CN104252527B - A kind of method and apparatus of the resident information of definite mobile subscriber - Google Patents
A kind of method and apparatus of the resident information of definite mobile subscriber Download PDFInfo
- Publication number
- CN104252527B CN104252527B CN201410443562.3A CN201410443562A CN104252527B CN 104252527 B CN104252527 B CN 104252527B CN 201410443562 A CN201410443562 A CN 201410443562A CN 104252527 B CN104252527 B CN 104252527B
- Authority
- CN
- China
- Prior art keywords
- hypothesis
- information
- cluster
- hypothesis quantity
- resident
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention proposes a kind of method for the resident information that mobile subscriber is determined in computer equipment, wherein, this method comprises the following steps:A. multiple event information of the mobile subscriber are obtained, wherein, the event information is used to indicate that the locus of the mobile subscriber and mobile subscriber are located at corresponding time point information during the locus;B. cluster analysis is carried out to the multiple event information based on clustering algorithm, to determine multiple resident information of the mobile subscriber.Scheme according to the present invention, can determine multiple resident information of mobile subscriber according to the event information of mobile subscriber, and determine the type of such resident information.
Description
Technical field
The present invention relates to field of computer technology, more particularly to a kind of determine that mobile subscriber's is resident in computer equipment
The method and apparatus of point information.
Background technology
In the prior art, the current location of mobile subscriber usually can be only obtained, such as by mobile subscriber's active
Report or triggering mobile subscriber such as report at the mode, obtain the current location of mobile subscriber.And then carried out based on the current location
The operation such as positioning.
The content of the invention
The object of the present invention is to provide it is a kind of in computer equipment determine mobile subscriber resident information method and
Device.
According to an aspect of the present invention, there is provided a kind of resident information that mobile subscriber is determined in computer equipment
Method, wherein, this method comprises the following steps:
A. multiple event information of the mobile subscriber are obtained, wherein, the event information is used to indicate the shifting
Employ the locus at family and mobile subscriber is located at corresponding time point information during the locus;
B. cluster analysis is carried out to the multiple event information based on clustering algorithm, to determine that the mobile subscriber's is more
A resident information.
According to another aspect of the present invention, a kind of resident point that mobile subscriber is determined in computer equipment is additionally provided
The device of information, wherein, which includes following device:
Device for the multiple event information for obtaining the mobile subscriber, wherein, the event information is used to refer to
Show that the locus of the mobile subscriber and mobile subscriber are located at corresponding time point information during the locus;
For carrying out cluster analysis to the multiple event information based on clustering algorithm, to determine the mobile subscriber's
The device of multiple resident information.
Compared with prior art, the present invention has the following advantages:1) can be by being carried out to the event information of mobile subscriber
Cluster analysis, to determine multiple resident points of mobile subscriber, so as to more accurately understand the scope of activities of mobile subscriber and
Rule of life;2) type of each resident point of mobile subscriber according to multiple resident information of mobile subscriber, can be determined, and
The probability that user occurs in certain resident point region is predicted to a certain extent.
Brief description of the drawings
By reading the detailed description made to non-limiting example made with reference to the following drawings, of the invention is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is the method for the resident information that mobile subscriber is determined in computer equipment of one embodiment of the invention
Flow diagram;
Fig. 2 is the method for the resident information that mobile subscriber is determined in computer equipment of another embodiment of the present invention
Flow diagram;
Fig. 3 is the device of the resident information that mobile subscriber is determined in computer equipment of one embodiment of the invention
Structure diagram;
Fig. 4 is the device of the resident information that mobile subscriber is determined in computer equipment of another embodiment of the present invention
Structure diagram.
The same or similar reference numeral represents the same or similar component in attached drawing.
Embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings.
Fig. 1 is the method for the resident information that mobile subscriber is determined in computer equipment of one embodiment of the invention
Flow diagram.
Wherein, the method for the present embodiment is mainly realized by computer equipment;The computer equipment is set including network
Standby and user equipment.The network equipment includes but not limited to the service of single network server, multiple webservers composition
Device group or the cloud being made of a large amount of computers or the webserver based on cloud computing (Cloud Computing), wherein, cloud meter
It is one kind of Distributed Calculation, a super virtual computer being made of the computer collection of a group loose couplings;The net
Network residing for network equipment includes but not limited to internet, wide area network, Metropolitan Area Network (MAN), LAN, VPN network etc..The user sets
Standby including but not limited to PC machine, tablet computer, smart mobile phone, PDA, IPTV etc..
It should be noted that the computer equipment is only for example, other calculating that are existing or being likely to occur from now on are set
It is standby to be such as applicable to the present invention, it should also be included within the scope of the present invention, and be incorporated herein by reference.
Step S1 and step S2 is included according to the method for the present embodiment.
In step sl, computer equipment obtains multiple event information of mobile subscriber.
Wherein, the event information is used to indicate that the locus of the mobile subscriber and mobile subscriber are located at the sky
Between position when corresponding time point information.Preferably, the event information can have many forms, including but unlimited
In:Point in one vector measured, hyperspace etc.;It is highly preferred that the event information is space-time vector.
For example, the event information of mobile subscriber is space-time vector α=(a, b, c, d), wherein, (a, b, c)
For the coordinate of the locus of mobile subscriber, d is located at corresponding time point information during the locus for mobile subscriber.
Specifically, computer equipment can obtain multiple event information of mobile subscriber in several ways.For example, calculate
Machine equipment receives from other computer equipments, mobile subscriber multiple event information;In another example mobile subscriber timing to
Computer equipment reports its event information, then within a period of time, computer equipment reception mobile subscriber is reported multiple
Event information etc..
It should be noted that the above-mentioned examples are merely illustrative of the technical solutions of the present invention, rather than the limit to the present invention
System, it should be appreciated by those skilled in the art that the implementation of any multiple event information for obtaining mobile subscriber, should all include
Within the scope of the invention.
In step s 2, computer equipment is based on clustering algorithm and carries out cluster analysis to the multiple event information, comes
Determine multiple resident information of the mobile subscriber.
Wherein, the clustering algorithm includes any algorithm that can be used for carrying out cluster analysis, for example, density clustering
Algorithm, EM algorithms etc..Preferably, the clustering algorithm needs to set the quantity of cluster centre;It is highly preferred that the clustering algorithm
For density-based algorithms.
Wherein, the resident information includes any information for being used to indicate the resident point of mobile subscriber;Preferably, it is described
Resident point information includes the relevant any information of resident point with mobile subscriber;Preferably, directly cluster analysis can be obtained
Class in cluster result is as resident point information.It is highly preferred that it can be come true by carrying out statistical analysis to the class in cluster result
Such fixed corresponding resident information, wherein, the resident information includes position attribution information and time attribute information, described
Position attribution information is used for locus or the position range for indicating the resident point, and the time attribute information is used to indicate to move
Multiple time point informations or time range when user is located at the resident point.
Specifically, computer equipment is based on clustering algorithm and carries out cluster analysis to the multiple event information, to obtain
Include the cluster result of multiple classes, and multiple resident information of mobile subscriber are determined according to the plurality of class.
For example, the quantity set of the cluster centre of clustering algorithm is predetermined quantity by computer equipment, such as 4;Computer is set
It is standby that 4 event information are selected from the multiple event information as cluster centre, in multiple event information
Each event information, computer equipment calculate the event information respectively the distance between with 4 cluster centres, and should
Event information categorization is to the cluster centre corresponding to minimum range;Afterwards, 4 in cluster result of computer equipment
Class, to determine the 4 of mobile subscriber resident point information.
As a kind of preferred solution of the present embodiment, the clustering algorithm needs to set the quantity of cluster centre.
Wherein, clustering algorithm of the computer equipment based on needs setting cluster centre, believes the multiple event
Breath carries out cluster analysis, and determines that the mode of multiple resident information of the mobile subscriber includes but not limited to:
1) quantity of the cluster centre of clustering algorithm has predefined, then computer equipment is directly based upon fixed cluster
The quantity at center, operation clustering algorithm determine the mobile subscriber to carry out cluster analysis to the multiple event information
Multiple resident information.
2) quantity of the cluster centre of clustering algorithm does not determine, then in the case, computer equipment needs first to determine
The quantity of one suitable cluster centre.
Specifically, in this implementation, computer equipment can determine a suitable hypothesis number from multiple hypothesis quantity
Amount, the quantity as cluster centre.The step S2 further comprises step S21 and step S22.
In the step s 21, will for each hypothesis quantity in all or part of multiple hypothesis quantity, computer equipment
The quantity set of the cluster centre of the clustering algorithm is the hypothesis quantity, and based on the clustering algorithm to the multiple space-time
Point information carries out cluster analysis, obtains cluster result corresponding with the hypothesis quantity, and correspond to respectively according to multiple hypothesis quantity
Multiple cluster results, select one hypothesis quantity.
Preferably, computer equipment is based at least one of following, corresponding multiple poly- according to multiple hypothesis quantity
Class result selects a hypothesis quantity:
The quantity for the event information that class in the corresponding cluster result of-hypothesis quantity includes.
Preferably, the quantity of the event information included in class is more, then usual cluster result is better.
The dispersion of class in the corresponding cluster result of-hypothesis quantity.
Preferably, the dispersion of class is lower, then usual cluster result is better.
Wherein, the dispersion is used for the dense degree for indicating class.Wherein, computer equipment can be using various ways come really
The fixed dispersion, e.g., all event information of the computer equipment in class determine the average of class, and calculate each space-time
Very poor, mean difference or standard deviation between point information and the average etc. represent such dispersion.
It should be noted that the step S21 can be realized using various ways.For example, the implementation bag of step S21
Include but be not limited to:
A) in this implementation, the step S21 further comprises step S2111, step S2112 and step S2113.
In step S2111, for a hypothesis for not determining its corresponding cluster result in the multiple hypothesis quantity
Quantity, the quantity set of the cluster centre of the clustering algorithm is the hypothesis quantity by computer equipment, and is based on the cluster
Algorithm carries out cluster analysis to the multiple event information, obtains cluster result corresponding with the hypothesis quantity.
In step S2112, when the corresponding cluster result of hypothesis quantity meets the first predetermined condition, computer equipment
Using the hypothesis quantity as the selected hypothesis quantity.
Wherein, first predetermined condition includes any predetermined condition for being used to select to assume quantity.Preferably, it is described
First predetermined condition includes but not limited to:
The quantity a predetermined level is exceeded threshold value for the event information that class in the-cluster result includes.
The dispersion of class in the-cluster result is less than predetermined dispersion threshold value.
For example, predetermined quantity threshold value is 100, it is assumed that the corresponding cluster result of quantity includes 4 classes, when in 4 classes
The quantity of null point information is respectively:120、110、108、150.Then in step S2112, computer equipment determines to assume quantity pair
The equal a predetermined level is exceeded threshold value of quantity of event information in each class for the cluster result answered, then computer equipment determine this
Cluster result meets the first predetermined condition, and using the hypothesis quantity as selected hypothesis quantity.
In step S2113, when the corresponding cluster result of hypothesis quantity does not meet first predetermined condition, calculate
Machine equipment repeating said steps S2111.
Specifically, when the corresponding cluster result of hypothesis quantity does not meet the first predetermined condition, computer equipment repeats
Step S2111, to obtain the cluster result corresponding to the hypothesis quantity for not determining its corresponding cluster result;And so on, directly
When assuming that the corresponding cluster result of quantity meets the first predetermined condition to one, using the hypothesis quantity as selected hypothesis number
Amount, and stop operation.
For example, multiple assume that quantity includes all natural numbers from 2 to 1000.When performing step S2111 for the first time, calculate
The hypothesis quantity of machine equipment selection is 2, and in the case where being 2 by the quantity set of cluster centre, based on clustering algorithm to institute
State multiple event information and carry out cluster analysis, obtain cluster result corresponding with assuming quantity " 2 ";Then, computer equipment
Judge that " 2 " corresponding cluster result does not meet first predetermined condition, in step S2113, computer equipment repeat step
S2111, selection do not determine the hypothesis quantity " 4 " of its corresponding cluster result, and determine its cluster result;Then, computer is set
It is standby to judge that " 4 " corresponding cluster result does not meet first predetermined condition, continue to execute step S2113;And so on, until
Computer equipment obtains the hypothesis quantity " 5 " for meeting the first predetermined condition, and performs step S2112, by " 5 " as selected
Assuming that quantity.
In this implementation, computer equipment only needs to obtain the hypothesis quantity for meeting the first predetermined condition, you can
Subsequent operation is performed based on the hypothesis quantity, without traveling through and obtaining the cluster result of all hypothesis quantity.
B) in this implementation, the multiple to assume increased number or successively decrease, the step S21 further comprises step
S2121, step S2122, step S2123 and step S2124.
In step S2121, one in the multiple hypothesis quantity is assumed quantity as current false by computer equipment
It is the current hypothesis quantity by the quantity set of the cluster centre of the clustering algorithm, and be based on the clustering algorithm if quantity
Cluster analysis is carried out to the multiple event information, obtains and current assumes the corresponding cluster result of quantity with this.
For example, multiple hypothesis quantity are included from 2 to 1000 incremental multiple natural numbers.In step S2121, computer is set
It is standby that " 2 " are assumed into quantity as current, and be " 2 " by the quantity set of the cluster centre of clustering algorithm, and it is based on the cluster
Algorithm carries out cluster analysis to the multiple event information, obtains cluster result corresponding with " 2 ".
In step S2122, the quantity set of the cluster centre of the clustering algorithm is the current vacation by computer equipment
If next hypothesis quantity of quantity, and cluster analysis is carried out to the multiple event information based on the clustering algorithm, obtain
With the corresponding cluster result of next hypothesis quantity.
For example, next hypothesis quantity of the computer equipment by the quantity set of the cluster centre of clustering algorithm for " 2 "
" 3 ", and cluster analysis is carried out to the multiple event information based on clustering algorithm, obtain corresponding with next hypothesis quantity
Cluster result.
In step S2123, when the corresponding cluster result of next hypothesis quantity is worse than the current hypothesis quantity pair
During the cluster result answered, computer equipment is using the current hypothesis quantity as the selected hypothesis quantity.
Preferably, the quantity of the space time information point that can be included according to the dispersion and/or class of class in cluster result determines
Whether the corresponding cluster result of next hypothesis quantity is worse than the corresponding cluster result of the current hypothesis quantity.
For example, the variance E in the corresponding cluster result of next hypothesis quantity between class can be calculated1, and current hypothesis
Variance E in the corresponding cluster result of quantity between class2, and compare E1And E2, work as E1More than E2, computer equipment can determine that next
The corresponding cluster result of a hypothesis quantity is worse than the corresponding cluster result of the current hypothesis quantity;Work as E1Less than E2, computer equipment
It can determine that the corresponding cluster result of next hypothesis quantity cluster result corresponding better than the current hypothesis quantity.
In step S2124, when the corresponding cluster result of next hypothesis quantity is better than the current hypothesis quantity pair
During the cluster result answered, computer equipment is using next hypothesis quantity as the current hypothesis quantity, repeating said steps
S2122。
For example, currently assume that quantity is " 2 ", and next hypothesis quantity of " 2 " is " 3 ", and " 3 " corresponding cluster result
Cluster result corresponding better than " 2 ", then " 3 " are used as by computer equipment currently assumes quantity, and repeat step S2122, obtains
The cluster result of " 4 ";Then, if " 4 " corresponding cluster result is better than " 3 " corresponding cluster result, computer equipment is by " 4 "
Assume quantity as current, continue repeat step S2122;And so on, until the corresponding cluster result of next hypothesis quantity
When being worse than the current hypothesis corresponding cluster result of quantity, in step S2123, computer equipment makees the current hypothesis quantity
For the selected hypothesis quantity.
Due to when it is multiple hypothesis quantity show increasing or decreasing relation when, an optimal corresponding cluster of hypothesis quantity
As a result, can be better than its corresponding cluster result of two neighboring hypothesis quantity, therefore, in this implementation, computer equipment can obtain
Obtain hypothesis quantity most preferably.Also, since hypothesis quantity execution subsequent operation can be based on after obtaining optimal hypothesis quantity, and
It need not continue to obtain other cluster results for assuming quantity, therefore under normal conditions, this implementation is without traveling through and obtaining all
Assuming that the cluster result of quantity.
C) the step S21 further comprises step S2131 and step S2132.
In step 2131, for each hypothesis quantity in multiple hypothesis quantity, computer equipment calculates the cluster
The quantity set of the cluster centre of method is the hypothesis quantity, and the multiple event information is carried out based on the clustering algorithm
Cluster analysis, obtains cluster result corresponding with the hypothesis quantity.
For example, there are 4 hypothesis quantity:2、3、4、5.Computer equipment is obtained in cluster respectively based on the clustering algorithm
It is poly- when cluster result, the quantity of cluster centre when the quantity of cluster result, cluster centre when the quantity of the heart is 2 is 3 are 4
Cluster result when class result and the quantity of cluster centre are 5.
In step 2132, computer equipment is according to the corresponding multiple cluster results of the multiple hypothesis quantity, choosing
Select a hypothesis quantity.
Wherein, computer equipment selects a hypothesis quantity according to the corresponding multiple cluster results of multiple hypothesis quantity
Implementation be described in detail above, details are not described herein.
It should be noted that multiple forms assumed quantity and can behave as set, are such as set [2,3,4 ..., 1000],
Then computer equipment can be read directly from the set assumes quantity.Alternatively, multiple forms assumed quantity and can behave as formula,
Such as k=K+n △;Wherein, k represents to assume quantity, and K is radix (usual K can use 2), △=1, n=0,1,2 ..., 998;Then count
The hypothesis quantity of its needs can be calculated by the formula by calculating machine equipment.
It should be noted that the above-mentioned examples are merely illustrative of the technical solutions of the present invention, rather than the limit to the present invention
System, it should be appreciated by those skilled in the art that each hypothesis quantity in any all or part for multiple hypothesis quantity, will
The quantity set of the cluster centre of the clustering algorithm is the hypothesis quantity, and based on the clustering algorithm to the multiple space-time
Point information carries out cluster analysis, obtains cluster result corresponding with the hypothesis quantity, and correspond to respectively according to multiple hypothesis quantity
Multiple cluster results, select one hypothesis quantity implementation, should be included in the scope of the present invention.
In step S22, computer equipment determines the movement according to the corresponding cluster result of selected hypothesis quantity
Multiple resident information of user.
Wherein, computer equipment can use various ways to be determined according to the corresponding cluster result of selected hypothesis quantity
Multiple resident information of the mobile subscriber.
For example, computer equipment can be directly using multiple classes of cluster result as mobile subscriber multiple resident information.
In another example for each class in cluster result, computer equipment can be by such progress statistical analysis, such as dividing
The locus of other all event information in such and time point information count, to determine that such is corresponding resident
Point information.
It should be noted that the above-mentioned examples are merely illustrative of the technical solutions of the present invention, rather than the limit to the present invention
System, it should be appreciated by those skilled in the art that it is any according to the corresponding cluster result of selected hypothesis quantity, determine the movement
The implementation of multiple resident information of user, should be included in the scope of the present invention.
, can be by carrying out cluster analysis to the event information of mobile subscriber, to determine shifting according to the scheme of the present embodiment
Multiple resident points at family are employed, so as to more accurately understand the scope of activities of mobile subscriber and rule of life.
Fig. 2 is the method for the resident information that mobile subscriber is determined in computer equipment of another embodiment of the present invention
Flow diagram.The method of the present embodiment mainly realized by computer equipment, wherein, with reference to right in embodiment illustrated in fig. 1
Any explanation that computer equipment is done, is incorporated herein by reference.
Step S1, step S2 and step S3 are included according to the method for the present embodiment.Wherein, the step S1 and step S2 be
It is described in detail with reference to the embodiment shown in FIG. 1, details are not described herein.
In step s3, computer equipment is determined in the multiple resident information according to the multiple resident information
Each resident point information type
Wherein, the type of the resident information is used for the property for indicating the resident point of mobile subscriber, such as family, dining room, joy
Happy place, place of working etc..
Specifically, for each resident information, computer equipment the resident point is determined by analyzing the resident point information
The type of information.
For example, according to resident information and map, computer equipment determines the position model corresponding to the resident point information
It is trapped among in a residential block, then computer equipment determines that the type of the resident point information is family.
Preferably, the step S3 further comprises to the step S31 each performed in the multiple resident information
With step S32.
In step S31, computer equipment obtains the position attribution information and time attribute information of the resident point information.
Wherein, computer equipment can use various ways to obtain the position attribution information and time attribute of residing point information
Information.
For example, when the resident information is the class in cluster result, computer equipment is to all space-times in such
Point information carries out statistical analysis, and the position attribution of the resident point information is obtained come the locus in all event information
Information, and the time point information in all event information obtains the time attribute information of the resident point information.
In another example when the resident information be by cluster result class carry out statistical analysis come obtain when,
Computer equipment can directly extract the position attribution information and time attribute information of the resident point information from the resident point information.
It should be noted that the above-mentioned examples are merely illustrative of the technical solutions of the present invention, rather than the limit to the present invention
System, it should be appreciated by those skilled in the art that any acquisition resident the position attribution information for putting information and time attribute information
Implementation, should be included in the scope of the present invention.
In step s 32, computer equipment determines the resident point according to the position attribution information and time attribute information
The type of information.
For example, the time range when time attribute information instruction mobile subscriber of resident point information is located at the resident point is concentrated
The 9 of Mon-Fri weekly:00 to 18:00, and the position attribution information of the resident point information indicates the position of the resident point
For an office building, then computer equipment determines that type of the resident point information is place of working.
In another example the time attribute information instruction mobile subscriber of resident point information is located at time range collection when this resides point
In the 21 of weekend:00 to 24:00, and the position attribution information of the resident point information indicates that the resident point is attached positioned at shopping centre
Closely, then computer equipment determines that the type of the resident point information is public place of entertainment.
It should be noted that the above-mentioned examples are merely illustrative of the technical solutions of the present invention, rather than the limit to the present invention
System, it should be appreciated by those skilled in the art that it is any according to the position attribution information and time attribute information, determine the resident point
The implementation of the type of information, should be included in the scope of the present invention.
According to the scheme of the present embodiment, it can determine that mobile subscriber's is each according to multiple resident information of mobile subscriber
The type of resident point, and the probability that user occurs in certain resident point region is predicted to a certain extent.
Fig. 3 is the device of the resident information that mobile subscriber is determined in computer equipment of one embodiment of the invention
Structure diagram.Include being used to obtain the mobile use according to the device of the resident information of the definite mobile subscriber of the present embodiment
The device (hereinafter referred to as " the first acquisition device 1 ") of multiple event information at family and for based on clustering algorithm to described more
A event information carries out cluster analysis, to determine the device of multiple resident information of the mobile subscriber (hereinafter referred to as " the
One determining device 2 ").
First acquisition device 1 obtains multiple event information of mobile subscriber.
Wherein, the event information is used to indicate that the locus of the mobile subscriber and mobile subscriber are located at the sky
Between position when corresponding time point information.Preferably, the event information can have many forms, including but unlimited
In:Point in one vector measured, hyperspace etc.;It is highly preferred that the event information is space-time vector.
For example, the event information of mobile subscriber is space-time vector α=(a, b, c, d), wherein, (a, b, c)
For the coordinate of the locus of mobile subscriber, d is located at corresponding time point information during the locus for mobile subscriber.
Specifically, the first acquisition device 1 can obtain multiple event information of mobile subscriber in several ways.For example,
First acquisition device 1 receives from other computer equipments, mobile subscriber multiple event information;In another example mobile use
Family timing reports its event information to computer equipment, then within a period of time, the first acquisition device 1 receives mobile subscriber institute
Multiple event information reported etc..
It should be noted that the above-mentioned examples are merely illustrative of the technical solutions of the present invention, rather than the limit to the present invention
System, it should be appreciated by those skilled in the art that the implementation of any multiple event information for obtaining mobile subscriber, should all include
Within the scope of the invention.
First determining device 2 is based on clustering algorithm and carries out cluster analysis to the multiple event information, described to determine
Multiple resident information of mobile subscriber.
Wherein, the clustering algorithm includes any algorithm that can be used for carrying out cluster analysis, for example, density clustering
Algorithm, EM algorithms etc..Preferably, the clustering algorithm needs to set the quantity of cluster centre;It is highly preferred that the clustering algorithm
For density-based algorithms.
Wherein, the resident information includes any information for being used to indicate the resident point of mobile subscriber;Preferably, it is described
Resident point information includes the relevant any information of resident point with mobile subscriber;Preferably, directly cluster analysis can be obtained
Class in cluster result is as resident point information.It is highly preferred that it can be come true by carrying out statistical analysis to the class in cluster result
Such fixed corresponding resident information, wherein, the resident information includes position attribution information and time attribute information, described
Position attribution information is used for locus or the position range for indicating the resident point, and the time attribute information is used to indicate to move
Multiple time point informations or time range when user is located at the resident point.
Specifically, the first determining device 2 is based on clustering algorithm and carries out cluster analysis to the multiple event information, to obtain
It must include the cluster result of multiple classes, and multiple resident information of mobile subscriber are determined according to the plurality of class.
For example, the quantity set of the cluster centre of clustering algorithm is predetermined quantity by the first determining device 2, such as 4;First is true
Determine device 2 and 4 event information are selected from the multiple event information as cluster centre, for multiple event information
In each event information, the first determining device 2 calculate the event information respectively between 4 cluster centres away from
From, and by the event information categorization to the cluster centre corresponding to minimum range;Afterwards, the first determining device 2 is according to cluster
As a result 4 classes in, to determine the 4 of mobile subscriber resident point information.
As a kind of preferred solution of the present embodiment, the clustering algorithm needs to set the quantity of cluster centre.
Wherein, clustering algorithm of first determining device 2 based on needs setting cluster centre, to the multiple event
Information carries out cluster analysis, and determines that the mode of multiple resident information of the mobile subscriber includes but not limited to:
1) quantity of the cluster centre of clustering algorithm has predefined, then the first determining device 2 is directly based upon fixed
The quantity of cluster centre, operation clustering algorithm determine the movement to carry out cluster analysis to the multiple event information
Multiple resident information of user.
2) quantity of the cluster centre of clustering algorithm does not determine, then in the case, the first determining device 2 needs first really
The quantity of a fixed suitable cluster centre.
Specifically, in this implementation, the first determining device 2 can determine a suitable hypothesis from multiple hypothesis quantity
Quantity, the quantity as cluster centre.First determining device 2 further comprises for the whole for multiple hypothesis quantity or portion
Each hypothesis quantity in point, is the hypothesis quantity by the quantity set of the cluster centre of the clustering algorithm, and based on described
Clustering algorithm carries out cluster analysis to the multiple event information, obtains cluster result corresponding with the hypothesis quantity, and root
According to the corresponding multiple cluster results of multiple hypothesis quantity, select a hypothesis quantity device (it is not shown, hereinafter referred to as
" selection device ") and for according to the corresponding cluster result of selected hypothesis quantity, determining that the mobile subscriber's is multiple
The device (not shown, hereinafter referred to as " the first sub- determining device ") of resident point information.
For each hypothesis quantity in all or parts of multiple hypothesis quantity, selection device is by the clustering algorithm
The quantity set of cluster centre is the hypothesis quantity, and the multiple event information is clustered based on the clustering algorithm
Analysis, obtains cluster result corresponding with the hypothesis quantity, and assumes the corresponding multiple cluster results of quantity according to multiple,
One hypothesis quantity of selection.
Preferably, selection device is based at least one of following, according to the corresponding multiple clusters of multiple hypothesis quantity
As a result a hypothesis quantity is selected:
The quantity for the event information that class in the corresponding cluster result of-hypothesis quantity includes.
Preferably, the quantity of the event information included in class is more, then usual cluster result is better.
The dispersion of class in the corresponding cluster result of-hypothesis quantity.
Preferably, the dispersion of class is lower, then usual cluster result is better.
Wherein, the dispersion is used for the dense degree for indicating class.Wherein, selection device can be determined using various ways
The dispersion, e.g., all event information of the computer equipment in class determine the average of class, and calculate each event
Very poor, mean difference or standard deviation between information and the average etc. represent such dispersion.
It should be noted that selection device can be realized using various ways.For example, the implementation of selection device includes
But it is not limited to:
A) in this implementation, selection device further comprises being used to not determine for one in the multiple hypothesis quantity
The hypothesis quantity of its corresponding cluster result, is the hypothesis quantity by the quantity set of the cluster centre of the clustering algorithm, and
Cluster analysis is carried out to the multiple event information based on the clustering algorithm, obtains cluster knot corresponding with the hypothesis quantity
The device (not shown, hereinafter referred to as " the first clustering apparatus ") of fruit, for meeting when the corresponding cluster result of hypothesis quantity
During one predetermined condition, using the hypothesis quantity as it is described it is selected assume quantity device (it is not shown, hereinafter referred to as " first
Setting device ") and for when the corresponding cluster result of hypothesis quantity does not meet first predetermined condition, described in triggering
First clustering apparatus repeats the device (not shown, hereinafter referred to as " the first trigger device ") of operation.
For the hypothesis quantity for not determining its corresponding cluster result in the multiple hypothesis quantity, the first cluster dress
It is the hypothesis quantity to put the quantity set of the cluster centre of the clustering algorithm, and based on the clustering algorithm to the multiple
Event information carries out cluster analysis, obtains cluster result corresponding with the hypothesis quantity.
When the corresponding cluster result of hypothesis quantity meets the first predetermined condition, the first setting device is by the hypothesis quantity
As the selected hypothesis quantity.
Wherein, first predetermined condition includes any predetermined condition for being used to select to assume quantity.Preferably, it is described
First predetermined condition includes but not limited to:
The quantity a predetermined level is exceeded threshold value for the event information that class in the-cluster result includes.
The dispersion of class in the-cluster result is less than predetermined dispersion threshold value.
For example, predetermined quantity threshold value is 100, it is assumed that the corresponding cluster result of quantity includes 4 classes, when in 4 classes
The quantity of null point information is respectively:120、110、108、150.Then the first setting device determines to assume the corresponding cluster result of quantity
Each class in event information the equal a predetermined level is exceeded threshold value of quantity, then the first setting device determine the cluster result accord with
The first predetermined condition is closed, and using the hypothesis quantity as selected hypothesis quantity.
When the corresponding cluster result of hypothesis quantity does not meet first predetermined condition, the first trigger device triggering institute
State the first clustering apparatus and repeat operation.
Specifically, when the corresponding cluster result of hypothesis quantity does not meet the first predetermined condition, the first trigger device touches
Sending out the first clustering apparatus described repeats operation, is not determined corresponding to the hypothesis quantity of its corresponding cluster result to obtain
Cluster result;And so on, until when the corresponding cluster result of a hypothesis quantity meets the first predetermined condition, the first setting dress
Put using the hypothesis quantity as selected hypothesis quantity, and stop operation.
For example, multiple assume that quantity includes all natural numbers from 2 to 1000.First clustering apparatus performs operation for the first time
When, the hypothesis quantity of selection is 2, and in the case where being 2 by the quantity set of cluster centre, based on clustering algorithm to described more
A event information carries out cluster analysis, obtains cluster result corresponding with assuming quantity " 2 ";Then, since " 2 " are corresponding poly-
Class result does not meet first predetermined condition, and the first trigger device triggers first clustering apparatus and repeats operation, selects
The hypothesis quantity " 4 " for not determining its corresponding cluster result is selected, and determines its cluster result;Then, due to " 4 " corresponding cluster
As a result first predetermined condition is not met, the first trigger device, which continues to trigger first clustering apparatus, repeats operation;
And so on, until obtaining the hypothesis quantity " 5 " for meeting the first predetermined condition, the first setting device is by " 5 " as selected
Assuming that quantity.
In this implementation, computer equipment only needs to obtain the hypothesis quantity for meeting the first predetermined condition, you can
Subsequent operation is performed based on the hypothesis quantity, without traveling through and obtaining the cluster result of all hypothesis quantity.
B) the multiple to assume increased number or successively decrease in this implementation, selection device further comprises being used for institute
State one in multiple hypothesis quantity and assume that quantity assumes quantity as current, by the quantity of the cluster centre of the clustering algorithm
It is set as the current hypothesis quantity, and cluster analysis is carried out to the multiple event information based on the clustering algorithm, obtains
With the current device (not shown, hereinafter referred to as " the second clustering apparatus ") for assuming the corresponding cluster result of quantity, for by institute
The quantity set for stating the cluster centre of clustering algorithm is next hypothesis quantity of the current hypothesis quantity, and is based on the cluster
Algorithm carries out cluster analysis to the multiple event information, obtains the dress with the corresponding cluster result of next hypothesis quantity
Put (not shown, hereinafter referred to as " the 3rd clustering apparatus "), for when the corresponding cluster result of next hypothesis quantity is worse than
During the current hypothesis corresponding cluster result of quantity, current assume quantity as the selected device for assuming quantity this
(not shown, hereinafter referred to as " the second setting device ") and for being better than when the corresponding cluster result of next hypothesis quantity
During the current hypothesis corresponding cluster result of quantity, using next hypothesis quantity as the current hypothesis quantity, described in triggering
For obtain with the device of the corresponding cluster result of next hypothesis quantity repeat operation device (it is not shown, below
Referred to as " the second trigger device ").
One in the multiple hypothesis quantity is assumed that quantity assumes quantity as current by the second clustering apparatus, by described in
The quantity set of the cluster centre of clustering algorithm is the current hypothesis quantity, and based on the clustering algorithm to the multiple space-time
Point information carries out cluster analysis, obtains and current assumes the corresponding cluster result of quantity with this.
For example, multiple hypothesis quantity are included from 2 to 1000 incremental multiple natural numbers.Second clustering apparatus is by " 2 " conduct
It is current to assume quantity, and be " 2 " by the quantity set of the cluster centre of clustering algorithm, and based on the clustering algorithm to described more
A event information carries out cluster analysis, obtains cluster result corresponding with " 2 ".
The quantity set of the cluster centre of the clustering algorithm is the next of the current hypothesis quantity by the 3rd clustering apparatus
A hypothesis quantity, and cluster analysis is carried out to the multiple event information based on the clustering algorithm, obtain next with this
Assuming that the corresponding cluster result of quantity.
For example, next hypothesis quantity of the 3rd clustering apparatus by the quantity set of the cluster centre of clustering algorithm for " 2 "
" 3 ", and cluster analysis is carried out to the multiple event information based on clustering algorithm, obtain corresponding with next hypothesis quantity
Cluster result.
When the corresponding cluster result of next hypothesis quantity is worse than the corresponding cluster result of the current hypothesis quantity,
Second setting device is using the current hypothesis quantity as the selected hypothesis quantity.
Preferably, the quantity of the space time information point that can be included according to the dispersion and/or class of class in cluster result determines
Whether the corresponding cluster result of next hypothesis quantity is worse than the corresponding cluster result of the current hypothesis quantity.
For example, the variance E in the corresponding cluster result of next hypothesis quantity between class can be calculated1, and current hypothesis
Variance E in the corresponding cluster result of quantity between class2, and compare E1And E2, work as E1More than E2, the second setting device can determine that down
One is assumed that the corresponding cluster result of quantity is worse than the corresponding cluster result of the current hypothesis quantity;Work as E1Less than E2, the second setting
Device can determine that the corresponding cluster result of next hypothesis quantity cluster result corresponding better than the current hypothesis quantity.
When the corresponding cluster result of next hypothesis quantity cluster result corresponding better than the current hypothesis quantity,
Second trigger device triggers the 3rd clustering apparatus and repeats operation.
For example, currently assume that quantity is " 2 ", and next hypothesis quantity of " 2 " is " 3 ", and " 3 " corresponding cluster result
Cluster result corresponding better than " 2 ", then " 3 " are assumed quantity by the second trigger device as current, and trigger the 3rd clustering apparatus
Operation is repeated, obtains the cluster result of " 4 ";Then, if " 4 " corresponding cluster result is better than " 3 " corresponding cluster result,
Then " 4 " are assumed quantity by the second trigger device as current, are continued the 3rd clustering apparatus of triggering and are repeated operation;With such
Push away, until when the corresponding cluster result of next hypothesis quantity is worse than the current hypothesis quantity corresponding cluster result, second sets
Device is determined using the current hypothesis quantity as the selected hypothesis quantity.
Due to when it is multiple hypothesis quantity show increasing or decreasing relation when, an optimal corresponding cluster of hypothesis quantity
As a result, its corresponding cluster result of two neighboring hypothesis quantity can be better than, therefore, in this implementation, optimal vacation can be obtained
If quantity.Also, since hypothesis quantity execution subsequent operation can be based on after obtaining optimal hypothesis quantity, without continuing to obtain
Other are taken to assume the cluster result of quantity, therefore under normal conditions, this implementation is without traveling through and obtaining all hypothesis quantity
Cluster result.
C) selection device further comprises for for each hypothesis quantity in multiple hypothesis quantity, the cluster to be calculated
The quantity set of the cluster centre of method is the hypothesis quantity, and the multiple event information is carried out based on the clustering algorithm
Cluster analysis, obtains the device (not shown, hereinafter referred to as " the 4th clustering apparatus ") of cluster result corresponding with the hypothesis quantity
And for according to the corresponding multiple cluster results of the multiple hypothesis quantity, selecting the device of a hypothesis quantity (to scheme
Do not show, hereinafter referred to as " sub- selection device ").
For each hypothesis quantity in multiple hypothesis quantity, the 4th clustering apparatus is by the cluster centre of the clustering algorithm
Quantity set be the hypothesis quantity, and based on the clustering algorithm to the multiple event information carry out cluster analysis, obtain
Obtain cluster result corresponding with the hypothesis quantity.
For example, there are 4 hypothesis quantity:2、3、4、5.4th clustering apparatus is clustered respectively based on the clustering algorithm
When cluster result, the quantity of cluster centre when the quantity of cluster result, cluster centre when the quantity at center is 2 is 3 are 4
The cluster result when quantity of cluster result and cluster centre is 5.
Sub- selection device selects a hypothesis number according to the corresponding multiple cluster results of the multiple hypothesis quantity
Amount.
Wherein, sub- selection device selects a hypothesis quantity according to the corresponding multiple cluster results of multiple hypothesis quantity
Implementation, select a hypothesis according to the corresponding multiple cluster results of multiple hypothesis quantity with hereinbefore selection device
The implementation of quantity is similar, and details are not described herein.
It should be noted that multiple forms assumed quantity and can behave as set, are such as set [2,3,4 ..., 1000],
Then computer equipment can be read directly from the set assumes quantity.Alternatively, multiple forms assumed quantity and can behave as formula,
Such as k=K+n △;Wherein, k represents to assume quantity, and K is radix (usual K can use 2), △=1, n=0,1,2 ..., 998;Then count
The hypothesis quantity of its needs can be calculated by the formula by calculating machine equipment.
It should be noted that the above-mentioned examples are merely illustrative of the technical solutions of the present invention, rather than the limit to the present invention
System, it should be appreciated by those skilled in the art that each hypothesis quantity in any all or part for multiple hypothesis quantity, will
The quantity set of the cluster centre of the clustering algorithm is the hypothesis quantity, and based on the clustering algorithm to the multiple space-time
Point information carries out cluster analysis, obtains cluster result corresponding with the hypothesis quantity, and correspond to respectively according to multiple hypothesis quantity
Multiple cluster results, select one hypothesis quantity implementation, should be included in the scope of the present invention.
First sub- determining device determines that the mobile subscriber's is more according to the corresponding cluster result of selected hypothesis quantity
A resident information.
Wherein, the first sub- determining device can use various ways to assume the corresponding cluster result of quantity according to selected,
Determine multiple resident information of the mobile subscriber.
For example, the first sub- determining device can directly be believed multiple classes of cluster result as multiple resident points of mobile subscriber
Breath.
In another example for each class in cluster result, the first sub- determining device can by such progress statistical analysis,
Such as the locus and time point information of all event information in such are counted respectively, to determine that such is corresponding
Resident point information.
It should be noted that the above-mentioned examples are merely illustrative of the technical solutions of the present invention, rather than the limit to the present invention
System, it should be appreciated by those skilled in the art that it is any according to the corresponding cluster result of selected hypothesis quantity, determine the movement
The implementation of multiple resident information of user, should be included in the scope of the present invention.
, can be by carrying out cluster analysis to the event information of mobile subscriber, to determine shifting according to the scheme of the present embodiment
Multiple resident points at family are employed, so as to more accurately understand the scope of activities of mobile subscriber and rule of life.
Fig. 4 is the device of the resident information that mobile subscriber is determined in computer equipment of another embodiment of the present invention
Structure diagram.First acquisition device 1, the are included according to the device of the resident information of the definite mobile subscriber of the present embodiment
One determining device 2 and for according to the multiple resident information, determining each resident point in the multiple resident information
The device (hereinafter referred to as " the second determining device 3 ") of the type of information.Wherein, first acquisition device 1 and first determines dress
Put 2 to be described in detail with reference to the embodiment shown in FIG. 3, details are not described herein.
Second determining device 3 determines each normal in the multiple resident information according to the multiple resident information
The type of stationary point information
Wherein, the type of the resident information is used for the property for indicating the resident point of mobile subscriber, such as family, dining room, joy
Happy place, place of working etc..
Specifically, determine that this is normal by analyzing the resident point information for each resident information, the second determining device 3
The type of stationary point information.
For example, according to resident information and map, the second determining device 3 determines the position corresponding to the resident point information
Scope is in a residential block, then the second determining device 3 determines that the type of the resident point information is family.
Preferably, second determining device 3 further comprises the position attribution information for obtaining the resident point information
With the device (not shown, hereinafter referred to as " the second acquisition device ") of time attribute information and for being believed according to the position attribution
Breath and time attribute information, determine that (not shown, hereinafter referred to as " the second son determines dress for the device of the type of the resident point information
Put ").
Second acquisition device obtains the position attribution information and time attribute information of the resident point information.
Wherein, the second acquisition device can use various ways to obtain position attribution information and the time category that this resides point information
Property information.
For example, when the resident information be cluster result in class when, the second acquisition device in such institute sometimes
Null point information carries out statistical analysis, and the position that the resident point information is obtained come the locus in all event information belongs to
Property information, and the time point information in all event information obtains the time attribute information of the resident point information.
In another example when the resident information be by cluster result class carry out statistical analysis come obtain when,
Second acquisition device can directly extract the position attribution information and time attribute letter of the resident point information from the resident point information
Breath.
It should be noted that the above-mentioned examples are merely illustrative of the technical solutions of the present invention, rather than the limit to the present invention
System, it should be appreciated by those skilled in the art that any acquisition resident the position attribution information for putting information and time attribute information
Implementation, should be included in the scope of the present invention.
Second sub- determining device determines the class for residing point information according to the position attribution information and time attribute information
Type.
For example, the time range when time attribute information instruction mobile subscriber of resident point information is located at the resident point is concentrated
The 9 of Mon-Fri weekly:00 to 18:00, and the position attribution information of the resident point information indicates the position of the resident point
For an office building, then the second sub- determining device determines that type of the resident point information is place of working.
In another example the time attribute information instruction mobile subscriber of resident point information is located at time range collection when this resides point
In the 21 of weekend:00 to 24:00, and the position attribution information of the resident point information indicates that the resident point is attached positioned at shopping centre
Closely, then the second sub- determining device determines that the type of the resident point information is public place of entertainment.
It should be noted that the above-mentioned examples are merely illustrative of the technical solutions of the present invention, rather than the limit to the present invention
System, it should be appreciated by those skilled in the art that it is any according to the position attribution information and time attribute information, determine the resident point
The implementation of the type of information, should be included in the scope of the present invention.
According to the scheme of the present embodiment, it can determine that mobile subscriber's is each according to multiple resident information of mobile subscriber
The type of resident point, and the probability that user occurs in certain resident point region is predicted to a certain extent.
It should be noted that the present invention can be carried out in the assembly of software and/or software and hardware, for example, this hair
Bright each device can using application-specific integrated circuit (ASIC) or any other realized similar to hardware device.In one embodiment
In, software program of the invention can be performed by processor to realize steps described above or function.Similarly, it is of the invention
Software program (including relevant data structure) can be stored in computer readable recording medium storing program for performing, for example, RAM memory,
Magnetically or optically driver or floppy disc and similar devices.In addition, some steps or function of the present invention can employ hardware to realize, example
Such as, as coordinating with processor so as to performing the circuit of each step or function.
It is obvious to a person skilled in the art that the invention is not restricted to the details of above-mentioned one exemplary embodiment, Er Qie
In the case of without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Profit requires rather than described above limits, it is intended that all in the implication and scope of the equivalency of claim by falling
Change is included in the present invention.Any reference numeral in claim should not be considered as to the involved claim of limitation.This
Outside, it is clear that one word of " comprising " is not excluded for other units or step, and odd number is not excluded for plural number.That is stated in system claims is multiple
Unit or device can also be realized by a unit or device by software or hardware.The first, the second grade word is used for table
Show title, and be not offered as any specific order.
Claims (16)
1. a kind of method for the resident information that mobile subscriber is determined in computer equipment, wherein, this method includes following step
Suddenly:
A. multiple event information of the mobile subscriber are obtained, wherein, the event information is used to indicate the mobile use
The locus at family and mobile subscriber are located at corresponding time point information during the locus;
B. cluster analysis is carried out to the multiple event information based on clustering algorithm, to determine that the mobile subscriber's is multiple normal
Stationary point information;And
X determines the type of each resident point information in the multiple resident information according to the multiple resident information, and
Predict the probability that user occurs in resident point region;
Wherein, the step x is included to the following steps each performed in the multiple resident information:
- the position attribution information and time attribute information of residing point information is obtained, wherein, the position attribution information is used to refer to
Show the position range of resident point, the time attribute information is used to indicate time range when mobile subscriber is located at the resident point;
- according to the position attribution information and time attribute information, determine the type for residing point information.
2. according to the method described in claim 1, wherein, the clustering algorithm needs to set the quantity of cluster centre.
3. according to the method described in claim 2, wherein, the step b comprises the following steps:
B1 is for each hypothesis quantity in all or part of multiple hypothesis quantity, by the cluster centre of the clustering algorithm
Quantity set is the hypothesis quantity, and carries out cluster analysis to the multiple event information based on the clustering algorithm, is obtained
Cluster result corresponding with the hypothesis quantity, and according to the corresponding multiple cluster results of multiple hypothesis quantity, select one
Assuming that quantity;
B2 is according to the selected multiple resident information assumed the corresponding cluster result of quantity, determine the mobile subscriber.
4. according to the method described in claim 3, wherein, the step b1 comprises the following steps:
B111 is for the hypothesis quantity for not determining its corresponding cluster result in the multiple hypothesis quantity, by the cluster
The quantity set of the cluster centre of algorithm is the hypothesis quantity, and based on the clustering algorithm to the multiple event information into
Row cluster analysis, obtains cluster result corresponding with the hypothesis quantity;
B112 is when the corresponding cluster result of hypothesis quantity meets the first predetermined condition, using the hypothesis quantity as described selected
The hypothesis quantity selected;
B113 is not when the corresponding cluster result of hypothesis quantity meets first predetermined condition, repeating said steps b111.
5. according to the method described in claim 3, wherein, the multiple to assume increased number or successively decrease, the step b1 includes
Following steps:
One in the multiple hypothesis quantity is assumed that quantity assumes quantity as current by b121, by the poly- of the clustering algorithm
The quantity set at class center is the current hypothesis quantity, and the multiple event information is gathered based on the clustering algorithm
Alanysis, obtains and current assumes the corresponding cluster result of quantity with this;
The quantity set of the cluster centre of the clustering algorithm is the current next hypothesis quantity for assuming quantity by b122, and
Cluster analysis is carried out to the multiple event information based on the clustering algorithm, is obtained corresponding with next hypothesis quantity
Cluster result;
B123 when the corresponding cluster result of next hypothesis quantity is worse than the corresponding cluster result of the current hypothesis quantity,
Using the current hypothesis quantity as the selected hypothesis quantity;
B124 when the corresponding cluster result of next hypothesis quantity cluster result corresponding better than the current hypothesis quantity,
Using next hypothesis quantity as the current hypothesis quantity, repeating said steps b122.
6. according to the method described in claim 3, wherein, the step b1 comprises the following steps:
- for each hypothesis quantity in multiple hypothesis quantity, being by the quantity set of the cluster centre of the clustering algorithm should
Assuming that quantity, and cluster analysis is carried out to the multiple event information based on the clustering algorithm, obtain and the hypothesis quantity
Corresponding cluster result;
- according to the corresponding multiple cluster results of the multiple hypothesis quantity, select a hypothesis quantity.
7. the method according to any one of claim 3 to 6, wherein, based at least one of following, according to multiple hypothesis
The corresponding multiple cluster results of quantity select a hypothesis quantity:
The quantity for the event information that class in the corresponding cluster result of-hypothesis quantity includes;
The dispersion of class in the corresponding cluster result of-hypothesis quantity.
8. according to the method described in claim 1, wherein, the event information is space-time vector.
9. a kind of device for the resident information that mobile subscriber is determined in computer equipment, wherein, which includes following dress
Put:
Device for the multiple event information for obtaining the mobile subscriber, wherein, the event information is used to indicate institute
State the locus of mobile subscriber and mobile subscriber is located at corresponding time point information during the locus;
For carrying out cluster analysis to the multiple event information based on clustering algorithm, to determine that the mobile subscriber's is multiple
The device of resident point information;And
For according to the multiple resident information, determining each class for residing point information in the multiple resident information
Type, and predict the device for the probability that user occurs in resident point region;
Wherein, the device for determining the type of resident point information is included to each holding in the multiple resident information
The following device of row operation:
For obtaining the resident point position attribution information of information and the device of time attribute information, wherein, the position attribution
Information is used for the position range for indicating resident point, and the time attribute information is used to indicating when mobile subscriber be located at this resident
Time range;
For according to the position attribution information and time attribute information, determining the device for residing the type of point information.
10. device according to claim 9, wherein, the clustering algorithm needs to set the quantity of cluster centre.
11. device according to claim 10, wherein, the device for being used to determine multiple resident information is including following
Device:
For each hypothesis quantity in all or part for multiple hypothesis quantity, by the cluster centre of the clustering algorithm
Quantity set be the hypothesis quantity, and based on the clustering algorithm to the multiple event information carry out cluster analysis, obtain
Cluster result corresponding with the hypothesis quantity is obtained, and according to the corresponding multiple cluster results of multiple hypothesis quantity, selects one
A device for assuming quantity;
For according to the corresponding cluster result of selected hypothesis quantity, determining multiple resident information of the mobile subscriber
Device.
12. according to the devices described in claim 11, wherein, the device for being used for one hypothesis quantity of selection includes following dress
Put:
For the hypothesis quantity for its definite corresponding cluster result in the multiple hypothesis quantity, by the cluster
The quantity set of the cluster centre of algorithm is the hypothesis quantity, and based on the clustering algorithm to the multiple event information into
Row cluster analysis, obtains the device of cluster result corresponding with the hypothesis quantity;
For when the corresponding cluster result of hypothesis quantity meets the first predetermined condition, using the hypothesis quantity as described selected
The device for the hypothesis quantity selected;
For when the corresponding cluster result of hypothesis quantity does not meet first predetermined condition, triggering to be described for for institute
The hypothesis quantity for not determining its corresponding cluster result in multiple hypothesis quantity is stated, is obtained corresponding with the hypothesis quantity poly-
The device of class result repeats the device of operation.
13. according to the devices described in claim 11, wherein, the multiple hypothesis or is successively decreased at increased number, described to be used to select
The device of one hypothesis quantity includes following device:
For one in the multiple hypothesis quantity to be assumed, quantity assumes quantity as current, by the poly- of the clustering algorithm
The quantity set at class center is the current hypothesis quantity, and the multiple event information is gathered based on the clustering algorithm
Alanysis, obtains the device with the corresponding cluster result of current hypothesis quantity;
For being the current next hypothesis quantity for assuming quantity by the quantity set of the cluster centre of the clustering algorithm, and
Cluster analysis is carried out to the multiple event information based on the clustering algorithm, is obtained corresponding with next hypothesis quantity
The device of cluster result;
For when the corresponding cluster result of next hypothesis quantity is worse than the corresponding cluster result of the current hypothesis quantity,
Using the current hypothesis quantity as the selected device for assuming quantity;
For when the corresponding cluster result of next hypothesis quantity cluster result corresponding better than the current hypothesis quantity,
Next hypothesis quantity is described corresponding with next hypothesis quantity for obtaining as the current hypothesis quantity, triggering
The device of cluster result repeats the device of operation.
14. according to the devices described in claim 11, wherein, the device for being used for one hypothesis quantity of selection includes following dress
Put:
For for each hypothesis quantity in multiple hypothesis quantity, the quantity set by the cluster centre of the clustering algorithm to be
The hypothesis quantity, and cluster analysis is carried out to the multiple event information based on the clustering algorithm, obtain and the hypothesis number
Measure the device of corresponding cluster result;
For according to the corresponding multiple cluster results of the multiple hypothesis quantity, selecting a device for assuming quantity.
15. the device according to any one of claim 11 to 14, wherein, based at least one of following, according to multiple
Assuming that the corresponding multiple cluster results of quantity select a hypothesis quantity:
The quantity for the event information that class in the corresponding cluster result of-hypothesis quantity includes;
The dispersion of class in the corresponding cluster result of-hypothesis quantity.
16. device according to claim 9, wherein, the event information is space-time vector.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410443562.3A CN104252527B (en) | 2014-09-02 | 2014-09-02 | A kind of method and apparatus of the resident information of definite mobile subscriber |
PCT/CN2014/093759 WO2016033901A1 (en) | 2014-09-02 | 2014-12-12 | Method and apparatus for determining resident point information about mobile user |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410443562.3A CN104252527B (en) | 2014-09-02 | 2014-09-02 | A kind of method and apparatus of the resident information of definite mobile subscriber |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104252527A CN104252527A (en) | 2014-12-31 |
CN104252527B true CN104252527B (en) | 2018-04-20 |
Family
ID=52187417
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410443562.3A Active CN104252527B (en) | 2014-09-02 | 2014-09-02 | A kind of method and apparatus of the resident information of definite mobile subscriber |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN104252527B (en) |
WO (1) | WO2016033901A1 (en) |
Families Citing this family (26)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105847310A (en) * | 2015-01-13 | 2016-08-10 | 中国移动通信集团江苏有限公司 | Position determination method and apparatus |
CN106153031B (en) * | 2015-04-13 | 2019-08-30 | 骑记(厦门)科技有限公司 | Motion profile representation method and device |
CN104765873B (en) * | 2015-04-24 | 2019-03-26 | 百度在线网络技术(北京)有限公司 | User's similarity determines method and apparatus |
CN106294485B (en) * | 2015-06-05 | 2019-11-01 | 华为技术有限公司 | Determine the method and device in significant place |
CN105307121B (en) * | 2015-10-16 | 2019-03-26 | 上海晶赞科技发展有限公司 | A kind of information processing method and device |
CN105843943B (en) * | 2016-04-08 | 2019-03-01 | 深圳广联赛讯有限公司 | Vehicle permanent residence analysis method |
CN106897331B (en) * | 2016-06-07 | 2020-09-11 | 阿里巴巴集团控股有限公司 | User key position data acquisition method and device |
CN106202236A (en) * | 2016-06-28 | 2016-12-07 | 联想(北京)有限公司 | A kind of customer location Forecasting Methodology and device |
CN106127487A (en) * | 2016-08-26 | 2016-11-16 | 成都市硕达科技股份有限公司 | A kind of illegal safe preservation system and using method thereof |
CN110832918B (en) * | 2017-06-30 | 2021-07-09 | Oppo广东移动通信有限公司 | User position identification method and device, storage medium and electronic equipment |
CN107515890A (en) * | 2017-07-04 | 2017-12-26 | 深圳市金立通信设备有限公司 | A kind of method and terminal for identifying resident point |
CN109428929A (en) * | 2017-08-31 | 2019-03-05 | 阿里巴巴集团控股有限公司 | Determination method, server and the user client of the location information of target object |
CN109543926B (en) * | 2017-09-21 | 2023-05-02 | 阿里巴巴集团控股有限公司 | Task verification method, mobile terminal and server |
CN108174350B (en) * | 2017-11-30 | 2020-12-11 | 北京三快在线科技有限公司 | Positioning method and device |
CN108122012B (en) * | 2017-12-28 | 2020-11-24 | 百度在线网络技术(北京)有限公司 | Method, device and equipment for determining center point of stationary point and storage medium |
CN108650632B (en) * | 2018-04-28 | 2020-05-26 | 广州市交通规划研究院 | Stagnation point judgment method based on occupational correspondence and time-space kernel clustering |
CN109672715A (en) * | 2018-09-13 | 2019-04-23 | 深圳壹账通智能科技有限公司 | User's permanent residence judgment method, device, equipment and computer readable storage medium |
CN109934265B (en) * | 2019-02-15 | 2021-06-11 | 同盾控股有限公司 | Method and device for determining address of ordinary station |
CN111861526B (en) * | 2019-04-30 | 2024-05-21 | 京东城市(南京)科技有限公司 | Method and device for analyzing object source |
CN112218230B (en) * | 2019-06-24 | 2023-03-24 | 中兴通讯股份有限公司 | Method and device for acquiring user resident position and computer readable storage medium |
CN110493706A (en) * | 2019-06-27 | 2019-11-22 | 中国移动通信集团海南有限公司 | The permanent residence of mobile subscriber determines method, apparatus and computer equipment |
CN112394647B (en) * | 2019-08-19 | 2024-04-19 | 中国移动通信有限公司研究院 | Control method, device, equipment and storage medium of household equipment |
CN111797181B (en) * | 2020-05-26 | 2023-09-05 | 北京城市象限科技有限公司 | Positioning method, device, control equipment and storage medium for user location |
CN112364907A (en) * | 2020-11-03 | 2021-02-12 | 北京红山信息科技研究院有限公司 | Method, system, server and storage medium for general investigation of frequent station of user to be tested |
CN113256405B (en) * | 2021-06-22 | 2021-10-12 | 平安科技(深圳)有限公司 | Method, device, equipment and storage medium for predicting cheating user concentrated area |
CN113688197A (en) * | 2021-08-26 | 2021-11-23 | 沈阳美行科技有限公司 | Resident point label determination method, device, equipment and storage medium |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103116696A (en) * | 2013-01-16 | 2013-05-22 | 上海美慧软件有限公司 | Personnel resident site recognizing method based on sparsely sampled mobile phone locating data |
CN103218442A (en) * | 2013-04-22 | 2013-07-24 | 中山大学 | Method and system for life mode analysis based on mobile device sensor data |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP5696903B2 (en) * | 2009-10-14 | 2015-04-08 | 日本電気株式会社 | Action type extraction system, apparatus, method, recording medium storing program |
CN102629297B (en) * | 2012-03-06 | 2016-01-06 | 北京建筑工程学院 | A kind of traveler activity rule analytical approach based on stroke identification |
CN103593361B (en) * | 2012-08-14 | 2017-02-22 | 中国科学院沈阳自动化研究所 | Movement space-time trajectory analysis method in sense network environment |
-
2014
- 2014-09-02 CN CN201410443562.3A patent/CN104252527B/en active Active
- 2014-12-12 WO PCT/CN2014/093759 patent/WO2016033901A1/en active Application Filing
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103116696A (en) * | 2013-01-16 | 2013-05-22 | 上海美慧软件有限公司 | Personnel resident site recognizing method based on sparsely sampled mobile phone locating data |
CN103218442A (en) * | 2013-04-22 | 2013-07-24 | 中山大学 | Method and system for life mode analysis based on mobile device sensor data |
Also Published As
Publication number | Publication date |
---|---|
CN104252527A (en) | 2014-12-31 |
WO2016033901A1 (en) | 2016-03-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104252527B (en) | A kind of method and apparatus of the resident information of definite mobile subscriber | |
TWI777004B (en) | Marketing information push equipment, devices and storage media | |
Zhang et al. | Evaluating geo-social influence in location-based social networks | |
CN106293914B (en) | A kind of method and terminal of task schedule | |
CN105677881A (en) | Information recommendation method and device and server | |
CN109299420A (en) | Social media account processing method, device, equipment and readable storage medium storing program for executing | |
WO2015154472A1 (en) | Operation mode switching method and mobile terminal | |
US10432491B2 (en) | Control device for estimation of power consumption and energy efficiency of application containers | |
CN108021492B (en) | Alarm merging method and equipment | |
CN106611366A (en) | Method and device for automatically selecting article | |
CN102915481B (en) | A kind of method, device and equipment for user account is managed | |
JP7105227B2 (en) | Simulation program, simulation method and simulation device | |
CN104424361B (en) | Automatic definition heat storage and big workload | |
JP2011138194A (en) | Information processing device, information processing method, and program | |
CN104182663B (en) | A kind of method and device for differentiating simulator operation | |
CN107704317B (en) | Intelligent device and application management method thereof and device with storage function | |
CN110968802B (en) | Analysis method and analysis device for user characteristics and readable storage medium | |
CN110796591B (en) | GPU card using method and related equipment | |
CN106776757A (en) | User completes the indicating means and device of Net silver operation | |
CN109410198A (en) | A kind of timing motion detection method, device and equipment | |
CN108563758B (en) | User quantity measuring and calculating method, device, equipment and storage medium | |
CN114300082B (en) | Information processing method and device and computer readable storage medium | |
Papadopoulou et al. | A machine-learning approach for communication prediction of large-scale applications | |
CN104731867B (en) | A kind of method and apparatus that object is clustered | |
CN113345594A (en) | Information prediction method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |