin and destination, device and storage medium, wherein this method comprises: travelling data collection includes at least one travelling data, and travelling data includes coordinate data firstly, obtaining vehicle-mounted user corresponding travelling data collection；Then, according to the corresponding local density of each travelling data and density distance, cluster numbers are determined；Then, according to preset rules and cluster numbers, the travelling data concentrated to travelling data is clustered, and obtains clustering cluster；Further, according to the cluster Nuclear Data in clustering cluster, cluster result is obtained, wherein cluster result indicates the beginning or end of vehicle-mounted user.Method provided by the invention improves the accuracy of cluster result by determining cluster numbers according to the corresponding local density of travelling data and density distance；Further, by obtaining cluster result, influence of the outlier to cluster result being reduced, to improve the accuracy of cluster result according to the cluster Nuclear Data in clustering cluster.
Description
Technical field
The present invention relates to field of intelligent transportation technology more particularly to a kind of clustering method of user origin and destination, device and
Storage medium.
Background technique
Car-mounted device usually will record the travelling data of vehicle, such as: latitude and longitude coordinates, travel speed and time etc.
Deng.(origin and destination: starting point and end are indicated by the origin and destination of these travelling datas, our available users gone on a journey every time
Point), to provide better vehicle-mounted service for vehicle-mounted user.
In the prior art, it generallys use and travelling data is gathered by the method based on density of representative of DBSCAN algorithm
Alanysis obtains the origin and destination of vehicle-mounted user.
But the above method is higher to the sensitivity of parameter, and when the sparse degree difference of cluster, identical judgement mark
Standard may destroy the natural structure of cluster, i.e., diluter cluster can be divided into multiple classes or density is biggish and closer from obtaining
Multiple classes can be merged into a cluster, lead to origin and destination mass center and practical origin and destination mass center have biggish offset, that is,
It says, cluster result accuracy is lower.
Summary of the invention
The present invention provides a kind of clustering method of user origin and destination, device and storage medium, to improve cluster result
Accuracy.
In a first aspect, the present invention provides a kind of clustering method of user origin and destination, comprising:
The corresponding travelling data collection of vehicle-mounted user in preset time period is obtained, the travelling data collection includes at least one row
Car data, the travelling data include coordinate data；
According to the corresponding local density of each travelling data and density distance, cluster numbers are determined；
According to preset rules and the cluster numbers, the travelling data concentrated to the travelling data is clustered, and is obtained
Clustering cluster；
According to the cluster Nuclear Data in the clustering cluster, cluster result is obtained, the cluster result indicates the vehicle-mounted user
Beginning or end.
Optionally, described according to the corresponding local density of each travelling data and density distance, determine cluster numbers
Before, further includes:
According to default truncation distance, the corresponding local density of each travelling data and density distance are obtained.
It is optionally, described that cluster numbers are determined according to the corresponding local density of each travelling data and density distance,
Include:
The corresponding local density of each travelling data and density distance are multiplied respectively, according to the product
Slope variation trend, be determined as cluster numbers.
Optionally, the cluster Nuclear Data according in the clustering cluster, obtain cluster result before, further includes:
The cluster Nuclear Data in the clustering cluster is determined in the following manner:
According to the corresponding local density of the clustering cluster middle rolling car data, the corresponding local density's ginseng of the clustering cluster is obtained
Examine value；
According to the corresponding local density of the clustering cluster middle rolling car data and the corresponding local density's ginseng of the clustering cluster
Value is examined, determines the cluster Nuclear Data in the clustering cluster.
Optionally, described according to the corresponding local density of the clustering cluster middle rolling car data, it is corresponding to obtain the clustering cluster
Local density's reference value, comprising:
By the average value of the corresponding local density of the clustering cluster middle rolling car data, it is determined as local density's reference
Value.
Optionally, described corresponding according to the corresponding local density of the clustering cluster middle rolling car data and the clustering cluster
Local density's reference value determines the cluster Nuclear Data in the clustering cluster, comprising:
If the corresponding local density of the clustering cluster middle rolling car data is greater than the corresponding local density's reference of the clustering cluster
Value, it is determined that the travelling data is cluster Nuclear Data；
If it is close that the corresponding local density of the clustering cluster middle rolling car data is less than or equal to the corresponding part of the clustering cluster
Spend reference value, it is determined that the travelling data is the dizzy data of cluster.
Optionally, described according to cluster Nuclear Data in each clustering cluster, obtain cluster result, comprising:
By the average value of the cluster Nuclear Data in each clustering cluster, it is determined as cluster result.
Second aspect, the present invention provide a kind of user origin and destination cluster analyzing device, which includes:
First obtains module, for obtaining the corresponding travelling data collection of vehicle-mounted user in preset time period, the driving number
It include at least one travelling data according to collection, the travelling data includes coordinate data；
First determining module, for determining according to the corresponding local density of each travelling data and density distance
Cluster numbers；
Cluster module obtains poly- for being clustered to the travelling data according to preset rules and the cluster numbers
Class cluster；
Computing module, for obtaining cluster result according to the cluster Nuclear Data in each clustering cluster, the cluster result is indicated
The beginning or end of the vehicle-mounted user.
The third aspect, the present invention provide a kind of user origin and destination cluster analyzing device, which includes: memory and processing
Device；
The memory stores program instruction；
Described program instruction by the processor when being executed, to execute method described in first aspect.
Fourth aspect, the present invention also provides a kind of storage mediums, comprising: program；
Described program is when being executed by processor, to execute method described in first aspect.
The present invention provides a kind of clustering method of user origin and destination, device and storage medium, wherein this method comprises:
Firstly, obtaining the corresponding travelling data collection of vehicle-mounted user, travelling data collection includes at least one travelling data, and travelling data includes
Coordinate data；Then, according to the corresponding local density of each travelling data and density distance, cluster numbers are determined；Then, according to
Preset rules and cluster numbers, the travelling data concentrated to travelling data cluster, and obtain clustering cluster；Further, according to poly-
Cluster Nuclear Data in class cluster obtains cluster result, wherein cluster result indicates the beginning or end of vehicle-mounted user.The present invention mentions
The method of confession, by determining cluster numbers, improving cluster result according to the corresponding local density of travelling data and density distance
Accuracy；Further, it by obtaining cluster result according to the cluster Nuclear Data in clustering cluster, reduces outlier and cluster is tied
The influence of fruit, to improve the accuracy of cluster result.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to do one simply to introduce, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without any creative labor, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the flow diagram of user origin and destination clustering method embodiment one provided by the invention；
Fig. 2 is the flow diagram of user origin and destination clustering method embodiment two provided by the invention；
Fig. 3 is the flow diagram of user origin and destination clustering method embodiment three provided by the invention；
Fig. 4 is the flow diagram of clustering method example IV in user origin and destination provided by the invention；
Fig. 5 is the structural schematic diagram of user origin and destination cluster analyzing device embodiment one provided by the invention；
Fig. 6 is the structural schematic diagram of user origin and destination cluster analyzing device embodiment two provided by the invention；
Fig. 7 is the structural schematic diagram of user origin and destination cluster analyzing device embodiment three provided by the invention；
Fig. 8 is the structural schematic diagram of user origin and destination cluster analyzing device example IV provided by the invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without making creative work, shall fall within the protection scope of the present invention.
Car-mounted device usually will record the travelling data of vehicle, such as: latitude and longitude coordinates, travel speed and time etc.
Deng.(origin and destination: starting point and end are indicated by the origin and destination of these travelling datas, our available users gone on a journey every time
Point).But since there are drift phenomenons for data itself, so the coordinate for same starting point will not when each run starts
It is consistent completely, but a scatterplot cluster for being polymerized to cluster similarly terminates for the coordinate of same terminal in each run
When will not be consistent completely and a scatterplot cluster for being polymerized to cluster.Therefore, it is necessary to be obtained to use according to these discrete datas
The true coordinate for the origin and destination that family is gone on a journey every time, to provide better vehicle-mounted service for vehicle-mounted user.
In the prior art, generally use use using DBSCAN algorithm as representative based on the method for density to travelling data into
Row clustering obtains the origin and destination of vehicle-mounted user.But the above method is higher to the sensitivity of parameter, it is dilute when cluster
When dredging degree difference, identical criterion may destroy the natural structure of cluster, i.e., diluter cluster can be divided into more
A class or density it is biggish and from nearlyr volume class can be merged into a cluster, lead to origin and destination mass center and practical origin and destination matter
The heart has biggish offset, that is to say, that cluster result accuracy is lower.
In the prior art, it can also use and travelling data is carried out based on the method for division using K-Means algorithm as representative
Clustering obtains the origin and destination of vehicle-mounted user.But need rule of thumb to estimate number of clusters using the above method, cause
Cluster result accuracy is lower.
Therefore, it is based on defect existing in the prior art, the present invention provides a kind of clustering method of user origin and destination, with
The accuracy for improving cluster result, to preferably provide vehicle-mounted service for vehicle-mounted user.
Fig. 1 is the flow diagram of user origin and destination clustering method embodiment one provided by the invention.The present embodiment
Shown in the clustering method of user origin and destination executing subject can for user origin and destination provided in an embodiment of the present invention cluster
Analytical equipment, the device can be computer, personal palm PC etc..
As shown in Figure 1, the method for the present embodiment includes:
S101, the corresponding travelling data collection of vehicle-mounted user in preset time period is obtained.
Wherein, travelling data collection includes at least one travelling data, and in various embodiments of the present invention, and travelling data includes
Coordinate data.
In general, vehicle installation car-mounted device can be bound with the information of vehicle-mounted user, vehicle during traveling,
Car-mounted device will record the travelling datas such as longitude and latitude, running speed, the timestamp of vehicle, these travelling datas can be stored in vehicle-mounted
In the memory module of device, stored alternatively, car-mounted device can also report to travelling data in corresponding server.This reality
Apply in example, user origin and destination cluster analyzing device can obtain above-mentioned travelling data, for example, if travelling data can be stored in it is vehicle-mounted
In the memory module of device, user origin and destination cluster analyzing device is connect with car-mounted device, obtains required travelling data；If row
Car data is stored in the corresponding server of car-mounted device, and user origin and destination cluster analyzing device can send number to the server
According to acquisition instruction, so that server based on data acquisition instruction, is sent to user origin and destination cluster point for corresponding travelling data
Analysis apparatus.
Alternatively possible implementation, user origin and destination cluster analyzing device can also be provided data input screen, can lead to
It crosses the data input screen and imports travelling data collection in the cluster analyzing device of user origin and destination in the form of document or table, or
Travelling data can also be input in the cluster analyzing device of user origin and destination by person, user by data input screen.
It should be noted that the specific implementation of travelling data is obtained for user origin and destination cluster analyzing device, this
Invention the present embodiment is with no restrictions.
S102, according to the corresponding local density of travelling data and density distance, determine cluster numbers.
In the present embodiment, local density indicates to concentrate in travelling data, be less than at a distance from travelling data default truncation away from
From other travelling datas number, wherein the distance between travelling data representation space geographic distance, the geographical space distance
Mercator projection principle in the prior art can be used to be calculated.Local density is bigger, indicates surrounding existing data
It is more；Local density is smaller, indicates that existing data are fewer around it.
Density distance, when some data have the maximum local density when, density distance indicate data set in the data away from
From farthest the distance between data point and the point；Otherwise, density distance indicates that all local densities are greater than in the data of the point,
With the data apart from the distance between the smallest data and the data.In the embodiment of the present invention, when the local density of travelling data
It is larger, and density distance it is also larger when, indicate at least exist a data more more dense than it, and they the distance between also compared with
Greatly；If the local density of travelling data is small, and density apart from it is larger when, indicate that the data are relatively isolated, and far from cluster numbers, i.e.,
The travelling data can be considered as outlier.
The corresponding local density of each travelling data and density distance are carried out phase by a kind of possible implementation respectively
Multiply, the quantity that product is greater than the travelling data of preset threshold is determined as cluster numbers.
Travelling data is concentrated the corresponding local density of each travelling data and close by another alternatively possible implementation
Degree distance is multiplied respectively, later according to the variation tendency of product, determines cluster numbers.It is, of course, also possible to using other modes
In conjunction with the characteristic parameter of travelling data, to determine cluster numbers.
In this step, by combine travelling data characteristic parameter, i.e., the corresponding local density of travelling data and density away from
From determining cluster numbers, the accuracy of cluster result can be effectively improved.
S103, according to preset rules and cluster numbers, the travelling data concentrated to travelling data clusters, and obtains cluster
Cluster.
In this step, cluster centre is determined according to preset rules and cluster numbers first, for example, N that can be forward by product
A travelling data is determined as cluster centre, wherein the quantity of cluster centre is consistent with cluster numbers under normal conditions, and N is big
In 0 integer.
Further, the travelling data at non-cluster center is concentrated to sort out travelling data, to obtain N number of cluster
Cluster.Specifically, by the travelling data at non-cluster center with it is bigger than its local density, and belonged to together apart from the smallest cluster centre
One kind clusters all travelling datas to realize, obtains N number of clustering cluster.
S104, according to the cluster Nuclear Data in clustering cluster, obtain cluster result.
Wherein, cluster result indicates the beginning or end of vehicle-mounted user.Specifically, if step S103 obtains the number of clustering cluster
Amount is one, then, according to the cluster Nuclear Data in unique clustering cluster, obtain cluster result；If step S103 obtains clustering cluster
Quantity be multiple, then, be respectively calculated for each clustering cluster, according to the cluster Nuclear Data in each clustering cluster, respectively
The corresponding cluster result of each clustering cluster is obtained, each cluster result indicates the beginning or end of vehicle-mounted user.Wherein, cluster core
Data are the biggish data of local density in clustering cluster, the core of corresponding clustering cluster, and opposite with cluster Nuclear Data for cluster
Dizzy data, cluster swoon data for the lesser data of local density in clustering cluster, correspond to the marginal portion of clustering cluster.
In a kind of possible implementation, is accurately identified to the travelling data in clustering cluster, determine cluster Nuclear Data
Later, cluster result can be obtained according to cluster Nuclear Data and preset computation rule, which indicates vehicle-mounted
The corresponding beginning or end in family.The preset computation rule can be realized by complicated mathematical algorithm, can also be user
Designed specific computation rule according to demand.
User origin and destination provided in this embodiment clustering method, firstly, obtaining the corresponding travelling data of vehicle-mounted user
Collection, travelling data collection includes at least one travelling data, and travelling data includes coordinate data；Then, according to each travelling data
Corresponding local density and density distance, determine cluster numbers；Then, according to preset rules and the cluster numbers, to driving
Travelling data in data set is clustered, and clustering cluster is obtained；Further, according to the cluster Nuclear Data in clustering cluster, cluster is obtained
As a result, wherein cluster result indicates the beginning or end of vehicle-mounted user.Method in the present embodiment, it is corresponding by travelling data
Local density and density distance, determine cluster numbers, improve the accuracy of cluster result；Further, by according to cluster
Cluster Nuclear Data in cluster obtains cluster result, influence of the outlier to cluster result is reduced, to improve cluster result
Accuracy.
Fig. 2 is the structural schematic diagram of user origin and destination clustering method embodiment two provided by the invention.Such as Fig. 2 institute
Show, the method for the present embodiment includes:
S201, the corresponding travelling data collection of vehicle-mounted user in preset time period is obtained.
Step S201 is similar with step S101 in embodiment illustrated in fig. 1 in the present embodiment, and details are not described herein again.
S202, distance is truncated according to default, obtains the corresponding local density of each travelling data and density distance.
In this step, using default truncation distance, calculates travelling data and concentrate the corresponding local density of each travelling data
And density distance.Specifically, the distance apart from travelling data is concentrated to be less than default other rows that distance is truncated travelling data
The quantity of car data is determined as the corresponding local density of travelling data.In practical applications, presetting truncation distance may be configured as
250 meters.
Further, according to the corresponding local density of all travelling datas, the corresponding density distance of each travelling data is obtained.
Further, on the basis of embodiment shown in Fig. 1, step S102, according to the corresponding local density of travelling data
And density distance, it determines cluster numbers, can be realized by S203:
S203, the corresponding local density of each travelling data and density distance are multiplied respectively, according to product
Slope variation trend, determines cluster numbers.
Specifically, the corresponding local density of each travelling data is multiplied with density distance, later carries out product
The travelling data that slope is greater than preset threshold is determined cluster according to the slope variation trend of the product after sequence by descending sort
Number.
In this step, by combine travelling data characteristic parameter, i.e., the corresponding local density of travelling data and density away from
From determining cluster numbers, the accuracy of cluster result can be effectively improved.
S204, according to preset rules and cluster numbers, the travelling data concentrated to travelling data clusters, and obtains cluster
Cluster.
Step S204 is similar with step S103 in embodiment illustrated in fig. 1 in the present embodiment, and details are not described herein again.
Further, it on the basis of embodiment shown in Fig. 1, step S104, according to the cluster Nuclear Data in clustering cluster, obtains
Cluster result can be realized by S205:
S205, by the average value of the cluster Nuclear Data in each clustering cluster, be determined as cluster result.
In the present embodiment, by using default truncation distance, obtaining the corresponding local density of each travelling data and close
Distance is spent, further, according to the slope variation trend of the corresponding local density of travelling data and the product of density distance, is determined poly-
Class number improves the accuracy of cluster result；Further, by obtaining cluster result, subtracting according to the cluster Nuclear Data in clustering cluster
Small influence of the outlier to cluster result, to improve the accuracy of cluster result.
Next, the specific implementation for how determining cluster Nuclear Data is described in detail by specific embodiment.
Fig. 3 is the flow diagram of user origin and destination clustering method embodiment three provided by the invention.It is shown in Fig. 1
On the basis of embodiment, step S104 can also include this before obtaining cluster result according to the cluster Nuclear Data in clustering cluster
Method shown in embodiment.
As shown in figure 3, the method for the present embodiment includes:
S301, according to the corresponding local density of clustering cluster middle rolling car data, obtain the corresponding local density's reference of clustering cluster
Value.
Local density's reference value is the standard reference value that can embody travelling data distribution situation, poly- for accurately distinguishing
Core data in class cluster.Specifically, local density's reference value can be true according to the corresponding local density of clustering cluster middle rolling car data
It is fixed, and each clustering cluster respectively corresponds local density's reference value, local density's reference value can be the same or different.Example
Such as, it can be determined according to the average value of the corresponding local density of clustering cluster middle rolling car data.
S302, it is referred to according to the corresponding local density of clustering cluster middle rolling car data and the corresponding local density of clustering cluster
Value, determines the cluster Nuclear Data in clustering cluster.
Specifically, by joining local density corresponding with the clustering cluster, the corresponding local density of clustering cluster middle rolling car data
It examines value to be compared, so that it is determined that the cluster Nuclear Data in clustering cluster, the cluster Nuclear Data is for calculating cluster result.
In the present embodiment, firstly, obtaining the corresponding office of clustering cluster according to the corresponding local density of clustering cluster middle rolling car data
Portion's density reference value, then, according to the corresponding local density of clustering cluster middle rolling car data and the corresponding local density of clustering cluster
Reference value determines the cluster Nuclear Data in clustering cluster.The present embodiment passes through the cluster Nuclear Data determined in clustering cluster, later can be according to poly-
Cluster Nuclear Data in class cluster obtains cluster result, influence of the outlier to cluster result is reduced, to improve cluster result
Accuracy.
Fig. 4 is the flow diagram of clustering method example IV in user origin and destination provided by the invention.Such as Fig. 4 institute
Show, it is step S301, corresponding according to clustering cluster middle rolling car data on the basis of the method for the present embodiment embodiment shown in Fig. 3
Local density obtains the corresponding local density's reference value of clustering cluster, can be realized by S401:
S401, by the average value of the corresponding local density of clustering cluster middle rolling car data, be determined as local density's reference value.
In this step, average value is an index for being able to reflect the central tendency of travelling data distribution.It is a kind of possible
In implementation, by the way that the corresponding local density of travelling datas all in clustering cluster is summed, and then divided by driving number
According to number, to obtain average value, which is local density's reference value.
It will, of course, be appreciated that average value can also be other kinds of average value, and such as: geometrical mean, weighting
Average value.
Optionally, S302, close according to the corresponding local density of clustering cluster middle rolling car data and the corresponding part of clustering cluster
Reference value is spent, the cluster Nuclear Data in clustering cluster is determined, can be realized by S402:
If S402, the corresponding local density of clustering cluster middle rolling car data are greater than the corresponding local density's reference value of clustering cluster,
Then determine that travelling data is cluster Nuclear Data.
Specifically, if the corresponding local density of clustering cluster middle rolling car data is greater than the corresponding part of clustering cluster middle rolling car data
The average value of density, it is determined that the travelling data is cluster Nuclear Data.
It will, of course, be appreciated that if the corresponding local density of clustering cluster middle rolling car data is less than or equal to clustering cluster pair
The local density's reference value answered, it is determined that travelling data is the dizzy data of cluster.Specifically, if the corresponding office of clustering cluster middle rolling car data
Portion's density is less than or equal to the average value of the corresponding local density of clustering cluster middle rolling car data, it is determined that the travelling data is dizzy for cluster
Data.
In the present embodiment, by the way that it is close to be determined as part by the average value of the corresponding local density of clustering cluster middle rolling car data
Reference value is spent, further, if the corresponding local density of clustering cluster middle rolling car data is greater than the corresponding local density's reference of clustering cluster
Value, it is determined that travelling data is cluster Nuclear Data.The cluster Nuclear Data in clustering cluster is determined by using the method in the present embodiment, it
Afterwards cluster result can be obtained, influence of the outlier to cluster result is reduced, to improve according to the cluster Nuclear Data in clustering cluster
The accuracy of cluster result.
Fig. 5 is the structural schematic diagram of user origin and destination cluster analyzing device embodiment one provided by the invention.Such as Fig. 3 institute
Show, the device 50 of the present embodiment includes: the first acquisition module 51, the first determining module 52, cluster module 53 and computing module
54。
Specifically, first module 51 is obtained, for obtaining the corresponding travelling data collection of vehicle-mounted user in preset time period, row
Car data collection includes at least one travelling data, and travelling data includes coordinate data.
First determining module 52, for determining poly- according to the corresponding local density of each travelling data and density distance
Class number.
Cluster module 53 obtains clustering cluster for being clustered to travelling data according to preset rules and cluster numbers.
Computing module 54, for obtaining cluster result according to the cluster Nuclear Data in each clustering cluster, cluster result indicates vehicle
Carry the beginning or end of user.
The device of the present embodiment can be used for executing the technical solution of embodiment of the method shown in Fig. 1, realization principle and skill
Art effect is similar, and details are not described herein again.
Fig. 6 is the structural schematic diagram of user origin and destination cluster analyzing device embodiment two provided by the invention.Such as Fig. 6 institute
Show, on the basis of the embodiment shown in Fig. 5 of device 60 of the present embodiment, further includes: second obtains module 55.
Wherein, second module 55 is obtained, for it is close to obtain the corresponding part of each travelling data according to default truncation distance
Degree and density distance.
Optionally, the first determining module 52, for dividing the corresponding local density of each travelling data and density distance
It is not multiplied, according to the slope variation trend of product, determines cluster numbers.
Optionally, computing module 54, for being determined as cluster knot for the average value of the cluster Nuclear Data in each clustering cluster
Fruit, wherein the cluster result indicates the beginning or end of vehicle-mounted user.
The device of the present embodiment can be used for executing the technical solution of embodiment of the method shown in Fig. 2, realization principle and skill
Art effect is similar, and details are not described herein again.
Fig. 7 is the structural schematic diagram of user origin and destination cluster analyzing device embodiment three provided by the invention.Such as Fig. 7 institute
Show, on the basis of the embodiment shown in Fig. 6 of device 70 of the present embodiment, further includes: third determining module 56.
Wherein, third determining module 56 is used to determine the cluster Nuclear Data in clustering cluster in the following manner:
According to the corresponding local density of clustering cluster middle rolling car data, the corresponding local density's reference value of clustering cluster is obtained；Root
According to the corresponding local density of clustering cluster middle rolling car data and the corresponding local density's reference value of clustering cluster, determine in clustering cluster
Cluster Nuclear Data.
In some embodiments, third determining module 56 includes: that third determines that submodule 561 and cluster Nuclear Data determine submodule
Block 562.
Wherein, third determines submodule 561, for obtaining poly- according to the corresponding local density of clustering cluster middle rolling car data
The corresponding local density's reference value of class cluster.
Cluster Nuclear Data determines submodule 562, for according to the corresponding local density of clustering cluster middle rolling car data and cluster
The corresponding local density's reference value of cluster, determines the cluster Nuclear Data in clustering cluster.
Optionally, in some embodiments, third determines submodule 561, is specifically used for clustering cluster middle rolling car data pair
The average value for the local density answered is determined as local density's reference value.
Cluster Nuclear Data determines submodule 562, gathers if being specifically used for the corresponding local density of clustering cluster middle rolling car data and being greater than
The corresponding local density's reference value of class cluster, it is determined that travelling data is cluster Nuclear Data；If the corresponding office of clustering cluster middle rolling car data
Portion's density is less than or equal to the corresponding local density's reference value of clustering cluster, it is determined that travelling data is the dizzy data of cluster.
The device of the present embodiment can be used for executing the technical solution of embodiment of the method shown in Fig. 3, Fig. 4, realization principle
Similar with technical effect, details are not described herein again.
Fig. 8 is the structural schematic diagram of user origin and destination cluster analyzing device example IV provided by the invention.Such as Fig. 8 institute
Show, the device 80 of the present embodiment includes: memory 81, processor 82.
Wherein, memory 81 can be independent physical unit, can be connect by bus 83 with processor 82.Memory
81, processor 82 also can integrate together, pass through hardware realization etc..
Memory 81 is used to store the program for realizing above method embodiment, the calling of processor 82 program, more than execution
The operation of embodiment of the method.
Optionally, when passing through software realization some or all of in the method for above-described embodiment, above-mentioned user origin and destination
Cluster analyzing device 80 can also only include processor.Memory for storing program is located at user origin and destination clustering dress
It sets except 80, processor is connect by circuit/electric wire with memory, for reading and executing the program stored in memory.
Processor 82 can be central processing unit (Central Processing Unit, referred to as: CPU), network processing unit
The combination of (Network Processor, referred to as: NP) or CPU and NP.
Processor 82 can further include hardware chip.Above-mentioned hardware chip can be specific integrated circuit
(Application-Specific Integrated Circuit, referred to as: ASIC), programmable logic device
(Programmable Logic Device, referred to as: PLD) or combinations thereof.Above-mentioned PLD can be Complex Programmable Logic Devices
(Complex Programmable Logic Device, referred to as: CPLD), field programmable gate array (Field-
Programmable Gate Array, referred to as: FPGA), Universal Array Logic (Generic Array Logic, referred to as: GAL)
Or any combination thereof.
Memory 81 may include volatile memory (Volatile Memory), such as random access memory
(Random-Access Memory, referred to as: RAM)；Memory also may include nonvolatile memory (Non-volatile
Memory), such as flash memory (Flash Memory), hard disk (Hard Disk Drive, referred to as: HDD) or solid state hard disk
(Solid-state Drive, referred to as: SSD)；Memory can also include the combination of the memory of mentioned kind.
The device of the present embodiment can be used for executing the technical solution of embodiment of the method shown in Fig. 1 to Fig. 4, realize former
Reason is similar with technical effect, and details are not described herein again.
In addition, the present invention also provides a kind of program product, for example, computer storage medium, comprising: program, program is in quilt
For executing above method when processor executes.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above-mentioned each method embodiment can lead to
The relevant hardware of program instruction is crossed to complete.Program above-mentioned can be stored in a computer readable storage medium.The journey
When being executed, execution includes the steps that above-mentioned each method embodiment to sequence；And storage medium above-mentioned include: ROM, RAM, magnetic disk or
The various media that can store program code such as person's CD.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (10)
1. a kind of clustering method of user origin and destination characterized by comprising
The corresponding travelling data collection of vehicle-mounted user in preset time period is obtained, the travelling data collection includes at least one driving number
According to the travelling data includes coordinate data；
According to the corresponding local density of each travelling data and density distance, cluster numbers are determined；
According to preset rules and the cluster numbers, the travelling data concentrated to the travelling data is clustered, and obtains cluster
Cluster；
According to the cluster Nuclear Data in the clustering cluster, cluster result is obtained, the cluster result indicates rising for the vehicle-mounted user
Point or terminal.
2. the method according to claim 1, wherein described close according to the corresponding part of each travelling data
Degree, density distance and preset threshold, before determining cluster numbers, further includes:
According to default truncation distance, the corresponding local density of each travelling data and density distance are obtained.
3. the method according to claim 1, wherein described close according to the corresponding part of each travelling data
Degree and density distance, determine cluster numbers, comprising:
The corresponding local density of each travelling data and density distance are multiplied respectively, according to the oblique of the product
Rate variation tendency, determines cluster numbers.
4. the method according to claim 1, wherein the cluster Nuclear Data according in the clustering cluster, obtains
Before cluster result, further includes:
The cluster Nuclear Data in the clustering cluster is determined in the following manner:
According to the corresponding local density of the clustering cluster middle rolling car data, the corresponding local density's reference of the clustering cluster is obtained
Value；
According to the corresponding local density of the clustering cluster middle rolling car data and the corresponding local density's reference value of the clustering cluster,
Determine the cluster Nuclear Data in the clustering cluster.
5. according to the method described in claim 4, it is characterized in that, described according to the corresponding office of the clustering cluster middle rolling car data
Portion's density obtains the corresponding local density's reference value of the clustering cluster, comprising:
By the average value of the corresponding local density of the clustering cluster middle rolling car data, it is determined as local density's reference value.
6. according to the method described in claim 4, it is characterized in that, described according to the corresponding office of the clustering cluster middle rolling car data
Portion's density and the corresponding local density's reference value of the clustering cluster, determine the cluster Nuclear Data in the clustering cluster, comprising:
If the corresponding local density of the clustering cluster middle rolling car data is greater than the corresponding local density's reference value of the clustering cluster,
Determine that the travelling data is cluster Nuclear Data；
If the corresponding local density of the clustering cluster middle rolling car data is less than or equal to the corresponding local density's ginseng of the clustering cluster
Examine value, it is determined that the travelling data is the dizzy data of cluster.
7. method according to claim 1-6, which is characterized in that described according to cluster nucleus number in each clustering cluster
According to acquisition cluster result, comprising:
By the average value of the cluster Nuclear Data in each clustering cluster, it is determined as cluster result.
8. a kind of user origin and destination cluster analyzing device characterized by comprising
First obtains module, for obtaining the corresponding travelling data collection of vehicle-mounted user in preset time period, the travelling data collection
Including at least one travelling data, the travelling data includes coordinate data；
First determining module, for determining cluster according to the corresponding local density of each travelling data and density distance
Number；
Cluster module obtains cluster for being clustered to the travelling data according to preset rules and the cluster numbers
Cluster；
Computing module, for according to the cluster Nuclear Data in each clustering cluster, obtaining cluster result, described in the cluster result expression
The beginning or end of vehicle-mounted user.
9. a kind of user origin and destination cluster analyzing device characterized by comprising memory and processor；
The memory stores program instruction；
Described program instruction requires the described in any item methods of 1-7 when being executed by the processor, with perform claim.
10. a kind of storage medium characterized by comprising program；
Described program requires the described in any item methods of 1-7 when being executed by processor, with perform claim.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811465250.7A CN109508750A (en) | 2018-12-03 | 2018-12-03 | The clustering method of user origin and destination, device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811465250.7A CN109508750A (en) | 2018-12-03 | 2018-12-03 | The clustering method of user origin and destination, device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109508750A true CN109508750A (en) | 2019-03-22 |
Family
ID=65749960
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811465250.7A Pending CN109508750A (en) | 2018-12-03 | 2018-12-03 | The clustering method of user origin and destination, device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109508750A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101887645A (en) * | 2010-07-09 | 2010-11-17 | 天津职业技术师范大学 | Wiring optimization method of wired vehicle detector in indoor parking lot |
| US8374828B1 (en) * | 2007-12-24 | 2013-02-12 | The University Of North Carolina At Charlotte | Computer implemented system for protein and drug target design utilizing quantified stability and flexibility relationships to control function |
| CN103337189A (en) * | 2013-06-08 | 2013-10-02 | 北京航空航天大学 | Dynamic road segment division based vehicle route guidance method |
| CN104517052A (en) * | 2014-12-09 | 2015-04-15 | 中国科学院深圳先进技术研究院 | Invasion detection method and device |
| CN105206048A (en) * | 2015-11-05 | 2015-12-30 | 北京航空航天大学 | Urban resident traffic transfer mode discovery system and method based on urban traffic OD data |
| CN106557829A (en) * | 2015-09-30 | 2017-04-05 | 百度在线网络技术（北京）有限公司 | Method and apparatus with demand and transport power mismatch region are obtained in car business |
| CN106600960A (en) * | 2016-12-22 | 2017-04-26 | 西南交通大学 | Traffic travel origin and destination identification method based on space-time clustering analysis algorithm |
| CN108595585A (en) * | 2018-04-18 | 2018-09-28 | 平安科技（深圳）有限公司 | Sample data sorting technique, model training method, electronic equipment and storage medium |
-
2018
- 2018-12-03 CN CN201811465250.7A patent/CN109508750A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8374828B1 (en) * | 2007-12-24 | 2013-02-12 | The University Of North Carolina At Charlotte | Computer implemented system for protein and drug target design utilizing quantified stability and flexibility relationships to control function |
| CN101887645A (en) * | 2010-07-09 | 2010-11-17 | 天津职业技术师范大学 | Wiring optimization method of wired vehicle detector in indoor parking lot |
| CN103337189A (en) * | 2013-06-08 | 2013-10-02 | 北京航空航天大学 | Dynamic road segment division based vehicle route guidance method |
| CN104517052A (en) * | 2014-12-09 | 2015-04-15 | 中国科学院深圳先进技术研究院 | Invasion detection method and device |
| CN106557829A (en) * | 2015-09-30 | 2017-04-05 | 百度在线网络技术（北京）有限公司 | Method and apparatus with demand and transport power mismatch region are obtained in car business |
| CN105206048A (en) * | 2015-11-05 | 2015-12-30 | 北京航空航天大学 | Urban resident traffic transfer mode discovery system and method based on urban traffic OD data |
| CN106600960A (en) * | 2016-12-22 | 2017-04-26 | 西南交通大学 | Traffic travel origin and destination identification method based on space-time clustering analysis algorithm |
| CN108595585A (en) * | 2018-04-18 | 2018-09-28 | 平安科技（深圳）有限公司 | Sample data sorting technique, model training method, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI746814B (en) | Computer readable medium, car insurance risk prediction device and server | |
| CN108833458B (en) | Application recommendation method, device, medium and equipment | |
| CN113312361B (en) | Track query method, device, equipment, storage medium and computer program product | |
| US11456932B2 (en) | System capacity heatmap | |
| US10977082B2 (en) | Resource allocation optimization support system and resource allocation optimization support method | |
| CN107944697B (en) | Supply and demand relationship-based heat map calculation method and system, server and medium | |
| US20220405806A1 (en) | Utilizing machine-learning models to create target audiences with customized auto-tunable reach and accuracy | |
| CN108182240A (en) | The new gaining rate prediction model training of point of interest and Forecasting Methodology, device and storage medium | |
| CN110895706B (en) | Method and device for acquiring target cluster number and computer system | |
| CN109978619A (en) | Method, system, equipment and the medium of air ticket pricing Policy Filtering | |
| US20090292715A1 (en) | System and Method for Determining Overall Utilization | |
| CN109508750A (en) | The clustering method of user origin and destination, device and storage medium | |
| CN110275895A (en) | It is a kind of to lack the filling equipment of traffic data, device and method | |
| CN110796115A (en) | Image detection method and device, electronic equipment and readable storage medium | |
| CN108133234B (en) | Sparse subset selection algorithm-based community detection method, device and equipment | |
| CN111523043B (en) | Resource information display and management method and device | |
| CN108280224A (en) | Ten thousand grades of dimension data generation methods, device, equipment and storage medium | |
| WO2022245295A2 (en) | System and method for predicting delivery time for batch orders | |
| CN115269600A (en) | Airplane data management method and device, electronic equipment and storage medium | |
| CN111160385B (en) | Method, device, equipment and storage medium for aggregating mass location points | |
| CN114021031A (en) | Financial product information pushing method and device | |
| CN108023734B (en) | Bandwidth acquisition method and device, data processing method and device and server | |
| CN110231969A (en) | User interface management method, apparatus, computer installation and readable storage medium storing program for executing | |
| CN112905782B (en) | Volume assembling method, device, equipment and storage medium | |
| CN110298679A (en) | A kind of method and apparatus calculating the distance between sample data |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190322 |