CN109992726A - Position predicting method, device and readable storage medium storing program for executing - Google Patents
Position predicting method, device and readable storage medium storing program for executing Download PDFInfo
- Publication number
- CN109992726A CN109992726A CN201811212184.2A CN201811212184A CN109992726A CN 109992726 A CN109992726 A CN 109992726A CN 201811212184 A CN201811212184 A CN 201811212184A CN 109992726 A CN109992726 A CN 109992726A
- Authority
- CN
- China
- Prior art keywords
- data
- prediction model
- point data
- target
- target position
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of position predicting method, which includes: the position data obtained in corresponding first preset time period of mobile terminal;Position data is clustered, and is formed according to the cluster and stops point data;Prediction model is constructed according to point data is stopped, the corresponding destination probability value in the target position of the mobile terminal and the target position is predicted by prediction model.The invention also discloses a kind of user location prediction meanss and a kind of readable storage medium storing program for executing.The present invention can excavate building prediction model from the distributed locations data obtained in application, predict the destination probability value that the target position of the mobile terminal and the target position occur, and recommend location-based products & services to user.
Description
Technical field
The present invention relates to field of communication technology more particularly to position predicting methods, device and readable storage medium storing program for executing.
Background technique
In recent years, constantly improve with mobile internet environment, becoming increasingly popular for intelligent mobile terminal, obtains user's
Track has simultaneously obtained quick development according to the trajectory predictions user behavior of user.How to be accurately identified based on geographical location information
The hobby of user, and recommending maximally related product, service to user on this basis is one of the hot spot paid close attention at present.
The location information of the available user of intelligent mobile terminal has contained the position motion track of user, different rails
Mark has reacted different user personalities.By excavating the motion track of user from the historical geography position data of user, in advance
The geographical location for surveying user's future can help the position of business the combination users such as mobile finance, be user's recommendation deeper into, more
Personalized service.
However, since the application such as mobile financial class is different from social category or navigation type application, user's usage frequency and viscous
Property it is lower, cause the location data obtained that all there is apparent dispersibility on Spatial Dimension and time dimension, it is difficult to straight
It connects and constructs prediction model to it, therefore, not yet occur the realization of the application user location prediction such as mobile finance, shopping software at present
Scheme.
Summary of the invention
It is a primary object of the present invention to propose a kind of position predicting method, device and readable storage medium storing program for executing, it is intended to realize
From the distributed locations data of acquisition excavate building prediction model, predict the mobile terminal target position and the mesh
The destination probability value that cursor position occurs recommends location-based products & services to user.
To achieve the above object, the present invention provides a kind of position predicting method, and the position predicting method includes following step
It is rapid:
Obtain the position data in corresponding first preset time period of mobile terminal;
Position data is clustered, and is formed according to the cluster and stops point data;
Prediction model is constructed according to point data is stopped, the target position of the mobile terminal is predicted by prediction model, with
And the corresponding destination probability value in the target position.
Preferably, described according to point data building prediction model is stopped, the mobile terminal is predicted by prediction model
The step of target position and the corresponding destination probability value in the target position, comprising:
The first prediction model is constructed according to point data is stopped, the mobile terminal is predicted by first prediction model
The corresponding first object probability value in first object position and the first object position.
Preferably, according to point data the first prediction model of building is stopped, pass through first prediction model and predict the shifting
The step of corresponding first object probability value in the first object position and the first object position of dynamic terminal, comprising:
Corresponding stop point data of each period is determined according to the stop point data, wherein the period includes
Workaday working hour, the workaday rest period, the working hour on day off and day off the rest period;
It is derived with Bayes' theorem and calculates the corresponding probability for stopping point data of each period;
The first prediction model, first prediction are constructed according to the corresponding probability for stopping point data of each period
Model includes the probability distribution for stopping point data and corresponding period;
Based on the probability distribution and the second preset time period for stopping point data, by described in first prediction model prediction
The corresponding first object probability value in the first object position of mobile terminal and the first object position.
Preferably, described according to point data building prediction model is stopped, the mobile terminal is predicted by prediction model
The step of target position and the corresponding destination probability value in the target position, comprising:
The second prediction model is constructed according to point data is stopped, the mobile terminal is predicted by second prediction model
The second destination probability value that second target position and second target position occur.
Preferably, according to point data the second prediction model of building is stopped, pass through second prediction model and predict the shifting
The step of corresponding second destination probability value in the second target position and second target position of dynamic terminal, comprising:
It is ranked up according to chronological order to point data is stopped;
According to time window and sliding
Step-length will stop point data and be divided into multiple dwell point subsequences along time shaft;
Based on the minimum threshold frequency of dwell point subsequence, with PrefixSpan Sequential Pattern Mining Algorithm, excavation stops
Stationary point trajectory model constructs the second prediction model, and second prediction model includes all dwell point trajectory model and its general
Rate;
Based on dwell point trajectory model and in wire position data, matched and searched is carried out by the second prediction model, predicts institute
State the corresponding second destination probability value in the second target position and second target position of mobile terminal.
Preferably, position data is clustered, and the step of stopping point data is formed according to the cluster, comprising:
Determine the corresponding central point of the position data, and the position that preset threshold will be less than or equal at a distance from central point
It sets data to cluster as target, calculates the stop central point of target cluster;
Using the rest position data in addition to target clusters corresponding position data as the position data, continue to execute
The step of determining the position data corresponding central point;
When all position datas complete cluster operation, formed according to the stop central point of target cluster and target cluster
Stop point data.
Preferably, position data is clustered, and before the step of forming stop point data according to the cluster, also wrapped
It includes:
The position data is cleaned, to remove the repeatable position data in preset duration, and filtering unauthorized and network
Postpone the illegal position data generated.
Preferably, described according to point data building prediction model is stopped, the mobile terminal is predicted by prediction model
After the step of target position and the corresponding destination probability value in the target position, further includes:
Target position is ranked up according to the destination probability value size order that target position occurs;
According to the sequence of target position to user's recommended products and/or service.
In addition, to achieve the above object, the present invention also provides a kind of user location prediction meanss, the user location prediction
Device includes: memory, processor and to be stored in the user location that can be run on the memory and on the processor pre-
Ranging sequence, the user location Prediction program realize the step of position predicting method as described above when being executed by the processor
Suddenly.
In addition, to achieve the above object, the present invention also provides a kind of readable storage medium storing program for executing, being deposited on the readable storage medium storing program for executing
User location Prediction program is contained, the user location Prediction program realizes position prediction as described above when being executed by processor
The step of method.
The present invention obtains the position data in corresponding first preset time period of mobile terminal;Position data is gathered
Class, and formed according to the cluster and stop point data;Prediction model is constructed according to point data is stopped, institute is predicted by prediction model
State the corresponding destination probability value in target position and the target position of mobile terminal.By the above-mentioned means, the present invention can
When the mobile terminal locations data dispersed in receiving user's usage frequency and the lower application of viscosity, prediction model is constructed,
Based on mobile terminal in line position and the second preset time period, predict the mobile terminal target position and the mesh
The probability value that cursor position occurs recommends the products kimonos such as mobile finance of respective target locations further according to prediction result to user
Business.
Detailed description of the invention
Fig. 1 is the terminal structure schematic diagram for the hardware running environment that the embodiment of the present invention is related to;
Fig. 2 is the flow diagram of position predicting method first embodiment of the present invention;
Fig. 3 is the flow diagram of position predicting method second embodiment of the present invention;
Fig. 4 is the flow diagram of position predicting method 3rd embodiment of the present invention;
Fig. 5 is the flow diagram of position predicting method fourth embodiment of the present invention;
Fig. 6 is the flow diagram of the 5th embodiment of position predicting method of the present invention;
Fig. 7 is the flow diagram of position predicting method sixth embodiment of the present invention;
Fig. 8 is the flow diagram of the 7th embodiment of position predicting method of the present invention;
Fig. 9 is the flow diagram of the 8th embodiment of position predicting method of the present invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
The primary solutions of the embodiment of the present invention are:
The applications such as existing mobile finance, shopping software class are due to user's usage frequency and sticky lower, the movement of acquisition
Corresponding position data all has apparent dispersibility on Spatial Dimension and time dimension in terminal, it is difficult to directly construct to it
Not yet there is the implementation of the application user location prediction such as mobile finance, shopping software at present in prediction model.
The mobile terminal locations number that the present invention can disperse in receiving user's usage frequency and the lower application of viscosity
According to when, construct prediction model, based on mobile terminal in line position and the second preset time period, predict the mesh of the mobile terminal
The probability value that cursor position and the target position occur, recommends the shifting of respective target locations further according to prediction result to user
The products & services such as dynamic finance.
As shown in Figure 1, Fig. 1 is the terminal structure schematic diagram for the hardware running environment that the embodiment of the present invention is related to.
The terminal of that embodiment of the invention can be PC, be also possible to smart phone, tablet computer, MP4 (Moving Picture
Experts Group Audio Layer IV, dynamic image expert's compression standard audio level 3) player, portable computer
Etc. packaged type terminal device having a display function.
As shown in Figure 1, the terminal may include: processor 1001, such as CPU, network interface 1004, user interface
1003, memory 1005, communication bus 1002.Wherein, communication bus 1002 is for realizing the connection communication between these components.
User interface 1003 may include display screen (Display), input unit such as keyboard (Keyboard), optional user interface
1003 can also include standard wireline interface and wireless interface.Network interface 1004 optionally may include that the wired of standard connects
Mouth, wireless interface (such as WI-FI interface).Memory 1005 can be high speed RAM memory, be also possible to stable memory
(non-volatile memory), such as magnetic disk storage.Memory 1005 optionally can also be independently of aforementioned processor
1001 storage device.
Preferably, terminal can also include camera, RF (Radio Frequency, radio frequency) circuit, sensor, audio
Circuit, WiFi module etc..Wherein, sensor such as optical sensor, motion sensor and other sensors.Specifically, light
Sensor may include ambient light sensor and proximity sensor, wherein ambient light sensor can according to the light and shade of ambient light come
The brightness of display screen is adjusted, proximity sensor can close display screen and/or backlight when mobile terminal is moved in one's ear.As
One kind of motion sensor, gravity accelerometer can detect the size of (generally three axis) acceleration in all directions, quiet
Size and the direction that can detect that gravity when only, the application that can be used to identify mobile terminal posture are (such as horizontal/vertical screen switching, related
Game, magnetometer pose calibrating), Vibration identification correlation function (such as pedometer, tap) etc.;Certainly, mobile terminal can also match
The other sensors such as gyroscope, barometer, hygrometer, thermometer, infrared sensor are set, details are not described herein.
It will be understood by those skilled in the art that the restriction of the not structure paired terminal of terminal structure shown in Fig. 1, can wrap
It includes than illustrating more or fewer components, perhaps combines certain components or different component layouts.
As shown in Figure 1, as may include that operating system, network are logical in a kind of memory 1005 of computer storage medium
Believe module, Subscriber Interface Module SIM and user location Prediction program.
In terminal shown in Fig. 1, network interface 1004 is mainly used for connecting background server, carries out with background server
Data communication;User interface 1003 is mainly used for connecting client (user terminal), carries out data communication with client;And processor
1001 can be used for calling the user location Prediction program stored in memory 1005, and execute following operation:
Obtain the position data in corresponding first preset time period of mobile terminal;
Position data is clustered, and is formed according to the cluster and stops point data;
Prediction model is constructed according to point data is stopped, the target position of the mobile terminal is predicted by prediction model, with
And the corresponding destination probability value in the target position.
Further, processor 1001 can call the user location Prediction program stored in memory 1005, also execute
It operates below: it is described according to point data building prediction model is stopped, the target position of the mobile terminal is predicted by prediction model
It sets and the step of the corresponding destination probability value in the target position, comprising:
The first prediction model is constructed according to point data is stopped, the mobile terminal is predicted by first prediction model
The corresponding first object probability value in first object position and the first object position.
Further, processor 1001 can call the user location Prediction program stored in memory 1005, also execute
It operates below: constructing the first prediction model according to point data is stopped, the mobile terminal is predicted by first prediction model
First object position and the step of the corresponding first object probability value in the first object position, comprising:
Corresponding stop point data of each period is determined according to the stop point data, wherein the period includes
Workaday working hour, the workaday rest period, the working hour on day off and day off the rest period;
It is derived with Bayes' theorem and calculates the corresponding probability for stopping point data of each period;
The first prediction model, first prediction are constructed according to the corresponding probability for stopping point data of each period
Model includes the probability distribution for stopping point data and corresponding period;
Based on the probability distribution and the second preset time period for stopping point data, by described in first prediction model prediction
The corresponding first object probability value in the first object position of mobile terminal and the first object position.
Further, processor 1001 can call the user location Prediction program stored in memory 1005, also execute
It operates below: it is described according to point data building prediction model is stopped, the target position of the mobile terminal is predicted by prediction model
It sets and the step of the corresponding destination probability value in the target position, comprising:
The second prediction model is constructed according to point data is stopped, the mobile terminal is predicted by second prediction model
The second destination probability value that second target position and second target position occur.
Further, processor 1001 can call the user location Prediction program stored in memory 1005, also execute
It operates below: constructing the second prediction model according to point data is stopped, the mobile terminal is predicted by second prediction model
The second target position and the step of the corresponding second destination probability value in second target position, comprising:
It is ranked up according to chronological order to point data is stopped;
Point data, which will be stopped, according to time window and sliding step is divided into multiple dwell point subsequences along time shaft;
Based on the minimum threshold frequency of dwell point subsequence, with PrefixSpan Sequential Pattern Mining Algorithm, excavation stops
Stationary point trajectory model constructs the second prediction model, and second prediction model includes all dwell point trajectory model and its general
Rate;
Based on dwell point trajectory model and in wire position data, matched and searched is carried out by the second prediction model, predicts institute
State the corresponding second destination probability value in the second target position and second target position of mobile terminal.
Further, processor 1001 can call the user location Prediction program stored in memory 1005, also execute
It operates below: position data is clustered, and the step of stopping point data is formed according to the cluster, comprising:
Determine the corresponding central point of the position data, and the position that preset threshold will be less than or equal at a distance from central point
It sets data to cluster as target, calculates the stop central point of target cluster;
Using the rest position data in addition to target clusters corresponding position data as the position data, continue to execute
The step of determining the position data corresponding central point;
When all position datas complete cluster operation, formed according to the stop central point of target cluster and target cluster
Stop point data.
Further, processor 1001 can call the user location Prediction program stored in memory 1005, also execute
It operates below: position data is clustered, and formed before the step of stopping point data according to the cluster, further includes:
The position data is cleaned, to remove the repeatable position data in preset duration, and filtering unauthorized and network
Postpone the illegal position data generated.
Further, processor 1001 can call the user location Prediction program stored in memory 1005, also execute
It operates below: it is described according to point data building prediction model is stopped, the target position of the mobile terminal is predicted by prediction model
Set and the step of the corresponding destination probability value in the target position after, further includes:
Target position is ranked up according to the destination probability value size order that target position occurs;
According to the sequence of target position to user's recommended products and/or service.
Based on above-mentioned hardware configuration, embodiment of the present invention method is proposed.
It is the flow diagram of position predicting method first embodiment of the present invention, the position prediction side referring to Fig. 2, Fig. 2
Method includes:
Step S10 obtains the position data in corresponding first preset time period of mobile terminal;
Starting is in application, the wireless location technology of mobile terminal adopts the position data of user in the terminal
Collection, position data includes device number, login position, login time data, in the positional number for receiving the mobile terminal upload
According to when, the position data is subjected to operation storage according to Hadoop distribution, Hadoop can be distributed mass data
Formula processing has high reliability, high scalability, high efficiency and high fault tolerance to the processing of data, obtains what mobile terminal uploaded
Position data in first preset time period.
Step S20, clusters position data, and is formed according to the cluster and stop point data;
The case where position data of acquisition is the single record of dispersion, corresponds to same position region there are a plurality of record, directly
It connects and constructs prediction model in the position data of dispersion, on the one hand will increase the complexity of algorithm, it on the other hand can be due to position
The potential Move Mode of mobile terminal is excessively dispersed and can not be excavated to data.Therefore it before constructing prediction model, needs pair
Position data in first preset time period is clustered, and is formed according to the cluster and stopped point data.
Step S30 constructs prediction model according to point data is stopped, the target of the mobile terminal is predicted by prediction model
The corresponding destination probability value in position and the target position.
The period is divided, the stop that each period occurs is derived by with Bayes' theorem derivation formula and counts
According to probability distribution, building prediction model include stop point data probability distribution and its corresponding period, be based on dwell point
The probability distribution of data and the second preset time period predict the first object position of the mobile terminal in the second preset time period
It sets and the probability value of first object position.
It is ranked up to point data is stopped according to chronological order, point data will be stopped and divide multiple stop idea sequences
Column, based on the minimum threshold frequency of dwell point subsequence, with PrefixSpan Sequential Pattern Mining Algorithm, building prediction mould
Type carries out matched and searched by prediction model, predicts the mobile terminal based on dwell point trajectory model and in wire position data
The second target position and second target position occur the second destination probability value, i.e., based on the trajectory model of dwell point
With current in wire position data, next target position that mobile terminal is likely to occur is predicted.
It further, is the flow diagram of position predicting method second embodiment of the present invention referring to Fig. 3, Fig. 3.Based on upper
The step of embodiment stated, S30, comprising:
Step S31 constructs the first prediction model according to point data is stopped, predicts the shifting by first prediction model
The corresponding first object probability value in first object position and the first object position of dynamic terminal.
The period is divided, corresponding stop point data of each period is determined according to the stop point data, with pattra leaves
This Theorem deduction derivation of equation obtains the probability distribution for the stop point data that each period occurs, and constructs prediction model packet
The probability distribution for stopping point data and its corresponding period are included, when default based on the probability distribution and its second for stopping point data
Between section, predict that the first object position and first object position of the mobile terminal in the second preset time period are corresponding general
Rate value.
It further, is the flow diagram of position predicting method 3rd embodiment of the present invention referring to Fig. 4, Fig. 4.Based on upper
The step of embodiment stated, S31, comprising:
Step S311 determines corresponding stop point data of each period according to the stop point data, wherein when described
Between section when including the rest on workaday working hour, workaday rest period, the working hour on day off and day off
Section;
Corresponding stop point data of each period is determined according to the stop point data, and the period includes workaday work
Period, the workaday rest period, the working hour on day off and day off four periods of rest period, for example, root
According to the work of user and time of having a rest rule, week age is divided into workaday working hour (Mon-Fri, 8:00-
18:00), workaday rest period (Mon-Fri, 0:00-8:00,18:00-24:00), (the week working hour on day off
End, 8:00-18:00), rest period on day off, (weekend, 0:00-8:00,18:00-24:00 determined that each period is corresponding
Stop point data.
Step S312 carries out deriving the general of the stop point data for calculating each period appearance with Bayes' theorem
Rate;
Bayes' theorem derivation formula can be with are as follows:
P (S | T)=P (T | S) * P (S)/P (T),
P (T)=current time segment length/total length of all periods
P (S)=stop point data S summary journal quantity/all summary journal quantity for stopping point data
P (T | S)=stop record quantity/stop point data S summary journal quantity of the point data S in period T
P (S | T) is the probability of the stop point data S occurred in period T;
The probability for the stop point data that each period occurs is calculated by Bayes' theorem derivation formula, it is assumed that
Occur stop point data S1, S2 in workaday working hour, then passes through formula P (S | T)=P (T | S) * P (S)/P (T) meter
Calculate the probability of dwell point S1 and S2 working hour on weekdays;Occur stopping in rest period on weekdays point data S1 and
S3 is then calculated by formula P (S | T)=P (T | S) * P (S)/P (T) general in the rest period of dwell point S1 and S3 on weekdays
Rate.
Step S313, according to corresponding probability distribution building the first prediction mould for stopping point data of each period
Type, first prediction model include the probability distribution for stopping point data and its corresponding period;
The first prediction model is constructed according to the probability of each stop point data of derivation and corresponding period, passes through the first mould
Type carries out the prediction of target position and the correspondence probability value of target position.
Step S314 passes through the first prediction mould based on the probability distribution and the second preset time period for stopping point data
Type predicts the corresponding first object probability value in first object position and the first object position of the mobile terminal.
According to the second preset time period, based on the dwell point data distribution in prediction model and the period is corresponded to, progress
With lookup, the first object position being likely to occur in the second preset time period, and the corresponding first object position are found out
Corresponding first object probability value is set, and by multiple first object positions of lookup according to corresponding probability value size order
It is ranked up.
It further, is the flow diagram of position predicting method fourth embodiment of the present invention referring to Fig. 5, Fig. 5.Based on upper
The step of embodiment stated, S30, comprising:
Step S32 constructs the second prediction model according to point data is stopped, predicts the shifting by second prediction model
The corresponding second destination probability value in the second target position and second target position of dynamic terminal.
It is ranked up to point data is stopped according to chronological order, point data will be stopped and divide multiple stop idea sequences
Column, based on the minimum threshold frequency of dwell point subsequence, with PrefixSpan Sequential Pattern Mining Algorithm, building prediction mould
Type carries out matched and searched by prediction model based on dwell point trajectory model and its in wire position data, predicts described mobile whole
The corresponding second destination probability value in second target position and second target position at end, i.e. the track mould based on dwell point
Formula, in wire position data, predicts next target position that mobile terminal is likely to occur with current.
It further, is the flow diagram of the 5th embodiment of position predicting method of the present invention referring to Fig. 6, Fig. 6.Based on upper
The step of embodiment stated, S32, comprising:
Step S321 is ranked up according to chronological order to point data is stopped;
It is ranked up according to the sequencing of time, for example, stopping point data is C1 (20180901 10:11:11);C1
(20180902 10:12:12);C1(20180903 10:12:12);C1(20180904 10:13:13);C2(20180901
16:12:12);C2(20180902 16:11:11);C2(20180903 17:12:12);C2(20180904 16:15:15);
C2(20180905 20:20:20);It is after sequence C1 (20180901 10:11:11);C2(20180901 16:12:12);C1
(20180902 10:12:12);C2(20180902 16:11:11);C1(20180903 10:12:12);C2(20180903
17:12:12);C1(20180904 10:13:13);C2(20180904 16:15:15);C2(20180905 20:20:20).
Step S322 will stop point data according to time window and sliding step and be divided into multiple stop ideas along time shaft
Sequence;
Before carrying out trajectory model excavation and establishing the second prediction model, it is based on time window and sliding step, according to
Time window and sliding step are divided into multiple dwell point subsequences for point data is stopped, the input as algorithm.Use sliding
The mode of time window is divided into multiple dwell point subsequences for point data is stopped, and the stop point data of mobile terminal is mapped to
On time shaft, multiple dwell point subsequences are divided into using time window, it is 24 hours that time window size, which is such as arranged, when
Between gap size (sliding step) between window be 24 hours, point data will be stopped and be divided into following multiple stop idea sequences
Column: Z1 { C1 (20180901 10:11:11), C2 (20180901 16:12:12) };Z2 C1 (20180902 10:12:12),
C2(20180902 16:11:11)};Z3 { C1 (20180903 10:12:12), C2 (20180903 17:12:12) };Z4{C1
(20180904 10:13:13), C2 (20180904 16:15:15) };Z5{C2(20180905 20:20:20)}.
Step S323 is calculated based on the minimum threshold frequency of dwell point subsequence with PrefixSpan sequential mode mining
Method excavates dwell point trajectory model and constructs the second prediction model, and second prediction model includes all stop locus of points moulds
Formula and its probability;
Based on the minimum threshold frequency of dwell point subsequence, with the PrefixSpan sequential mode mining based on prefix trees
Algorithm excavates the trajectory model of user, and PrefixSpan algorithm generates candidate sequence, and data for projection library due to not having to
(the sequence suffix set about a certain prefix) reduces quickly, and memory consumption is more stable, so having preferable digging efficiency.
The process for excavating user trajectory mode using PrefixSapn algorithm is as shown in table 1, inputs dwell point arrangement set
S finds out the prefix and corresponding data for projection library that length is 1 based on the minimum threshold frequency of dwell point subsequence from S;It saves
All frequent 1 sequence sets for meeting minimum threshold frequency;By the corresponding sequence of prefix for being unsatisfactory for minimum threshold frequency from S
It deletes;The prefix for meeting minimum threshold is added in frequent 1 sequence sets;Recursive calculation Frequent episodes collection F;Mining Frequent sequence
Recursive function, prefixs is prefix sets;Find out the corresponding data for projection library of prefix;If data for projection library is sky, recurrence
It returns;If all items are all unsatisfactory for minimum threshold frequency in data for projection library, recurrence is returned;Minimum threshold frequency will be met
Each item obtains new prefix sets in conjunction with current prefix;New prefix sets are added in Set of Frequent Sequential Patterns;To new
Prefix sets carry out recursive calculation.
Table 1
Sequence pattern is arranged in input of the ready-portioned dwell point subsequence as Sequential Pattern Mining Algorithm prefixspan
Minimum threshold frequency (minimum support), the second prediction model of mining track mode construction, second prediction model includes
All dwell point trajectory models and its probability.
Step S324 carries out matching by the second prediction model and looks into based on dwell point trajectory model and in wire position data
It looks for, predicts the corresponding second destination probability value in the second target position and second target position of the mobile terminal.
Based on the dwell point trajectory model excavated and in wire position data, pattern match is carried out by the second prediction model
It searches, the second target position is predicted, for example, the minimum threshold frequency being arranged in step S323 is 0.5, excavating equipment pair
The trajectory model answered is C1- > C2, is C1 dwell point in wire position data, the second target position of prediction is C2.
It further, is the flow diagram of position predicting method sixth embodiment of the present invention referring to Fig. 7, Fig. 7.Based on upper
Embodiment shown in Fig. 2 is stated, step S20 may include:
Step S21 determines the corresponding central point of the position data, and default by being less than or equal at a distance from central point
The position data of threshold value is clustered as target, calculates the stop central point of target cluster;
It will determine the corresponding central point of position data of the first preset time period, traverse other position datas, it will be in
The position data that the distance of heart point is less than or equal to preset threshold is clustered as target, and calculates the stop center of target cluster
Point.
Step S22, using the rest position data in addition to target clusters corresponding position data as the position data,
Continue to execute the step of determining the position data corresponding central point;
Using the rest position data in addition to target clusters corresponding position data as the position data, determine described in
The central point of position data continues target cluster, calculates the stop central point of target cluster.
Step S23, when all position datas complete cluster operation, according in the stop of target cluster and target cluster
Heart point forms dwell point data.
When all position datas of the first preset time period complete cluster operation, clustered according to target cluster and target
Stop central point formed stop point data, such as: obtain mobile terminal in corresponding first preset time period position data
Are as follows: R1 (20180901 10:11:11);R2(20180902 10:12:12);R3(20180903 10:12:12);R4
(20180904 10:13:13);R5(20180901 16:12:12);R6(20180902 16:11:11);R7(20180903
17:12:12);R8(20180904 16:15:15);R9(20180905 20:20:20);
After completing the position data cluster operation in all first preset time periods, two target clusters of P1, P2 are obtained,
Middle P1 target cluster includes that position records R1, R2, R3, R4, is to stop central point by calculating C1, P2 target cluster includes position
R5, R6, R7, R8, R9 are recorded, is to stop central point by calculating C2;
It is stopped in central point and P2 target cluster and its corresponding C2 stop according to P1 target cluster and its corresponding C1
Heart point forms dwell point data: C1 (20180901 10:11:11);C1(20180902 10:12:12);C1(20180903
10:12:12);C1(20180904 10:13:13);C2(20180901 16:12:12);C2(20180902 16:11:11);
C2(20180903 17:12:12);C2(20180904 16:15:15);C2(20180905 20:20:20);
The algorithm flow of the position data cluster of user is as shown in table 2, inputs the distributed locations data P of user, and setting is pre-
If threshold value, position data each in P is handled;Determine whether to find the position data for meeting preset threshold;Traversal is current
Stop each position data in point data;Distance of the calculating position data to dwell data center;It is default if it is less than being equal to
Threshold value, position data P, which is added to, to stop in point data SP;Recalculate the center for stopping point data SP;Find the default threshold of satisfaction
The position data of value, isFind are set as true;Interior loop is jumped out, if not finding the position data for meeting preset threshold;
S is added as a new stop point data in position data P, P is as the new central point for stopping point data.
Table 2
It further, is the flow diagram of the 7th embodiment of position predicting method of the present invention referring to Fig. 8, Fig. 8.Based on upper
The embodiment stated, before S20 step, further includes:
Step S40 cleans the position data, and to remove the repeatable position data in preset duration, and filtering is not awarded
The illegal position data that power and network delay generate.
Position data is cleaned, including removes, filter, the position of application program in the terminal to user
In data acquisition, duplicate position data can be generated because of application is repeatedly started in same position in user's short time,
The position data by redundancy in this part preset duration is needed to be removed place when getting the corresponding position data of mobile terminal
Reason, in addition to this, also comprising the illegal position number due to equipment unauthorized and network delay generation in the position data got
According to, need this part invalid data being filtered processing, make establish prediction model it is more accurate.
It further, is the flow diagram of the 8th embodiment of position predicting method of the present invention referring to Fig. 9, Fig. 9.Based on upper
After the step of embodiment stated, S30, further includes:
Step S50 is ranked up target position according to the destination probability value size order that target position occurs;
Based on user's dwell point trajectory model, in conjunction with user currently in wire position data, to the progress of user trajectory mode
Match, predicts the next target position of user, i.e. the second target position, to multiple next target positions of prediction according to corresponding rail
The probability value of mark mode is ranked up from high to low;
Based on user's dwell point probability distribution mode and corresponding period, second time period is preset, prediction user is default
First object position locating for second time period, to multiple first object positions of prediction according to target position probability from height to
It is low to be ranked up.
Step S60, according to the sequence of target position to user's recommended products and/or service.
It takes the first object position of default top n prediction or takes default preceding M first object position, according to target position
Value sort and recommend products and/or the service such as finance, such as coupons push to user.
The present invention also provides a kind of user location prediction meanss.
User location prediction meanss of the present invention include: memory, processor and are stored on the memory and can be in institute
The user location Prediction program run on processor is stated, is realized such as when the user location Prediction program is executed by the processor
Above the step of position predicting method.
Wherein, the user location Prediction program run on the processor, which is performed realized method, can refer to this
The each embodiment of invention position predicting method, details are not described herein again.
The present invention also provides a kind of readable storage medium storing program for executing.
User location Prediction program is stored on readable storage medium storing program for executing of the present invention, the user location Prediction program is processed
The step of device realizes position predicting method as described above when executing.
Wherein, the user location Prediction program run on the processor, which is performed realized method, can refer to this
The each embodiment of invention position predicting method, details are not described herein again.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the system that include a sequence element not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or system institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or system.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art
The part contributed out can be embodied in the form of software products, which is stored in one as described above
In storage medium (such as ROM/RAM, magnetic disk, CD), including some instructions are used so that terminal device (it can be mobile phone,
Computer, server, air conditioner or network equipment etc.) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of position predicting method, which is characterized in that the position predicting method includes the following steps:
Obtain the position data in corresponding first preset time period of mobile terminal;
Position data is clustered, and is formed according to the cluster and stops point data;
Prediction model is constructed according to point data is stopped, the target position of the mobile terminal, Yi Jisuo are predicted by prediction model
State the corresponding destination probability value in target position.
2. position predicting method as described in claim 1, which is characterized in that described according to stop point data building prediction mould
Type predicts the corresponding destination probability value in the target position of the mobile terminal and the target position by prediction model
Step, comprising:
The first prediction model is constructed according to point data is stopped, predicts the first of the mobile terminal by first prediction model
The corresponding first object probability value in target position and the first object position.
3. position predicting method as claimed in claim 2, which is characterized in that predict mould according to point data building first is stopped
Type predicts the first object position and the first object position pair of the mobile terminal by first prediction model
The step of first object probability value answered, comprising:
Corresponding stop point data of each period is determined according to the stop point data, wherein the period includes work
Day working hour, the workaday rest period, the working hour on day off and day off the rest period;
It is derived with Bayes' theorem and calculates the corresponding probability for stopping point data of each period;
The first prediction model, first prediction model are constructed according to the corresponding probability for stopping point data of each period
Including stopping the probability distribution of point data and corresponding to the period;
Based on the probability distribution and the second preset time period for stopping point data, the movement is predicted by first prediction model
The corresponding first object probability value in the first object position of terminal and the first object position.
4. position predicting method as described in claim 1, which is characterized in that described according to stop point data building prediction mould
Type predicts the corresponding destination probability value in the target position of the mobile terminal and the target position by prediction model
Step, comprising:
The second prediction model is constructed according to point data is stopped, predicts the second of the mobile terminal by second prediction model
The second destination probability value that target position and second target position occur.
5. position predicting method as claimed in claim 4, which is characterized in that predict mould according to point data building second is stopped
Type predicts the second target position and second target position pair of the mobile terminal by second prediction model
The step of the second destination probability value answered, comprising:
It is ranked up according to chronological order to point data is stopped;
Point data, which will be stopped, according to time window and sliding step is divided into multiple dwell point subsequences along time shaft;
Dwell point is excavated with PrefixSpan Sequential Pattern Mining Algorithm based on the minimum threshold frequency of dwell point subsequence
Trajectory model constructs the second prediction model, and second prediction model includes all dwell point trajectory model and its probability;
Based on dwell point trajectory model and in wire position data, matched and searched is carried out by the second prediction model, predicts the shifting
The corresponding second destination probability value in the second target position and second target position of dynamic terminal.
6. position predicting method as described in any one in claim 1-5, which is characterized in that position data is clustered, and
The step of stopping point data is formed according to the cluster, comprising:
Determine the corresponding central point of the position data, and the positional number that preset threshold will be less than or equal at a distance from central point
It is clustered according to as target, calculates the stop central point of target cluster;
Using the rest position data in addition to target clusters corresponding position data as the position data, determination is continued to execute
The step of position data corresponding central point;
When all position datas complete cluster operation, is formed and stopped according to the stop central point of target cluster and target cluster
Point data.
7. position predicting method as claimed in claim 6, which is characterized in that clustered to position data, and according to described
Cluster was formed before the step of stopping point data, further includes:
The position data is cleaned, to remove the repeatable position data in preset duration, and filtering unauthorized and network delay
The illegal position data of generation.
8. position predicting method as claimed in claim 7, which is characterized in that described according to stop point data building prediction mould
Type predicts the corresponding destination probability value in the target position of the mobile terminal and the target position by prediction model
After step, further includes:
Target position is ranked up according to the destination probability value size order that target position occurs;
According to the sequence of target position to user's recommended products and/or service.
9. a kind of user location prediction meanss, which is characterized in that the user location prediction meanss include: memory, processor
And it is stored in the user location Prediction program that can be run on the memory and on the processor, the user location prediction
It realizes when program is executed by the processor such as the step of position predicting method described in any item of the claim 1 to 8.
10. a kind of readable storage medium storing program for executing, which is characterized in that user location Prediction program is stored on the readable storage medium storing program for executing,
Such as position prediction side described in any item of the claim 1 to 8 is realized when the user location Prediction program is executed by processor
The step of method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811212184.2A CN109992726A (en) | 2018-10-17 | 2018-10-17 | Position predicting method, device and readable storage medium storing program for executing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811212184.2A CN109992726A (en) | 2018-10-17 | 2018-10-17 | Position predicting method, device and readable storage medium storing program for executing |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109992726A true CN109992726A (en) | 2019-07-09 |
Family
ID=67128265
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811212184.2A Pending CN109992726A (en) | 2018-10-17 | 2018-10-17 | Position predicting method, device and readable storage medium storing program for executing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109992726A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110909592A (en) * | 2019-10-11 | 2020-03-24 | 重庆特斯联智慧科技股份有限公司 | Target tracking method and system based on multi-scale characteristic quantity |
CN111107319A (en) * | 2019-12-25 | 2020-05-05 | 眸芯科技(上海)有限公司 | Target tracking method, device and system based on regional camera |
CN111291092A (en) * | 2020-02-14 | 2020-06-16 | 腾讯科技(深圳)有限公司 | Data processing method, device, server and storage medium |
CN111459162A (en) * | 2020-04-07 | 2020-07-28 | 珠海格力电器股份有限公司 | Standby position planning method and device, storage medium and computer equipment |
CN112541134A (en) * | 2020-12-07 | 2021-03-23 | 韩珍 | Sequence position recommendation method based on geographical perception |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102289461A (en) * | 2010-07-07 | 2011-12-21 | 微软公司 | Recommendations and targeted advertising based upon directions requesting activity and data |
CN102509170A (en) * | 2011-10-10 | 2012-06-20 | 浙江鸿程计算机系统有限公司 | Location prediction system and method based on historical track data mining |
CN104408203A (en) * | 2014-12-18 | 2015-03-11 | 西安电子科技大学宁波信息技术研究院 | Method for predicting path destination of moving object |
CN104462190A (en) * | 2014-10-24 | 2015-03-25 | 中国电子科技集团公司第二十八研究所 | On-line position prediction method based on mass of space trajectory excavation |
CN104931041A (en) * | 2015-05-03 | 2015-09-23 | 西北工业大学 | Method for predicting place sequence based on user track data |
CN107402931A (en) * | 2016-05-19 | 2017-11-28 | 滴滴(中国)科技有限公司 | Recommend method and apparatus to a kind of trip purpose |
-
2018
- 2018-10-17 CN CN201811212184.2A patent/CN109992726A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102289461A (en) * | 2010-07-07 | 2011-12-21 | 微软公司 | Recommendations and targeted advertising based upon directions requesting activity and data |
CN102509170A (en) * | 2011-10-10 | 2012-06-20 | 浙江鸿程计算机系统有限公司 | Location prediction system and method based on historical track data mining |
CN104462190A (en) * | 2014-10-24 | 2015-03-25 | 中国电子科技集团公司第二十八研究所 | On-line position prediction method based on mass of space trajectory excavation |
CN104408203A (en) * | 2014-12-18 | 2015-03-11 | 西安电子科技大学宁波信息技术研究院 | Method for predicting path destination of moving object |
CN104931041A (en) * | 2015-05-03 | 2015-09-23 | 西北工业大学 | Method for predicting place sequence based on user track data |
CN107402931A (en) * | 2016-05-19 | 2017-11-28 | 滴滴(中国)科技有限公司 | Recommend method and apparatus to a kind of trip purpose |
Non-Patent Citations (1)
Title |
---|
叶谦: "《基于路径模式挖掘的个人连续路径预测》", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110909592A (en) * | 2019-10-11 | 2020-03-24 | 重庆特斯联智慧科技股份有限公司 | Target tracking method and system based on multi-scale characteristic quantity |
CN111107319A (en) * | 2019-12-25 | 2020-05-05 | 眸芯科技(上海)有限公司 | Target tracking method, device and system based on regional camera |
CN111107319B (en) * | 2019-12-25 | 2021-05-28 | 眸芯科技(上海)有限公司 | Target tracking method, device and system based on regional camera |
CN111291092A (en) * | 2020-02-14 | 2020-06-16 | 腾讯科技(深圳)有限公司 | Data processing method, device, server and storage medium |
CN111459162A (en) * | 2020-04-07 | 2020-07-28 | 珠海格力电器股份有限公司 | Standby position planning method and device, storage medium and computer equipment |
CN111459162B (en) * | 2020-04-07 | 2021-11-16 | 珠海格力电器股份有限公司 | Standby position planning method and device, storage medium and computer equipment |
CN112541134A (en) * | 2020-12-07 | 2021-03-23 | 韩珍 | Sequence position recommendation method based on geographical perception |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109992726A (en) | Position predicting method, device and readable storage medium storing program for executing | |
CN109598777B (en) | Image rendering method, device and equipment and storage medium | |
Conti et al. | Looking ahead in pervasive computing: Challenges and opportunities in the era of cyber–physical convergence | |
US20130009994A1 (en) | Methods and apparatus to generate virtual-world environments | |
Picco et al. | Software engineering for mobility: reflecting on the past, peering into the future | |
CN108304758A (en) | Facial features tracking method and device | |
CN104584601A (en) | Discovery method and apparatuses and system for discovery | |
US9224100B1 (en) | Method and apparatus using accelerometer data to serve better ads | |
CN107450841B (en) | Interactive object control method and device | |
US20090167768A1 (en) | Selective frame rate display of a 3D object | |
WO2013062237A1 (en) | System and method for managing social relationship information | |
CN110351662B (en) | Method, platform and device for end cloud cooperation | |
CN106157602A (en) | The method and apparatus of calling vehicle | |
US11654372B2 (en) | Methods, systems, and devices for identifying a portion of video content from a video game for a player or spectator | |
CN112328911B (en) | Place recommending method, device, equipment and storage medium | |
CN108074009A (en) | Motion route generation method and device, mobile terminal and server | |
CN105955715A (en) | Information processing method, device and intelligent terminal | |
CN106598222A (en) | Scene mode switching method and system | |
CN110494194A (en) | Object control system, program and method in location game | |
CN104809416A (en) | display screen shielding method, electronic device and computer program product | |
CN110213591A (en) | A kind of video motion estimating method, device and storage medium | |
Ruan et al. | Wireless sensor network deployment in mobile phones assisted environment | |
CN111797867A (en) | System resource optimization method and device, storage medium and electronic equipment | |
CN108829595A (en) | Test method, device, storage medium and electronic equipment | |
CN107832848A (en) | application management method, device, storage medium and electronic equipment |
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 |