CN102509170A - Location prediction system and method based on historical track data mining - Google Patents

Location prediction system and method based on historical track data mining Download PDF

Info

Publication number
CN102509170A
CN102509170A CN201110308289XA CN201110308289A CN102509170A CN 102509170 A CN102509170 A CN 102509170A CN 201110308289X A CN201110308289X A CN 201110308289XA CN 201110308289 A CN201110308289 A CN 201110308289A CN 102509170 A CN102509170 A CN 102509170A
Authority
CN
China
Prior art keywords
track data
module
path
server
data
Prior art date
Application number
CN201110308289XA
Other languages
Chinese (zh)
Inventor
潘士渠
陈岭
吕明琪
赵江奇
Original Assignee
浙江鸿程计算机系统有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 浙江鸿程计算机系统有限公司 filed Critical 浙江鸿程计算机系统有限公司
Priority to CN201110308289XA priority Critical patent/CN102509170A/en
Publication of CN102509170A publication Critical patent/CN102509170A/en

Links

Abstract

The invention relates to the technical field of data mining, in particular to a location prediction system and a method based on historical track data mining. In the method, the location prediction problem is decomposed into two sub-problems of offline track mining and online mode matching, wherein the offline track mining mainly adopts the data mining technology to process and analyze the historical track data of a user; and the online mode matching is mainly based on the online motion conditions of the user and the mined motion mode to perform matching and searching. According to the invention, various needs based on the location-based services (such as location-based advertisement, recommendation, reminding and other services) are effectively met, and the problems of the prior art are solved.

Description

A kind of position prediction system and method based on the historical track data mining

Technical field

The present invention relates to the data mining technology field, be specifically related to a kind of position prediction system and method based on the historical track data mining.

Background technology

The position prediction technology can be location based services (LBS, location-based service) and provides support, like intelligent navigation, advertisement putting, service reservation, place prompting etc.Position prediction to the user mainly concentrates under mobile network or the wireless network environment at present; User's motion conditions in all kinds of radiobeacon (like mobile base station, WiFi node, the RFID node etc.) scope in environment is analyzed; Obtain the user movement rule, and then carry out position prediction.For based on the position prediction of mobile base station technology, the mobile base station bearing accuracy is lower, the location-based service requirement of the exact position that is difficult to satisfy the demand; For based on the position prediction of radio node technology, need in environment, dispose a large amount of radio nodes, and the position of radio node or be unknown, or need carry out a large amount of positions marks work.

At present to the reckoning of user's short-term exact position and prediction mainly based on the combination of GPS+ dead reckoning.Calculate the position when yet these class methods are generally used for the GPS inefficacy; Can only predict the position of user in the short time; The GPS out-of-service time is in a single day long, and then forecasting efficiency reduces greatly, and this method needs hardware devices such as high-precision gyroscope, acceleration transducer, angular-rate sensor; Cost is higher, is difficult on mobile devices such as mobile phone, promote.

Summary of the invention

The present invention overcomes above-mentioned weak point; Purpose is to provide a kind of position prediction system based on the historical track data mining; This system may operate on the mobile device; For Location-Based Services provides user's destination and following routing information, effectively satisfy the demand of all kinds of Location-Based Services (like services such as location-based advertisement, recommendation, promptings), solved the problem that exists in the prior art.

Another object of the present invention is to provide a kind of position predicting method that excavates based on track; This method is excavated the position prediction PROBLEM DECOMPOSITION and two subproblems of line model coupling for the off-line track; The off-line track excavates the main data mining technology of using user's historical track data is handled and analyzed; The line model coupling is mainly mated based on online motion conditions of user and the motor pattern of excavating and is searched; Effectively satisfy the demand of all kinds of Location-Based Services (like services such as location-based advertisement, recommendation, promptings), solved the problem that exists in the prior art.

The present invention achieves the above object through following technical scheme: a kind of position prediction system based on the historical track data mining; Comprise mobile client and server end, client is installed on the mobile device, comprises collection of GPS track data and pre-processing module; Motor pattern excavates preparation module; Online position prediction module, four modules of client communication module, the GPS track data is gathered and pre-processing module is responsible for gathering the GPS location point with certain SF; Be recorded as the GPS track data, and track data is cleaned and cuts apart pre-service; Motor pattern excavates preparation module and is responsible for candidate's Origin And Destination is extracted; Then track data being carried out abstract handles; Send Origin And Destination and abstract path data to server end through the client communication module at last, and call its motor pattern mining algorithm; Online position prediction module is responsible for the user movement pattern through excavating that server end returns is carried out modeling, and structural model is set, and real-time estimate is carried out in user's destination and following path; The client communication module is responsible for communicating with server end; Server end comprises that motor pattern excavates module and server end communication module; Motor pattern excavates module and is responsible for according to the time series mining algorithm path abstract data of client upload being excavated, and obtains the motor pattern collection of organizing according to starting point and endpoint data; The server end communication module is responsible for communicating with the client communication module, accepts the path abstract data of client, sends it to motor pattern and excavates module, and will excavate the result and return to client.

As preferably, the client and server end communicates through GPRS.

A kind of position predicting method based on the historical track data mining may further comprise the steps:

1) on user's mobile device, settles gps receiver, gather the GPS location point, be recorded as the GPS track data, and protect to preserve with certain SF;

2) the GPS track data is cleaned, the location point continuous to the time carries out cluster according to its speed, incorporates speed into same cluster less than the continuous position point of threshold value, and filters wherein irrational location point data;

3) according to cutting apart at interval the writing time of continuous position point, if then track is divided into two paths greater than threshold value at interval the writing time of continuous two location points;

4) to every paths along the time shaft forward with oppositely respectively location point is carried out clustering processing; Time is continuous and location point close together is incorporated same cluster into; The forward direction cluster of finding out the path and back are to cluster; Then forward direction cluster set and the cluster of back in cluster set are mated, cluster centre point distance then merges it less than threshold value becomes candidate's beginning or end;

5) appointed area is divided into the square net that the length of side equates, location point is replaced by the grid that comprises it, and the path is converted into the grid sequence data, then track data is carried out abstract and handle;

6) send Origin And Destination and abstract path data to server end through the client communication module and send the abstract path data to server; And call its motor pattern mining algorithm the user path abstract data that receives is excavated; Obtain the motor pattern collection organized to terminal according to starting point, and its form that is organized into file is returned;

7) utilization PrefixSpan sequential mode mining algorithm carries out modeling to motor pattern, the structural model tree, and scheme-tree comprises all motor patterns and adopts the probability of different starting points and terminal point;

8) based on user movement pattern of excavating and online exercise data thereof, through carrying out matched and searched in the scheme-tree, obtain lookup result, its destination and following path are predicted.

As preferably, the described gps receiver of step 1) is built-in GPS module of mobile device or the external GPS module that connects through bluetooth.

As preferably, step 1), step 2) and step 3) can synchronous processing.

As preferably, the described server end of step 6) receives only Origin And Destination and the abstract path data that do not comprise any particular geographic location information.

As preferably, step 7) is described scheme-tree result be retained in the Installed System Memory.

Beneficial effect of the present invention:

1) obtains its characteristics of motion through analysis, and carry out online position prediction, need not additional hardware equipment, not influenced by the environmental basis facility based on this to the historical GPS track data of user;

2) associated prediction is carried out in user's destination and following path, and the following path of user is predicted continuously, avoided the short shortcoming of path prediction length by destination information;

3) to the user movement rule, introduce motor pattern element continuity notion, expansion PrefixSpan algorithm is used for the motor pattern of digging user, is guaranteeing to have improved the length of motor pattern under the true and reliable prerequisite of motor pattern;

4) through motor pattern is set up suffix tree; In the branch of all motor patterns and suffix mode index to a shared prefix thereof; Make the follow-up mode coupling need not to adopt searching algorithm to search the coupling starting point, improved the speed of pattern match in the online position prediction.

Description of drawings

Fig. 1 is based on the position prediction system architecture synoptic diagram that track excavates;

Fig. 2 position prediction software general frame figure;

Fig. 3 motor pattern excavates prepares process flow diagram;

The online position prediction process flow diagram of Fig. 4;

Fig. 5 server process flow diagram.

Specific embodiment

Below in conjunction with accompanying drawing the present invention is done further explanation:

Embodiment 1: based on the position prediction system that track excavates, structure is as shown in Figure 1, comprises mobile client and server end; Client is installed on the mobile device; Comprise collection of GPS track data and pre-processing module, motor pattern excavates preparation module, online position prediction module; Four modules of client communication module; The GPS track data is gathered with pre-processing module and is responsible for being recorded as the GPS track data with certain SF collection GPS location point, and track data is cleaned and cuts apart pre-service; Motor pattern excavates preparation module and is responsible for candidate's Origin And Destination is extracted; Then track data being carried out abstract handles; Send Origin And Destination and abstract path data to server end through the client communication module at last, and call its motor pattern mining algorithm; Online position prediction module is responsible for the user movement pattern through excavating that server end returns is carried out modeling, and structural model is set, and real-time estimate is carried out in user's destination and following path; The client communication module is responsible for communicating with server end; Server end comprises that motor pattern excavates module and server end communication module; Motor pattern excavates module and is responsible for according to the time series mining algorithm path abstract data of client upload being excavated, and obtains the motor pattern collection of organizing according to starting point and endpoint data; The server end communication module is responsible for communicating with the client communication module, accepts the path abstract data of client, sends it to motor pattern and excavates module, and will excavate the result and return to client; The client and server end communicates through GPRS.

The concrete steps that realize are following:

Use prerequisite: user's mobile device need possess GPS positioning function (needing internal or external gps receiver), and has collected a certain amount of GPS track data.

Step 1: track data pre-service.The track data preprocessing process comprises that track data cleans and two parts are cut apart in the path.Aspect the track data cleaning, because the location of GPS equipment is uncertain, the GPS track that collects comprises outlier.The location point continuous to the time carries out cluster according to its speed, with speed rationally the continuous position point of (speed is less than a certain threshold value) incorporate same cluster into, when detecting unreasonable speed, and current cluster comprises location point and filters out location point data wherein more at least.Aspect cutting apart in the path, according to cutting apart at interval the writing time of continuous position point, if then track is divided into two paths greater than certain threshold value at interval the writing time of continuous two location points.

Step 2: candidate's start point/end point is extracted.The place that starting point and terminal point often leave and go to for the user; Adopt a kind of time-based clustering algorithm of expansion; To every paths; Along the time shaft forward and oppositely respectively to location point carry out clustering processing (with the time continuously and the location point of close together incorporate same cluster into); And find out first cluster (forward direction cluster) and last cluster (back is to cluster) in path, and then forward direction cluster set and the cluster of back in cluster set being mated, cluster centre point becomes candidate's start point/end point apart from then merging it less than a certain threshold value.

Step 3: the path is abstract.Based on space-division method; At first the appointed area is divided into the square net that the length of side equates, location point is replaced by the grid that comprises it, and the user path is converted into grid sequence; Starting and terminal point situation according to the path of passing through each grid merges relevant grid with structure realm then; Make the path through the same area have similar starting point and terminal point, at last with the user path abstract be regional sequence, be called sequence zone time.

Step 4: motor pattern excavates.Motor pattern is used to describe place and the path of walking and both relations that the user often leaves and goes to.Based on PrefixSpan sequential mode mining algorithm; Introduce mode time continuity notion; Require the motor pattern element to satisfy time continuity constraint (because the motor pattern element is the zone of band timestamp, promptly the interval time of adjacent area is less than a certain threshold value), make that the motor pattern element satisfies temporal continuity on the one hand; Motor pattern can be tolerated the interference in the track data on the other hand, thereby excavates longer motor pattern.

5) step 5: online position prediction.Based on user movement pattern of excavating and online exercise data thereof, associated prediction is carried out in its destination and following path.At first based on a kind of data structure of scheme-tree motor pattern is carried out modeling, motor pattern carries out index through its prefix in tree, so that pattern is searched.Behind the online exercise data of given user (comprising starting point and the information such as regional sequence of visiting recently), the regional sequence of visiting recently with the user carries out matched and searched as prefix in scheme-tree.After obtaining lookup result (i.e. the motor pattern of coupling); At first based on user's starting point and the motor pattern that finds; The destination most possible according to the probability model predictive user is then based on customer objective ground and the motor pattern that finds, the following path most possible according to the probability model predictive user (being following moving region sequence); And be principle to approach the destination, predict the outcome and predict repeatedly through splicing following path to improve following path prediction length.

The present invention includes two parts program: client-side program and server end program.Client-side program is installed on the mobile device, mainly is divided into four modules: the GPS track data is gathered and pre-processing module, and motor pattern excavates preparation module, online position prediction module, and and server com-munication module; The server major function is that motor pattern excavates, and this is because mode excavation algorithm computation pressure is bigger on the one hand, is the protection privacy of user on the other hand, and server is only handled the user path (not comprising actual geography information) of abstract mistake.Client-side program and server end program communicate through GPRS, and its software general frame is as shown in Figure 2.

Client-side program:

Client is a direct user oriented part in the native system, and it operates on the mobile device fully, respectively each module is elaborated below:

(1) the GPS track data is gathered and pre-processing module:

This module is applicable to GPS module that mobile device is built-in or the external GPS module that connects through bluetooth, gathers the GPS location point with certain SF, is recorded as the GPS track data, and carries out persistence and handle (being recorded in the mobile phone SD card like the form with file).In addition, the cleaning that track data is carried out and pre-service work such as cut apart and can carry out simultaneously with the position data sampling is handled required time to save the later stage track data.

(2) motor pattern excavates preparation module:

It is as shown in Figure 3 that motor pattern excavates the preparation module workflow, and module comprises that candidate's start point/end point is extracted and the function of abstract two aspects, path.Motor pattern excavates preparation module and can be called by user's demonstration, also can after the user collects a certain amount of new track data, be called automatically by system.After calling this module with thread independently at running background; At first candidate's start point/end point of user is extracted; Track data to the user carries out the abstract processing then; Obtain sequence sets zone time,, and call its motor pattern mining algorithm at last through sending the abstract path data to server with server com-munication module.Motor pattern excavates preparation module and is responsible for the information to obtaining in the operational process, comprises that information such as candidate's start point/end point, set of regions, zone time sequence carries out persistence and handle, and the motor pattern that also server is returned carries out persistence to be handled.

(3) online position prediction module:

When online position prediction module begins one section new distance the user, move simultaneously on backstage and track data acquisition module with the form of a separate threads, and real-time estimate is carried out in user's destination and following path.Online position prediction module workflow is as shown in Figure 4:

1) be enabled in the line position prediction module, can at first carry out modeling the user movement pattern that is stored on the mobile device, the structural model tree, scheme-tree comprises all motor patterns and adopts the probability of different starting points and terminal point.Scheme-tree is configured in and only carries out once after this module starts, and the result is retained in the Installed System Memory, is convenient to use repeatedly after online exercise data upgrades.

2) may change at any time owing to the user movement behavior; Online position prediction algorithm is called (calling once as 5 minutes) repeatedly with certain interval; All can produce new online exercise data at every turn after being called, at this moment need in scheme-tree, to carry out again matched and searched.Because there is most important influence in the path of user's visit recently to its following motion, for reducing calculated amount, uses nearest several access region in the online exercise data to carry out matched and searched.

3) matched and searched can be returned some candidate's motor patterns, supports situation (being starting point and terminal point and the number of times of user when adopting this motor pattern) and user's origin information to predict its destination according to start point/end point in candidate's motor pattern.

4) according to the probability of going to this destination in the destination of predicting and the candidate's motor pattern candidate's motor pattern is traveled through to predict its following moving region sequence.

5) for obtaining longer predicted path; One take turns prediction and finish after; If prediction algorithm end condition (the following path and the destination that dope are enough approaching, can't dope any following path again, or reach a predetermined prediction number of times) does not satisfy; Then splice nearest access region sequence and the following moving region sequence that obtains of prediction in the online exercise data obtaining new online exercise data, and restart matched and searched.

Destination and following path on-line prediction algorithm are as shown in table 1; The given path mode collection PS that from the user trajectory data, excavates; Algorithm at first makes up scheme-tree; And from online exercise data online_data, obtaining starting point and nearest routing information (1-2 is capable), algorithm carries out path mode matched and searched process (3-8 is capable) then.Owing to possibly can't in scheme-tree, find the path mode with its coupling for an online regional sequence.In this case, we shorten online regional sequence, remove its regional element (9-10 is capable) the earliest, and restart matching process up to the path mode that finds coupling.The route matching process will be accomplished, and this is because under worst case, online regional sequence is shortened into single active regions, then must in the child node of root node, find the coupling inlet.After the coupling path mode found, algorithm carried out the degree of depth with next regional probability to scheme-tree based on starting point-terminal point and searches with prediction destination and following path (12-14 is capable).

Table 1

(4) and server com-munication module

When needs excavate track data again; To send new path abstract data to server with server com-munication module; Server will be carried out the motor pattern mining algorithm, and the motor pattern collection that obtains is sent back client-side program with document form.

Server:

Server is actually a Web Service, mainly carries out the bigger motor pattern excacation of calculating pressure, and it communicates through GPRS and client-side program, and its flow process is as shown in Figure 5, respectively each module is elaborated below:

(1) motor pattern excavates module:

This module is excavated the user path abstract data that receives based on the time series mining algorithm that proposes, and obtains according to the motor pattern collection of starting and terminal point to organizing, and its form that is organized into file is returned.

The motor pattern mining algorithm is as shown in table 2, and most important notion is prefix (prefix) in the algorithm, and this prefix notion is similar to the prefix notion in the PrefixSpan algorithm, but different aspect two.The first, adjacent element must satisfy time continuity constraint (the 9th row) in the prefix.The second, when each prefix generated, we supported set attribute for it is provided with a starting and terminal point, made motor pattern comprise starting point-terminal point and supported situation information.The continuous recurrence of algorithm is expanded prefix to generate longer pattern.When algorithm was called for the first time, initial parameter projections was user's sequence sets zone time, and initial prefix is all frequent zones that occur.Each time in the recurrence; To each mapping in the current mapping ensemblen; Algorithm is searched and last element that is complementary of current prefix (the 2nd row) therein, generates new mapping ensemblen (the 3rd row) based on this element, and in new mapping ensemblen, seeks frequent zone as new sub-prefix set (the 4th row).If sub-prefix non-NULL, algorithm is through splicing current prefix and certain sub-prefix to generate new prefix, and current prefix and sub-prefix to be spliced must satisfy time continuity constraint (8-10 is capable).Based on newly-generated prefix, a new path mode of algorithm construction (11-12 is capable).

Table 2

(2) with the client communication module:

This module mainly is responsible for communicating with client-side program, accepts the path abstract data of client, sends it to motor pattern and excavates module, and will excavate the result and return to client.Because server receives only the path abstract data (being the time series of zone number) that does not comprise any particular geographic location information, can protect privacy of user preferably.

Above said be specific embodiment of the present invention and the know-why used, if the change of doing according to conception of the present invention, when the function that it produced does not exceed spiritual that instructions and accompanying drawing contain yet, must belong to protection scope of the present invention.

Claims (7)

1. the position prediction system based on the historical track data mining is characterized in that, comprises mobile client and server end; Client is installed on the mobile device; Comprise collection of GPS track data and pre-processing module, motor pattern excavates preparation module, online position prediction module; Four modules of client communication module; The GPS track data is gathered with pre-processing module and is responsible for being recorded as the GPS track data with certain SF collection GPS location point, and track data is cleaned and cuts apart pre-service; Motor pattern excavates preparation module and is responsible for candidate's Origin And Destination is extracted; Then track data being carried out abstract handles; Send Origin And Destination and abstract path data to server end through the client communication module at last, and call its motor pattern mining algorithm; Online position prediction module is responsible for the user movement pattern through excavating that server end returns is carried out modeling, and structural model is set, and real-time estimate is carried out in user's destination and following path; The client communication module is responsible for communicating with server end; Server end comprises that motor pattern excavates module and server end communication module; Motor pattern excavates module and is responsible for according to the time series mining algorithm path abstract data of client upload being excavated, and obtains the motor pattern collection of organizing according to starting point and endpoint data; The server end communication module is responsible for communicating with the client communication module, accepts the path abstract data of client, sends it to motor pattern and excavates module, and will excavate the result and return to client.
2. a kind of position prediction system based on the historical track data mining according to claim 1 is characterized in that the client and server end communicates through GPRS.
3. the position predicting method based on the historical track data mining is characterized in that, may further comprise the steps:
1) on user's mobile device, settles gps receiver, gather the GPS location point, be recorded as the GPS track data, and protect to preserve with certain SF;
2) the GPS track data is cleaned, the location point continuous to the time carries out cluster according to its speed, incorporates speed into same cluster less than the continuous position point of threshold value, and filters wherein irrational location point data;
3) according to cutting apart at interval the writing time of continuous position point, if then track is divided into two paths greater than threshold value at interval the writing time of continuous two location points;
4) to every paths along the time shaft forward with oppositely respectively location point is carried out clustering processing; Time is continuous and location point close together is incorporated same cluster into; The forward direction cluster of finding out the path and back are to cluster; Then forward direction cluster set and the cluster of back in cluster set are mated, cluster centre point distance then merges it less than threshold value becomes candidate's beginning or end;
5) appointed area is divided into the square net that the length of side equates, location point is replaced by the grid that comprises it, and the path is converted into the grid sequence data, then track data is carried out abstract and handle;
6) send Origin And Destination and abstract path data to server end through the client communication module and send the abstract path data to server; And call its motor pattern mining algorithm the user path abstract data that receives is excavated; Obtain the motor pattern collection organized to terminal according to starting point, and its form that is organized into file is returned;
7) utilization PrefixSpan sequential mode mining algorithm carries out modeling to motor pattern, the structural model tree, and scheme-tree comprises all motor patterns and adopts the probability of different starting points and terminal point;
8) based on user movement pattern of excavating and online exercise data thereof, through carrying out matched and searched in the scheme-tree, obtain lookup result, its destination and following path are predicted.
4. a kind of position predicting method based on the historical track data mining according to claim 3 is characterized in that, the described gps receiver of step 1) is the built-in GPS module of mobile device or passes through the external GPS module that bluetooth connects.
5. a kind of position predicting method based on the historical track data mining according to claim 3 is characterized in that step 1), step 2) and step 3) can synchronous processing.
6. a kind of position predicting method based on the historical track data mining according to claim 3 is characterized in that, the described server end of step 6) receives only Origin And Destination and the abstract path data that do not comprise any particular geographic location information.
7. according to claim 3 or the described a kind of position predicting method of 4 or 5 or 6 arbitrary claims, it is characterized in that step 7) is described scheme-tree result be retained in the Installed System Memory based on the historical track data mining.
CN201110308289XA 2011-10-10 2011-10-10 Location prediction system and method based on historical track data mining CN102509170A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201110308289XA CN102509170A (en) 2011-10-10 2011-10-10 Location prediction system and method based on historical track data mining

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201110308289XA CN102509170A (en) 2011-10-10 2011-10-10 Location prediction system and method based on historical track data mining

Publications (1)

Publication Number Publication Date
CN102509170A true CN102509170A (en) 2012-06-20

Family

ID=46221250

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201110308289XA CN102509170A (en) 2011-10-10 2011-10-10 Location prediction system and method based on historical track data mining

Country Status (1)

Country Link
CN (1) CN102509170A (en)

Cited By (38)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102853840A (en) * 2012-09-10 2013-01-02 无锡坦程物联网科技有限公司 Method for discovering routine vehicle parking region based on grids
CN102903038A (en) * 2012-09-29 2013-01-30 威海智联信息网络有限公司 Method and system for small object logistics distribution based on user track
CN103024669A (en) * 2012-11-30 2013-04-03 广东工业大学 Mesh network node positioning method
CN103149577A (en) * 2013-02-28 2013-06-12 山东大学 Combined navigation method of Beidou navigation, GPS (global positioning system) navigation and historical data fusion
CN103593349A (en) * 2012-08-14 2014-02-19 中国科学院沈阳自动化研究所 Movement position analysis method in sense network environment
CN103870892A (en) * 2014-03-26 2014-06-18 北京清软英泰信息技术有限公司 Method and system for achieving railway locomotive operation control from off-line mode to on-line mode
CN103970972A (en) * 2013-01-31 2014-08-06 中国移动通信集团公司 Motion information processing method and device
CN104270714A (en) * 2014-09-11 2015-01-07 华为技术有限公司 Method and device for determining action trails of user
CN104298669A (en) * 2013-07-16 2015-01-21 江苏宏联物联网信息技术有限公司 Person geographic information mining model based on social network
CN104464344A (en) * 2014-11-07 2015-03-25 湖北大学 Vehicle driving path prediction method and system
CN104598992A (en) * 2015-01-12 2015-05-06 北京中交兴路车联网科技有限公司 Common route analysis method for vehicle
CN104636816A (en) * 2013-11-07 2015-05-20 财团法人资讯工业策进会 Apparatus and method for creating a power consumption model
CN104754730A (en) * 2013-12-27 2015-07-01 华为技术有限公司 Method and device for position prediction
CN104917698A (en) * 2015-06-15 2015-09-16 中国科学院自动化研究所 Resource organization and schedule optimization method of internet platform
CN104931041A (en) * 2015-05-03 2015-09-23 西北工业大学 Method for predicting place sequence based on user track data
CN105678246A (en) * 2015-12-31 2016-06-15 浙江工业大学 Motion mode excavation method based on base station label locus
CN105933857A (en) * 2015-11-25 2016-09-07 中国银联股份有限公司 Mobile terminal position prediction method and apparatus
TWI558249B (en) * 2014-09-23 2016-11-11 Chunghwa Telecom Co Ltd People 's potential information generation method
CN106408239A (en) * 2016-09-19 2017-02-15 上海德启信息科技有限公司 Express delivery service information monitoring method, system, client and server
CN106462627A (en) * 2014-12-29 2017-02-22 谷歌公司 Analyzing semantic places and related data from a plurality of location data reports
CN106568445A (en) * 2016-11-11 2017-04-19 吉林大学 Indoor track prediction method based on bidirectional circulation neural network
CN106595665A (en) * 2016-11-30 2017-04-26 耿生玲 Prediction method for spatial-temporal trajectory of moving object in obstructed space
CN106682093A (en) * 2016-11-30 2017-05-17 贵州智通天下信息技术有限公司 Method for generating route according to track record
CN103747523B (en) * 2014-01-14 2017-09-22 上海河广信息科技有限公司 A kind of customer location forecasting system and method based on wireless network
WO2017215322A1 (en) * 2016-06-13 2017-12-21 Beijing Didi Infinity Technology And Development Co., Ltd. Systems and methods for recommending recommended service location
CN107657007A (en) * 2017-09-22 2018-02-02 广东欧珀移动通信有限公司 Information-pushing method, device, terminal, readable storage medium storing program for executing and system
US9906913B2 (en) 2013-10-28 2018-02-27 Huawei Technologies Co., Ltd. Method for predicting position of mobile user, and equipment
CN107869991A (en) * 2017-09-20 2018-04-03 百度在线网络技术(北京)有限公司 Acquisition method and device, the computer equipment and computer-readable recording medium of indoor location data
CN107889061A (en) * 2017-11-08 2018-04-06 洛阳师范学院 The ambulant appraisal procedure of vehicle and in field of video transmission application in car networking
CN107895103A (en) * 2017-12-05 2018-04-10 北京科技大学 A kind of track data cleaning method
CN107979528A (en) * 2017-11-17 2018-05-01 广东乐心医疗电子股份有限公司 A kind of weather based reminding method and system
CN108153867A (en) * 2017-12-25 2018-06-12 山东大学 User trajectory Forecasting Methodology and device based on temporal regularity
WO2018166280A1 (en) * 2017-03-14 2018-09-20 广州市动景计算机科技有限公司 Information recommendation method, apparatus, system, and storage medium
CN109357682A (en) * 2018-09-19 2019-02-19 潍坊工程职业学院 A kind of road navigation method
CN109697221A (en) * 2018-11-22 2019-04-30 东软集团股份有限公司 Method for digging, device, storage medium and the electronic equipment of track rule
CN110264244A (en) * 2019-05-25 2019-09-20 浙江华坤道威数据科技有限公司 A kind of advertising user track following management system and method
CN106682093B (en) * 2016-11-30 2020-06-26 贵州智通天下信息技术有限公司 Method for generating route according to track record
CN111078754B (en) * 2019-12-19 2020-08-25 南京柏跃软件有限公司 Frequent trajectory extraction method based on massive space-time data and mining system thereof

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
叶谦: "《基于路径模式挖掘的个人连续路径预测》", 《中国优秀硕士论文全文数据库》, 31 December 2008 (2008-12-31), pages 25 - 76 *

Cited By (51)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103593349A (en) * 2012-08-14 2014-02-19 中国科学院沈阳自动化研究所 Movement position analysis method in sense network environment
CN102853840A (en) * 2012-09-10 2013-01-02 无锡坦程物联网科技有限公司 Method for discovering routine vehicle parking region based on grids
CN102903038A (en) * 2012-09-29 2013-01-30 威海智联信息网络有限公司 Method and system for small object logistics distribution based on user track
CN102903038B (en) * 2012-09-29 2016-03-16 威海智联信息网络有限公司 Based on smallclothes Logistics Distribution Method and the system of user trajectory
CN103024669B (en) * 2012-11-30 2015-03-25 广东工业大学 Mesh network node positioning method
CN103024669A (en) * 2012-11-30 2013-04-03 广东工业大学 Mesh network node positioning method
CN103970972A (en) * 2013-01-31 2014-08-06 中国移动通信集团公司 Motion information processing method and device
CN103970972B (en) * 2013-01-31 2018-08-17 中国移动通信集团公司 Movable information treating method and apparatus
CN103149577A (en) * 2013-02-28 2013-06-12 山东大学 Combined navigation method of Beidou navigation, GPS (global positioning system) navigation and historical data fusion
CN103149577B (en) * 2013-02-28 2016-04-13 山东大学 The Combinated navigation method that " Big Dipper " navigation, GPS navigation and historical data merge
CN104298669A (en) * 2013-07-16 2015-01-21 江苏宏联物联网信息技术有限公司 Person geographic information mining model based on social network
US9906913B2 (en) 2013-10-28 2018-02-27 Huawei Technologies Co., Ltd. Method for predicting position of mobile user, and equipment
CN104636816A (en) * 2013-11-07 2015-05-20 财团法人资讯工业策进会 Apparatus and method for creating a power consumption model
CN104754730A (en) * 2013-12-27 2015-07-01 华为技术有限公司 Method and device for position prediction
CN104754730B (en) * 2013-12-27 2018-07-31 华为技术有限公司 The method and device of position prediction
CN103747523B (en) * 2014-01-14 2017-09-22 上海河广信息科技有限公司 A kind of customer location forecasting system and method based on wireless network
CN103870892A (en) * 2014-03-26 2014-06-18 北京清软英泰信息技术有限公司 Method and system for achieving railway locomotive operation control from off-line mode to on-line mode
CN103870892B (en) * 2014-03-26 2017-05-10 中车信息技术有限公司 Method and system for achieving railway locomotive operation control from off-line mode to on-line mode
CN104270714B (en) * 2014-09-11 2018-04-10 华为技术有限公司 The method and apparatus for determining user movement track
CN104270714A (en) * 2014-09-11 2015-01-07 华为技术有限公司 Method and device for determining action trails of user
TWI558249B (en) * 2014-09-23 2016-11-11 Chunghwa Telecom Co Ltd People 's potential information generation method
CN104464344A (en) * 2014-11-07 2015-03-25 湖北大学 Vehicle driving path prediction method and system
CN106462627A (en) * 2014-12-29 2017-02-22 谷歌公司 Analyzing semantic places and related data from a plurality of location data reports
CN104598992A (en) * 2015-01-12 2015-05-06 北京中交兴路车联网科技有限公司 Common route analysis method for vehicle
CN104931041B (en) * 2015-05-03 2017-12-01 西北工业大学 A kind of location sequence Forecasting Methodology based on user trajectory data
CN104931041A (en) * 2015-05-03 2015-09-23 西北工业大学 Method for predicting place sequence based on user track data
CN104917698B (en) * 2015-06-15 2018-10-30 中国科学院自动化研究所 The resource tissue and method for optimizing scheduling of internet platform
CN104917698A (en) * 2015-06-15 2015-09-16 中国科学院自动化研究所 Resource organization and schedule optimization method of internet platform
CN105933857B (en) * 2015-11-25 2019-05-14 中国银联股份有限公司 A kind of mobile terminal locations prediction technique and device
CN105933857A (en) * 2015-11-25 2016-09-07 中国银联股份有限公司 Mobile terminal position prediction method and apparatus
CN105678246B (en) * 2015-12-31 2018-09-18 浙江工业大学 A kind of motor pattern method for digging based on base station label track
CN105678246A (en) * 2015-12-31 2016-06-15 浙江工业大学 Motion mode excavation method based on base station label locus
WO2017215322A1 (en) * 2016-06-13 2017-12-21 Beijing Didi Infinity Technology And Development Co., Ltd. Systems and methods for recommending recommended service location
CN106408239A (en) * 2016-09-19 2017-02-15 上海德启信息科技有限公司 Express delivery service information monitoring method, system, client and server
CN106568445A (en) * 2016-11-11 2017-04-19 吉林大学 Indoor track prediction method based on bidirectional circulation neural network
CN106682093A (en) * 2016-11-30 2017-05-17 贵州智通天下信息技术有限公司 Method for generating route according to track record
CN106595665B (en) * 2016-11-30 2019-10-11 耿生玲 The prediction technique of mobile object space-time trajectory in a kind of space with obstacle
CN106595665A (en) * 2016-11-30 2017-04-26 耿生玲 Prediction method for spatial-temporal trajectory of moving object in obstructed space
CN106682093B (en) * 2016-11-30 2020-06-26 贵州智通天下信息技术有限公司 Method for generating route according to track record
WO2018166280A1 (en) * 2017-03-14 2018-09-20 广州市动景计算机科技有限公司 Information recommendation method, apparatus, system, and storage medium
CN107869991A (en) * 2017-09-20 2018-04-03 百度在线网络技术(北京)有限公司 Acquisition method and device, the computer equipment and computer-readable recording medium of indoor location data
CN107657007A (en) * 2017-09-22 2018-02-02 广东欧珀移动通信有限公司 Information-pushing method, device, terminal, readable storage medium storing program for executing and system
CN107889061B (en) * 2017-11-08 2020-06-02 洛阳师范学院 Method for evaluating vehicle mobility in Internet of vehicles and application of method in video transmission field
CN107889061A (en) * 2017-11-08 2018-04-06 洛阳师范学院 The ambulant appraisal procedure of vehicle and in field of video transmission application in car networking
CN107979528A (en) * 2017-11-17 2018-05-01 广东乐心医疗电子股份有限公司 A kind of weather based reminding method and system
CN107895103A (en) * 2017-12-05 2018-04-10 北京科技大学 A kind of track data cleaning method
CN108153867A (en) * 2017-12-25 2018-06-12 山东大学 User trajectory Forecasting Methodology and device based on temporal regularity
CN109357682A (en) * 2018-09-19 2019-02-19 潍坊工程职业学院 A kind of road navigation method
CN109697221A (en) * 2018-11-22 2019-04-30 东软集团股份有限公司 Method for digging, device, storage medium and the electronic equipment of track rule
CN110264244A (en) * 2019-05-25 2019-09-20 浙江华坤道威数据科技有限公司 A kind of advertising user track following management system and method
CN111078754B (en) * 2019-12-19 2020-08-25 南京柏跃软件有限公司 Frequent trajectory extraction method based on massive space-time data and mining system thereof

Similar Documents

Publication Publication Date Title
Ahmed et al. A comparison and evaluation of map construction algorithms using vehicle tracking data
EP3342192B1 (en) Method and arrangement for locating a mobile device
Zhang et al. GROPING: Geomagnetism and crowdsensing powered indoor navigation
CN104520881B (en) Predict that the future to place accesses based on possibility is accessed to neighbouring destination ranking and from position history
US20190156661A1 (en) Method and system for modeling and processing vehicular traffic data and information and applying thereof
US8457880B1 (en) Telematics using personal mobile devices
CN105354196B (en) Information-pushing method and information push-delivery apparatus
Chen et al. Reliable shortest path finding in stochastic networks with spatial correlated link travel times
CN102023018B (en) Method of predicting energy consumption, apparatus for predicting energy consumption, and terminal apparatus
US10571288B2 (en) Searching similar trajectories by locations
US8165800B2 (en) Apparatus for and method of providing data to an external application
CN104380293B (en) Method and computing device for providing information of relevant elements from maps history based on location
CN102762957B (en) Historial traffic data compression
CN104541527B (en) For inferring the method for the interest of user and calculating equipment
CN104899263B (en) A kind of ship track mining analysis and monitoring method based on specific region
CN104049575B (en) It is collected in Process Control System and delivers data to big data machine
Nadembega et al. A destination and mobility path prediction scheme for mobile networks
CN100541130C (en) Navigational system
CN103593361B (en) Movement space-time trajectory analysis method in sense network environment
Chen et al. Trajectory simplification method for location-based social networking services
CN101109804B (en) System for fast three-dimensional locating wireless sensor network and method thereof
Zheng et al. Learning location correlation from gps trajectories
CN101936739B (en) Navigation device, route-search server, and route-search system
CN103152823B (en) A kind of wireless indoor location method
CN104240163A (en) Intelligent district recommendation method and intelligent district recommendation 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
C53 Correction of patent for invention or patent application
CB02 Change of applicant information

Address after: Hangzhou City, Zhejiang Province, Binjiang District Puyan street 310053 Albert Road No. 1 Building 2 Zhejiang Hongcheng computer system Co. Ltd.

Applicant after: Zhejiang Hongcheng Computer Systems Co., Ltd.

Address before: 1, building 11, building 1, No. 310013, staff Road, Hangzhou, Zhejiang

Applicant before: Zhejiang Hongcheng Computer Systems Co., Ltd.

C02 Deemed withdrawal of patent application after publication (patent law 2001)
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20120620