CN109272032A - Trip mode recognition methods, device, computer equipment and storage medium - Google Patents
Trip mode recognition methods, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109272032A CN109272032A CN201811032855.7A CN201811032855A CN109272032A CN 109272032 A CN109272032 A CN 109272032A CN 201811032855 A CN201811032855 A CN 201811032855A CN 109272032 A CN109272032 A CN 109272032A
- Authority
- CN
- China
- Prior art keywords
- base station
- data
- track
- characteristic
- signaling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The application provides a kind of trip mode recognition methods, device, computer equipment and storage medium, wherein, method includes: the base station data for acquiring signaling Internet data and communication base station, and track pair is generated according to the collected data, track is extracted to characteristic, characteristic includes speed, based on velocity characteristic, update default cluster learning algorithm training pattern, obtain updated cluster learning algorithm training pattern, it carries in cluster learning algorithm training pattern in the updated using speed as the data of core, and different trip modes most significantly be also speed in terms of difference, therefore, based on updated cluster learning algorithm training pattern, trip mode can be accurately identified.
Description
Technical field
This application involves big data identification technology fields, more particularly to a kind of trip mode recognition methods, device, calculating
Machine equipment and storage medium.
Background technique
The trip mode for identifying resident, facilitates urban transportation infrastructure construction planning and intelligent transportation decision support.
The conventional differentiation to resident trip mode, main comprehensive three classes data are analyzed, statistical data, movement including traffic department
The position of equipment and temporal information, city map network structure.Due to traffic data timeliness and be located on map
Accuracy, position and temporal information of the identifying schemes of current majority trip modes only with mobile device.
The mobile phone signaling of operator can with magnanimity, in time acquire resident position and temporal information, be current technical side
The basic data that case is relied on, basic ideas are to navigate to a little and calculate when consumption between points, further according to distance and when consume
The speed of calculating corresponds to the corresponding vehicles.This kind of technical solution relies on the priori knowledge of vehicles travel speed, and
There are certain deviations for the differentiation result of trip mode.
Summary of the invention
Based on this, it is necessary in view of the above technical problems, provide a kind of accurate trip mode recognition methods of identification, dress
It sets, computer equipment and storage medium.
A kind of trip mode recognition methods, which comprises
Obtain the base station data of signaling Internet data and communication base station；
According to the signaling Internet data and the base station data, track pair is generated；
It obtains track and default cluster learning algorithm training mould is updated according to the track to characteristic to characteristic
Type, the track include speed to characteristic；
According to updated cluster learning algorithm training pattern, trip mode is identified.
The signaling Internet data includes terminal identity identification marking, communication base station identity in one of the embodiments,
Identification marking and acquisition signaling time, the base station data include base station identity identification marking, latitude and longitude of base station data and
Base station range.
It is described according to the signaling Internet data and the base station data in one of the embodiments, generate track
To including:
According to the signaling Internet data, the corresponding first base station of current single signaling Internet data and the second base station are identified
And acquisition signaling time；
According to the base station data, the corresponding longitude and latitude data of the first base station and second base station are obtained；
According to the longitude and latitude data of acquisition, the first base station is calculated at a distance from second base station；
According to the base station data, the coverage area of the first base station Yu second base station is obtained；
According to the first base station at a distance from second base station, the covering of the first base station and second base station
Range and the acquisition signaling time generate track pair.
In one of the embodiments, it is described according to the first base station at a distance from second base station, described first
The coverage area and the acquisition signaling time of base station and second base station generate track to before, further includes:
Judge whether the first base station is greater than first base station covering radius and the second base at a distance from second base station
It stands the sum of covering radius；
If more than, into it is described according to the first base station at a distance from second base station, the first base station and institute
State the second base station coverage area and the acquisition signaling time, generate track pair step；
If being not more than, using next single signaling Internet data as current single signaling Internet data, the basis is returned to
The signaling Internet data identifies the corresponding first base station of current single signaling Internet data and the second base station and acquisition signaling
The step of time.
The track includes maximum distance, minimum range and average distance to characteristic in one of the embodiments,
And maximum speed, minimum speed and average speed, it is described according to the track to characteristic, update default clustering learning
Algorithm training pattern includes:
Same trajectories are obtained to corresponding friction speed data, generate characteristic training set；
Default cluster learning algorithm training pattern is updated to the characteristic training set.
It is described according to updated cluster learning algorithm training pattern in one of the embodiments, identify trip mode
Include:
Update the model that multiple clusters are preset in the updated cluster learning algorithm training pattern, wherein each cluster
Model carries different velocity characteristic data；
The corresponding velocity characteristic data of trip mode to be selected are preset in acquisition；
According to the model of multiple clusters of update and it is described preset the corresponding velocity characteristic data of trip mode to be selected, identification
Trip mode.
It is described in one of the embodiments, that default cluster learning algorithm training is updated to characteristic according to the track
Before model, further includes:
Obtain the base station data of history signaling Internet data and historical communication base station；
According to the history signaling Internet data and the history base station data, historical track pair is generated；
It obtains historical track and is generated by default clustering learning and is calculated according to the historical track to characteristic for characteristic
Method training pattern.
A kind of trip mode identification device, described device include:
Data acquisition module, for obtaining the base station data of signaling Internet data and communication base station；
Track is to generation module, for generating track pair according to the signaling Internet data and the base station data；
Training module is clustered, characteristic is updated default poly- according to the track to characteristic for obtaining track
Class learning algorithm training pattern, the track include speed to characteristic；
Identification module, for identifying trip mode according to updated cluster learning algorithm training pattern.
A kind of computer equipment, including memory and processor, the memory are stored with computer program, and feature exists
In the processor is realized when executing the computer program such as the step of the above method.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
It is real such as the step of above-mentioned method when row.
Above-mentioned trip mode recognition methods, device, computer equipment and storage medium acquire signaling Internet data and lead to
Believe the base station data of base station, and generate track pair according to the collected data, extracts track to characteristic, characteristic includes
Speed is based on velocity characteristic, updates default cluster learning algorithm training pattern, obtains updated cluster learning algorithm training mould
Type is carried using speed as the data of core in cluster learning algorithm training pattern in the updated, and different trip modes is most
Therefore difference in terms of apparent and speed is based on updated cluster learning algorithm training pattern, can be recognized accurately
Line mode.
Detailed description of the invention
Fig. 1 is the applied environment figure of trip mode recognition methods in one embodiment；
Fig. 2 is the flow diagram of above-mentioned trip mode recognition methods one embodiment；
Fig. 3 is tracing point sequence chart；
The flow diagram of above-mentioned second embodiment of trip mode recognition methods of Fig. 4；
Fig. 5 is the structural block diagram of trip mode identification device in one embodiment；
Fig. 6 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, not
For limiting the application.
Trip mode recognition methods provided by the present application, can be applied in application environment as shown in Figure 1.Wherein, eventually
End 102 is communicated with server 104 by network by network.104 acquisition terminal of server, 102 signaling Internet data and
Track pair is generated according to signaling Internet data and base station data with the base station data of 102 communication base station of terminal, obtains track
To characteristic, cluster learning algorithm processing is carried out to characteristic according to track, track includes speed to characteristic, according to
Cluster learning algorithm processing result identifies that trip mode, the result for exporting identification show or store to third party.Wherein, terminal
102 can be, but not limited to be various personal computers, laptop, smart phone, tablet computer and portable wearable set
Standby, server 104 can be realized with the server cluster of the either multiple server compositions of independent server.
In one embodiment, as shown in Fig. 2, providing a kind of trip mode recognition methods, it is applied to Fig. 1 in this way
In server for be illustrated, comprising the following steps:
S200: the base station data of signaling Internet data and communication base station is obtained.
Terminal can be communicated constantly with base station during online.By taking terminal is smart phone as an example, user exists at present
It often will use smart phone during trip, therefore the motion track (trip mode) of smart phone generally just represents user's
Trip mode, user is during trip, and using smart phone online, object for appreciation game, chat etc., during these, intelligence is eventually
Data interaction is constantly carried out between end and base station, terminal also constantly generates signaling Internet data, meanwhile, as mobile terminal (is used
Family) movement, user may be moved to the coverage area of the base station B by the coverage area of the base station A, corresponding, with the mobile terminal
Base station B is switched to by A, along with the change of base station data in the handoff procedure of base station.Herein, netting index in signaling is acquired
According to carrying acquisition time, terminal identity identification marking and communication base station identity identification in signaling Internet data.Separately
Outside, the base station data of communication base station is also acquired, communication base station refers to the base station communicated with present terminal, carries in base station data
There are base station identity identification marking, latitude and longitude of base station data and base station range, base station identity identification marking is for distinguishing not
Same base station, latitude and longitude of base station data are used to determine the geographical location of base station, and the longitude and latitude data based on two base stations can be accurate
Distance between two base stations is calculated, base station range is for determining in terminal mobile communication process whether cut
It changes, such as when terminal moves out in current base station coverage area, it is necessary to switch to other base stations and be communicated.More specifically
For, by taking terminal is smart phone as an example, the signaling Internet data of smart phone, including the signaling netted from 2G, 3G and 4G and
Internet data, essential record cell-phone number, base station number, the time point for acquiring signaling etc..
S400: according to signaling Internet data and base station data, track pair is generated.
Terminal identifies that it moves corresponding serving BS, that is, produces its trace information in moving process, track letter
Breath also represents the motion track of user.Specifically, according to signaling Internet data and base station data, each terminal can be identified
In motion profile of the different time sections between different base station, many terminals may be corresponding with for identical two base stations
There are identical tracks pair for the multiple terminals of energy.Based on signaling Internet data and base station data, it is corresponding each that each terminal can be distinguished
The motion profile of base station can be depicted as tracing point sequence chart based on these data.
In practical applications, the track point sequence for counting each cell-phone number (terminal identity identification marking), such as Fig. 3 tracing point
Sequence.The base station locating for the initial time point is to start tracing point, and one signaling Internet data of every generation calculates two base stations and passes through
Distance between latitude, horizontal axis indicate time point, and the longitudinal axis indicates tracing point to the distance for starting tracing point.Using haversine function come
The distance between calculation base station longitude and latitude two o'clock, calculated is the shortest distance of the point between earth surface, and calculation formula is such as
Under:
Wherein: R be the earth radius, value 637km, d be the distance between two o'clock,For 1 (the first base of latitude point
Stand) latitude, indicated in the form of radian,It for the latitude of latitude point 2 (the second base station), is indicated in the form of radian, Δ γ
For location point 1 and 2 difference of longitudes are put, are indicated in the form of radian.
S600: it obtains track and default cluster learning algorithm training mould is updated according to track to characteristic to characteristic
Type, track include speed to characteristic.
There is different track datas for different users, obtains track to characteristic, include in this feature data
Speed data.Specifically, different trip modes (including single trip mode or Combined mode) its corresponding speed
Degree be it is different, therefore, first for track to first extracting characteristic, then characteristic is carried out at cluster learning algorithm
Reason.Speed in characteristic may include three maximum speed, minimum speed and average speed aspects.For identical two
Track pair between a base station, different users is different using the time of its consumption of different trip modes, corresponding speed
With regard to difference, conversely, carrying out cluster learning algorithm training to characteristic based on the track comprising speed, can be obtained based on speed
The model of the clustering cluster of different trip modes.Non-essential, track can also include maximum outside speed in characteristic to including
Distance, minimum range and average distance, the covering of maximum distance, that is, two base stations the base station center distance d+ of track centering two
The sum of radius r1 and r2, the covering radius r1 and r2 of minimum range, that is, base station center distance d of track centering two and two base stations
Difference.Default cluster learning algorithm training pattern is the mould constructed in advance based on historical data and cluster learning algorithm training
Type, it can be understood as what is obtained in above-mentioned steps S200 is the corresponding signaling Internet data of current t moment and communication base station
Base station data, default cluster learning algorithm training pattern then can be based on signaling Internet data before t moment and communication base
The base station data stood, using the model of cluster learning algorithm training building.
S800: according to updated cluster learning algorithm training pattern, trip mode is identified.
For step S600 cluster learning algorithm processing result, since the processing result is using speed as the core of cluster,
And the maximum difference of different trip modes lies also in speed, therefore, trip mode can be recognized accurately.Step S600 more
Newly after default cluster learning algorithm training pattern, updated cluster learning algorithm training pattern is obtained, after the update
Cluster learning algorithm training pattern be the signaling Internet data and communication base station that are obtained by step S200 base station data
Reason, which updates, to be obtained, therefore can accurately identify the base station data in the step S200 signaling Internet data obtained and communication base station
Corresponding trip mode.Non-essential, what step S200 was obtained can be current time corresponding signaling Internet data and leads to
Believe the base station data of base station, then updated cluster learning algorithm training pattern is can to characterize current trip mode, thus real
Now current trip mode is accurately identified.
Above-mentioned trip mode recognition methods, the base station data of acquisition signaling Internet data and communication base station, and according to
The data of acquisition generate track pair, extract track to characteristic, and characteristic includes speed, is based on velocity characteristic, update pre-
If cluster learning algorithm training pattern, updated cluster learning algorithm training pattern is obtained, clustering learning in the updated is calculated
The area using speed as the data of core, and in terms of different trip modes is most significantly also speed is carried in method training pattern
Not, therefore, it is based on updated cluster learning algorithm training pattern, trip mode can be accurately identified.
As shown in figure 4, S400 includes: in one of the embodiments,
S410: according to signaling Internet data, the corresponding first base station of current single signaling Internet data and the second base are identified
Stand and acquire signaling time.
Terminal identity identification marking is carried in signaling Internet data, shows that the terminal of present communications is (i.e. current to use
Family), the first base station and the second base station and acquire the signaling time which is communicated.Current single signaling Internet data
Refer to that in operations described below be the operation that the signaling Internet data being presently processing for each carries out, i.e., in each signaling
Network data can be handled using following same ways, to generate corresponding track pair.
S420: according to base station data, first base station and the corresponding longitude and latitude data in the second base station are obtained.
Latitude and longitude of base station data are carried in base station data, and current first base station and the second base are obtained from base station data
It stands corresponding longitude and latitude data.
S430: according to the longitude and latitude data of acquisition, first base station is calculated at a distance from the second base station.
S440: according to base station data, the coverage area of first base station and the second base station is obtained；
S450: according to first base station at a distance from the second base station, the coverage area of first base station and the second base station and adopt
Collect signaling time, generates track pair.
According to longitude and latitude data and above-mentioned formula, the distance d of first base station and the second base station can be calculated, is in addition obtained
The covering radius r1 of first base station and the covering radius r2 of the second base station are taken, distance d, covering radius r1 and covering radius are based on
R2, it is d+r1+r2, minimum range d-r1-r2, base that tracing point, which can be obtained, and correspond to maximum distance in first base station and the second base station
In the time t that step S410 is obtained, can calculate maximum speed is d+r1+r2/t, minimum speed d-r1-r2/t.For
Different signaling Internet datas can generate track pair using above-mentioned same way, by the track of generation to can use track
The record of point sequence seal.
In one of the embodiments, according to first base station at a distance from the second base station, first base station and the second base station
Coverage area and acquisition signaling time, generate track to before, further includes:
Step 1: judge whether first base station is greater than first base station covering radius and the second base station at a distance from the second base station
The sum of covering radius.
Step 2: if more than, into according to first base station at a distance from the second base station, first base station covers with the second base station
Lid range and acquisition signaling time, generate the step of track pair.
Step 3: if being not more than, using next single signaling Internet data as current single signaling Internet data, root is returned to
It is believed that Internet data is enabled, when identifying the corresponding first base station of current single signaling Internet data and the second base station and acquisition signaling
Between the step of.
In the present embodiment, need further to identify obtained track to whether being effective track pair, when first base station with
The distance d of second base station be greater than first base station covering radius r1 and the second base station covering radius r2 and when, show the track to being
Effective track pair, into according to first base station with the second base station at a distance from, the coverage area of first base station and the second base station and
Signaling time is acquired, the step of track pair is generated；When first base station covers half no more than first base station with the second base station distance d
Diameter r1 and the second base station covering radius r2 and when, show the track to being invalid track pair, current signaling Internet data is to this Shen
Please be for scheme it is invalid, next signaling Internet data is re-used as current signaling Internet data, starts next round
Identification process.In simple terms, effective track pair is calculated from the point sequence of track, it is assumed that the coverage area of two base stations does not have intersection
Be exactly it is primary effective advance, effective track is to meeting formula d > r1+r2, in which: d indicates central point between two base stations
Distance, r1, r2 respectively indicate the covering radius of two base stations.
In one of the embodiments, track to characteristic include maximum distance, minimum range and average distance and
Maximum speed, minimum speed and average speed update default cluster learning algorithm training pattern according to track to characteristic
Include:
Step 1: same trajectories are obtained to corresponding friction speed data, generate characteristic training set.
Step 2: default cluster learning algorithm training pattern is updated to characteristic training set.
Because each track to it is corresponding be two base stations, might have different practical sections, institute between two base stations
The training set of the track pair is used as with the record all users (terminal) in the same track pair.Such purpose is to protect
Each track pair is demonstrate,proved, has different travel speeds under different trip mode difference road conditions.
In one of the embodiments, according to updated cluster learning algorithm training pattern, identify that trip mode includes:
Step 1: the model that multiple clusters are preset in updated cluster learning algorithm training pattern is updated, wherein each cluster
Model carry different velocity characteristic data.
Step 2: the corresponding velocity characteristic data of trip mode to be selected are preset in acquisition.
Step 3: according to the model of multiple clusters of update and presetting the corresponding velocity characteristic data of trip mode to be selected,
Identify trip mode.
Different velocity characteristic data are carried in the model of each cluster, are based on the corresponding velocity characteristic of trip mode to be selected
Trip mode can be recognized accurately in data.The model of default cluster refers to based on the first of conventional terminal (user) track data building
The model of beginning cluster, firstly, the model of each default cluster is updated first according to truthful data cluster learning algorithm training result,
The model of updated each cluster can accurate characterization present terminal (user) track data, that is, carry characterization present terminal (use
Family) according to the corresponding velocity characteristic data of trip mode to be selected trip side can be recognized accurately in corresponding velocity characteristic data
Formula.Non-essential, the model of default cluster may include the mould of the default cluster of private car, bus, bicycle, walking and subway
Type, updates the model of 5 clusters according to cluster learning algorithm training result, obtain private car, bus, bicycle, walking with
And the corresponding velocity characteristic data of subway trip mode, according to the model of updated 5 clusters and private car, bus, from
Driving, walking and the corresponding velocity characteristic data of subway trip mode identify trip mode.
In one of the embodiments, according to track to characteristic update default cluster learning algorithm training pattern it
Before, further includes:
Step 1: the base station data of history signaling Internet data and historical communication base station is obtained.
Step 2: according to history signaling Internet data and history base station data, historical track pair is generated.
Step 3: it obtains historical track and default clustering learning is generated according to historical track to characteristic to characteristic
Algorithm training pattern.
In one of the embodiments, according to cluster learning algorithm processing result, after identifying trip mode, further includes:
The true trip mode of feedback is received as a result, according to true trip mode as a result, to the model continuous learnings of multiple clusters and updating.
The process that the model of multiple clusters can keep a continuous learning to update, so that model becomes closer to true value.
True trip mode result can be the data that user is uploaded by terminal, such as it is public transport that user, which uploads this trip mode,
Vehicle services further according to true trip mode model continuous learning and update as a result, to cluster.Specifically, with daily signaling
The increase of Internet data, is continuously increased sample, to model continuous learning, keeps model abundant enough to the training of each track pair,
So as to adapt to the travel speed of the vehicles under different road conditions, model will be more and more accurate, and can have timeliness,
For example the section is just repaired the roads, during which speed can reduce.
It should be understood that although each step in the flow chart of Fig. 2 and Fig. 4 is successively shown according to the instruction of arrow,
But these steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly state otherwise herein, these
There is no stringent sequences to limit for the execution of step, these steps can execute in other order.Moreover, in Fig. 2 and Fig. 4
At least part step may include that perhaps these sub-steps of multiple stages or stage are not necessarily same to multiple sub-steps
One moment executed completion, but can execute at different times, and the execution in these sub-steps or stage sequence is also not necessarily
Be successively carry out, but can at least part of the sub-step or stage of other steps or other steps in turn or
Alternately execute.
5 clustering models non-essential, that the track of generation trains training set and its clustering learning method, can pass through
The characteristic of terminal identity identification marking (cell-phone number) and base station itself is verified.
Such as: subway trip mode
Assuming that certain track will judge that one of cluster is subway to the model for having trained 5 clusters based on cluster, need to only look for
The cell-phone number that can be clearly gone on a journey out by subway, then extracts in the signaling Internet data input model of cell-phone number and predicts ownership
Cluster, which is subway.How to determine cell-phone number in some track pair or continuous path to being above subway trip? first,
Judge live it is whether close with the base station distance of subway with the base station in place of working, such as table 1 live and work point base stations with it is iron-based
Stop spacing is from close；Second, continuous path to whether it is most of belong to subway base station, if 2 track of table is to upper base station；It is comprehensive this two
Point, so that it may clearly judge.
The inhabitation of table 1 and work point base stations and subway base station distance
2 track of table is to upper base station
| Serial number | Subway station name | The shortest distance (m) |
| 1 | A subway station | 284 |
| 2 | B subway station | 439 |
| 3 | C subway station | 207 |
As shown in figure 5, a kind of trip mode identification device, device include:
Data acquisition module 200, for obtaining the base station data of signaling Internet data and communication base station；
Track is to generation module 400, for generating track pair according to signaling Internet data and base station data；
Training module 600 is clustered, for obtaining track to characteristic, clustering learning is carried out to characteristic according to track
Algorithm process, track include speed to characteristic；
Identification module 800, for identifying trip mode according to cluster learning algorithm processing result.
Above-mentioned trip mode identification device, data acquisition module 200 acquire the base of signaling Internet data and communication base station
It stands data, track generates track pair to generation module 400 according to the collected data, and cluster training module 600 extracts track to spy
Data are levied, track includes speed to characteristic, is based on velocity characteristic, updates default cluster learning algorithm training pattern, obtains
Updated cluster learning algorithm training pattern, identification module 800 identify trip side according to cluster learning algorithm processing result
Formula is carried using speed as the data of core in cluster learning algorithm training pattern in the updated, and different trip modes is most
Therefore difference in terms of apparent and speed is based on updated cluster learning algorithm training pattern, can be recognized accurately
Line mode.
Signaling Internet data includes terminal identity identification marking, communication base station identification in one of the embodiments,
Mark and acquisition signaling time, base station data include base station identity identification marking, latitude and longitude of base station data and base station covering
Range.
Track is also used to be identified current single according to signaling Internet data to generation module 400 in one of the embodiments,
The corresponding first base station of signaling Internet data and the second base station and acquisition signaling time；According to base station data, first is obtained
The corresponding longitude and latitude data in base station and the second base station；According to the longitude and latitude data of acquisition, first base station and the second base station are calculated
Distance；According to base station data, the coverage area of first base station and the second base station is obtained；According to first base station and the second base station away from
Coverage area and acquisition signaling time from, first base station and the second base station, generate track pair.
In one of the embodiments, track to generation module 400 be also used to judge first base station and the second base station away from
From whether greater than the sum of first base station covering radius and the second base station covering radius；If more than into according to first base station and the
Distance, first base station and the coverage area of the second base station and acquisition signaling time of two base stations, generate the operation of track pair；If
It is not more than, using next single signaling Internet data as current single signaling Internet data, returns according to signaling Internet data, know
The corresponding first base station of not current single signaling Internet data and the second base station and the operation for acquiring signaling time.
In one of the embodiments, track to characteristic include maximum distance, minimum range and average distance and
Maximum speed, minimum speed and average speed, cluster training module 600 are also used to obtain same trajectories to corresponding not synchronized
Degree evidence generates characteristic training set, updates default cluster learning algorithm training pattern to characteristic training set.
Identification module 800 is also used to update updated cluster learning algorithm training pattern in one of the embodiments,
In preset the models of multiple clusters, wherein the model of each cluster carries different velocity characteristic data, and trip side to be selected is preset in acquisition
The corresponding velocity characteristic data of formula according to the model of multiple clusters of update and preset the corresponding velocity characteristic of trip mode to be selected
Data identify trip mode.
Above-mentioned trip mode identification device in one of the embodiments, further include:
Learn update module, for receiving the true trip mode result of feedback；According to true trip mode as a result, to more
The model continuous learning of a cluster and update.
Specific about trip mode identification device limits the limit that may refer to above for trip mode recognition methods
Fixed, details are not described herein.Modules in above-mentioned trip mode identification device can fully or partially through software, hardware and its
Combination is to realize.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also be with
It is stored in the memory in computer equipment in a software form, in order to which processor calls the above modules of execution corresponding
Operation.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 6.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is calculated for storing the model for presetting multiple clusters, trip mode characteristic to be selected and default clustering learning
The data such as method training pattern.The network interface of the computer equipment is used to communicate with external terminal by network connection.The meter
To realize a kind of trip mode recognition methods when calculation machine program is executed by processor.
It will be understood by those skilled in the art that structure shown in Fig. 6, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment is provided, including memory, processor and storage are on a memory
And the computer program that can be run on a processor, processor perform the steps of when executing computer program
Obtain the base station data of signaling Internet data and communication base station；
According to signaling Internet data and base station data, track pair is generated；
It obtains track and default cluster learning algorithm training pattern, rail is updated according to track to characteristic to characteristic
Mark includes speed to characteristic；
According to updated cluster learning algorithm training pattern, trip mode is identified.
Signaling Internet data includes terminal identity identification marking, communication base station identification in one of the embodiments,
Mark and acquisition signaling time, base station data include base station identity identification marking, latitude and longitude of base station data and base station covering
Range.
It is also performed the steps of when processor executes computer program in one of the embodiments,
According to signaling Internet data, identify the corresponding first base station of current single signaling Internet data and the second base station and
Acquire signaling time；According to base station data, first base station and the corresponding longitude and latitude data in the second base station are obtained；According to the warp of acquisition
Latitude data calculates first base station at a distance from the second base station；According to base station data, acquisition first base station is covered with the second base station
Lid range；According to first base station at a distance from the second base station, the coverage area and acquisition signaling of first base station and the second base station
Time generates track pair.
Judgement first base station is also performed the steps of when processor executes computer program in one of the embodiments,
With whether be greater than the sum of first base station covering radius and the second base station covering radius at a distance from the second base station；If more than into root
According to first base station at a distance from the second base station, the coverage area and acquisition signaling time of first base station and the second base station, generate
The step of track pair；If being not more than, using next single signaling Internet data as current single signaling Internet data, basis is returned to
Signaling Internet data identifies the corresponding first base station of current single signaling Internet data and the second base station and acquisition signaling time
The step of.
In one of the embodiments, track to characteristic include maximum distance, minimum range and average distance and
Maximum speed, minimum speed and average speed, processor also performs the steps of when executing computer program obtains identical rail
Mark generates characteristic training set to corresponding friction speed data, updates default clustering learning to characteristic training set and calculates
Method training pattern.
It is also performed the steps of when processor executes computer program in one of the embodiments,
Update the model that multiple clusters are preset in updated cluster learning algorithm training pattern, wherein the model of each cluster
Carry different velocity characteristic data；The corresponding velocity characteristic data of trip mode to be selected are preset in acquisition；According to the multiple of update
The model of cluster and the corresponding velocity characteristic data of trip mode to be selected are preset, identifies trip mode.
It is also performed the steps of when processor executes computer program in one of the embodiments,
Obtain the base station data of history signaling Internet data and historical communication base station；According to history signaling Internet data with
And history base station data, generate historical track pair；Historical track is obtained to characteristic, according to historical track to characteristic,
Generate default cluster learning algorithm training pattern.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of when being executed by processor
Obtain the base station data of signaling Internet data and communication base station；
According to signaling Internet data and base station data, track pair is generated；
It obtains track and default cluster learning algorithm training pattern, rail is updated according to track to characteristic to characteristic
Mark includes speed to characteristic；
According to updated cluster learning algorithm training pattern, trip mode is identified.
Signaling Internet data includes terminal identity identification marking, communication base station identification in one of the embodiments,
Mark and acquisition signaling time, base station data include base station identity identification marking, latitude and longitude of base station data and base station covering
Range.
It is also performed the steps of when computer program is executed by processor in one of the embodiments,
According to signaling Internet data, identify the corresponding first base station of current single signaling Internet data and the second base station and
Acquire signaling time；According to base station data, first base station and the corresponding longitude and latitude data in the second base station are obtained；According to the warp of acquisition
Latitude data calculates first base station at a distance from the second base station；According to base station data, acquisition first base station is covered with the second base station
Lid range；According to first base station at a distance from the second base station, the coverage area and acquisition signaling of first base station and the second base station
Time generates track pair.
It is also performed the steps of when computer program is executed by processor in one of the embodiments, and judges the first base
It stands and whether is greater than the sum of first base station covering radius and the second base station covering radius at a distance from the second base station；If more than entrance
It is raw according to first base station at a distance from the second base station, the coverage area and acquisition signaling time of first base station and the second base station
At the step of track pair；If being not more than, using next single signaling Internet data as current single signaling Internet data, root is returned to
It is believed that Internet data is enabled, when identifying the corresponding first base station of current single signaling Internet data and the second base station and acquisition signaling
Between the step of.
In one of the embodiments, track to characteristic include maximum distance, minimum range and average distance and
It is identical also to perform the steps of acquisition for maximum speed, minimum speed and average speed when computer program is executed by processor
Track generates characteristic training set to corresponding friction speed data, updates default clustering learning to characteristic training set
Algorithm training pattern.
It is also performed the steps of when computer program is executed by processor in one of the embodiments,
Update the model that multiple clusters are preset in updated cluster learning algorithm training pattern, wherein the model of each cluster
Carry different velocity characteristic data；The corresponding velocity characteristic data of trip mode to be selected are preset in acquisition；According to the multiple of update
The model of cluster and the corresponding velocity characteristic data of trip mode to be selected are preset, identifies trip mode.
It is also performed the steps of when computer program is executed by processor in one of the embodiments,
Obtain the base station data of history signaling Internet data and historical communication base station；According to history signaling Internet data with
And history base station data, generate historical track pair；Historical track is obtained to characteristic, according to historical track to characteristic,
Generate default cluster learning algorithm training pattern.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Instruct relevant hardware to complete by computer program, computer program to can be stored in a non-volatile computer readable
It takes in storage medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, this Shen
Please provided by any reference used in each embodiment to memory, storage, database or other media, may each comprise
Non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield all should be considered as described in this specification.
Above embodiments only express the several embodiments of the application, and the description thereof is more specific and detailed, but can not
Therefore it is construed as limiting the scope of the patent.It should be pointed out that for those of ordinary skill in the art,
Under the premise of not departing from the application design, various modifications and improvements can be made, these belong to the protection scope of the application.
Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of trip mode recognition methods, which comprises
Obtain the base station data of signaling Internet data and communication base station；
According to the signaling Internet data and the base station data, track pair is generated；
It obtains track and default cluster learning algorithm training pattern, institute is updated according to the track to characteristic to characteristic
Stating track includes speed to characteristic；
According to updated cluster learning algorithm training pattern, trip mode is identified.
2. the method according to claim 1, wherein the signaling Internet data includes terminal identity identification mark
Knowledge, communication base station identity identification and acquisition signaling time, the base station data includes base station identity identification marking, base station
Longitude and latitude data and base station range.
3. according to the method described in claim 2, it is characterized in that, described according to the signaling Internet data and the base station
Data generate track to including:
According to the signaling Internet data, identify the corresponding first base station of current single signaling Internet data and the second base station and
Acquire signaling time；
According to the base station data, the corresponding longitude and latitude data of the first base station and second base station are obtained；
According to the longitude and latitude data of acquisition, the first base station is calculated at a distance from second base station；
According to the base station data, the coverage area of the first base station Yu second base station is obtained；
According to the first base station at a distance from second base station, the coverage area of the first base station and second base station
And the acquisition signaling time, generate track pair.
4. according to the method described in claim 3, it is characterized in that, described according to the first base station and second base station
Distance, the coverage area of the first base station and second base station and the acquisition signaling time, generate track to before,
Further include:
Judge whether the first base station is greater than first base station covering radius at a distance from second base station and covers with the second base station
The sum of lid radius；
If more than, into it is described according to the first base station at a distance from second base station, the first base station and described the
The coverage area of two base stations and the acquisition signaling time generate the step of track pair；
If being not more than, using next single signaling Internet data as current single signaling Internet data, return described according to
Signaling Internet data identifies the corresponding first base station of current single signaling Internet data and the second base station and acquisition signaling time
The step of.
5. the method according to claim 1, wherein the track includes maximum distance, minimum to characteristic
Distance and average distance and maximum speed, minimum speed and average speed, it is described according to the track to characteristic, more
Newly default cluster learning algorithm training pattern includes:
Same trajectories are obtained to corresponding friction speed data, generate characteristic training set；
Default cluster learning algorithm training pattern is updated to the characteristic training set.
6. the method according to claim 1, wherein described according to updated cluster learning algorithm training mould
Type, identification trip mode include:
Update the model that multiple clusters are preset in the updated cluster learning algorithm training pattern, wherein the model of each cluster
Carry different velocity characteristic data；
The corresponding velocity characteristic data of trip mode to be selected are preset in acquisition；
According to the model of multiple clusters of update and it is described preset the corresponding velocity characteristic data of trip mode to be selected, identification trip
Mode.
7. the method according to claim 1, wherein described update default gather to characteristic according to the track
Before class learning algorithm training pattern, further includes:
Obtain the base station data of history signaling Internet data and historical communication base station；
According to the history signaling Internet data and the history base station data, historical track pair is generated；
It obtains historical track and default cluster learning algorithm instruction is generated according to the historical track to characteristic to characteristic
Practice model.
8. a kind of trip mode identification device, described device include:
Data acquisition module, for obtaining the base station data of signaling Internet data and communication base station；
Track is to generation module, for generating track pair according to the signaling Internet data and the base station data；
Training module is clustered, for obtaining track to characteristic, default cluster is updated to characteristic according to the track and is learned
Algorithm training pattern is practised, the track includes speed to characteristic；
Identification module, for identifying trip mode according to updated cluster learning algorithm training pattern.
9. a kind of computer equipment, including memory and processor, the memory are stored with computer program, feature exists
In the step of processor realizes any one of claims 1 to 7 the method when executing the computer program.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of method described in any one of claims 1 to 7 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811032855.7A CN109272032A (en) | 2018-09-05 | 2018-09-05 | Trip mode recognition methods, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811032855.7A CN109272032A (en) | 2018-09-05 | 2018-09-05 | Trip mode recognition methods, device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109272032A true CN109272032A (en) | 2019-01-25 |
Family
ID=65187248
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811032855.7A Pending CN109272032A (en) | 2018-09-05 | 2018-09-05 | Trip mode recognition methods, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109272032A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109947822A (en) * | 2019-03-19 | 2019-06-28 | 成都市公安科学技术研究所 | A kind of method of movement track pattern-recognition working condition |
| CN111190984A (en) * | 2019-12-30 | 2020-05-22 | 上海炬宏信息技术有限公司 | Job and residence extraction method and device and computer readable storage medium |
| CN111653093A (en) * | 2020-05-29 | 2020-09-11 | 南京瑞栖智能交通技术产业研究院有限公司 | Urban trip mode comprehensive identification method based on mobile phone signaling data |
| CN111653096A (en) * | 2020-05-29 | 2020-09-11 | 南京瑞栖智能交通技术产业研究院有限公司 | Urban trip mode identification method based on mobile phone signaling data |
| CN112637781A (en) * | 2020-11-23 | 2021-04-09 | 广州大学 | User traffic mode judging method based on base station track |
| CN112653995A (en) * | 2019-10-12 | 2021-04-13 | 中国移动通信有限公司研究院 | User identity recognition method and device and computer readable storage medium |
| CN113192341A (en) * | 2021-04-02 | 2021-07-30 | 天地（常州）自动化股份有限公司 | Estimation method for average speed of moving target in underground area positioning interval |
| WO2022012486A1 (en) * | 2020-07-13 | 2022-01-20 | 华为技术有限公司 | Method and apparatus for determining intimate contact with terminal device |
| CN114245314A (en) * | 2021-12-17 | 2022-03-25 | 高创安邦(北京)技术有限公司 | Personnel trajectory correction method and device, storage medium and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106197458A (en) * | 2016-08-10 | 2016-12-07 | 重庆邮电大学 | A kind of cellphone subscriber's trip mode recognition methods based on mobile phone signaling data and navigation route data |
| US20170048772A1 (en) * | 2015-08-14 | 2017-02-16 | Qualcomm Incorporated | Mobility enhancements for high speed scenarios |
| CN107784597A (en) * | 2017-09-19 | 2018-03-09 | 平安科技（深圳）有限公司 | Trip mode recognition methods, device, terminal device and storage medium |
| CN107845260A (en) * | 2017-10-26 | 2018-03-27 | 杭州东信北邮信息技术有限公司 | A kind of recognition methods of user's bus trip mode |
| CN109284773A (en) * | 2018-08-15 | 2019-01-29 | 西南交通大学 | Traffic trip endpoint recognition methods based on multilayer Agglomerative Hierarchical Clustering algorithm |
-
2018
- 2018-09-05 CN CN201811032855.7A patent/CN109272032A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170048772A1 (en) * | 2015-08-14 | 2017-02-16 | Qualcomm Incorporated | Mobility enhancements for high speed scenarios |
| CN106197458A (en) * | 2016-08-10 | 2016-12-07 | 重庆邮电大学 | A kind of cellphone subscriber's trip mode recognition methods based on mobile phone signaling data and navigation route data |
| CN107784597A (en) * | 2017-09-19 | 2018-03-09 | 平安科技（深圳）有限公司 | Trip mode recognition methods, device, terminal device and storage medium |
| CN107845260A (en) * | 2017-10-26 | 2018-03-27 | 杭州东信北邮信息技术有限公司 | A kind of recognition methods of user's bus trip mode |
| CN109284773A (en) * | 2018-08-15 | 2019-01-29 | 西南交通大学 | Traffic trip endpoint recognition methods based on multilayer Agglomerative Hierarchical Clustering algorithm |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109947822A (en) * | 2019-03-19 | 2019-06-28 | 成都市公安科学技术研究所 | A kind of method of movement track pattern-recognition working condition |
| CN112653995A (en) * | 2019-10-12 | 2021-04-13 | 中国移动通信有限公司研究院 | User identity recognition method and device and computer readable storage medium |
| CN112653995B (en) * | 2019-10-12 | 2023-03-28 | 中国移动通信有限公司研究院 | User identity recognition method and device and computer readable storage medium |
| CN111190984A (en) * | 2019-12-30 | 2020-05-22 | 上海炬宏信息技术有限公司 | Job and residence extraction method and device and computer readable storage medium |
| CN111190984B (en) * | 2019-12-30 | 2024-03-08 | 北京世纪高通科技有限公司 | Method and device for extracting job place and computer readable storage medium |
| CN111653093B (en) * | 2020-05-29 | 2022-06-17 | 南京瑞栖智能交通技术产业研究院有限公司 | Urban trip mode comprehensive identification method based on mobile phone signaling data |
| CN111653096A (en) * | 2020-05-29 | 2020-09-11 | 南京瑞栖智能交通技术产业研究院有限公司 | Urban trip mode identification method based on mobile phone signaling data |
| CN111653093A (en) * | 2020-05-29 | 2020-09-11 | 南京瑞栖智能交通技术产业研究院有限公司 | Urban trip mode comprehensive identification method based on mobile phone signaling data |
| WO2022012486A1 (en) * | 2020-07-13 | 2022-01-20 | 华为技术有限公司 | Method and apparatus for determining intimate contact with terminal device |
| CN112637781A (en) * | 2020-11-23 | 2021-04-09 | 广州大学 | User traffic mode judging method based on base station track |
| CN112637781B (en) * | 2020-11-23 | 2023-10-03 | 广州大学 | User traffic mode distinguishing method based on base station track |
| CN113192341A (en) * | 2021-04-02 | 2021-07-30 | 天地（常州）自动化股份有限公司 | Estimation method for average speed of moving target in underground area positioning interval |
| CN114245314A (en) * | 2021-12-17 | 2022-03-25 | 高创安邦(北京)技术有限公司 | Personnel trajectory correction method and device, storage medium and electronic equipment |
| CN114245314B (en) * | 2021-12-17 | 2024-01-05 | 高创安邦(北京)技术有限公司 | Personnel track correction method and device, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109272032A (en) | Trip mode recognition methods, device, computer equipment and storage medium | |
| CN106912018B (en) | Map matching method and system based on signaling track | |
| CN111343585B (en) | Mobile user track map matching method based on hidden Markov model | |
| CN105491532B (en) | A kind of mobile phone SIP signaling filtering method and apparatus for road network running state analysis | |
| CN105528359B (en) | For storing the method and system of travel track | |
| CN102521973B (en) | A kind of mobile phone switches the road matching method of location | |
| CN109996186A (en) | A kind of network coverage problem identification method and device, read/write memory medium | |
| CN107018493B (en) | Geographic position prediction method based on continuous time sequence Markov model | |
| CN109766777A (en) | Detection method, device, storage medium and the electronic equipment of abnormal track | |
| CN104217593B (en) | A kind of method for obtaining road condition information in real time towards mobile phone travelling speed | |
| CN109688532A (en) | A kind of method and device dividing city function region | |
| CN104884899A (en) | Method of determining trajectories through one or more junctions of a transportation network | |
| CN104596507A (en) | Method for determining travel track of mobile terminal | |
| CN105933857B (en) | A kind of mobile terminal locations prediction technique and device | |
| CN110598917B (en) | Destination prediction method, system and storage medium based on path track | |
| CN108806301A (en) | A kind of automatic identifying method of public bus information | |
| CN105635956A (en) | Positioning method and apparatus based on indoor virtual landmark | |
| CN112815948B (en) | Method, device, computer equipment and storage medium for identifying yaw mode | |
| CN108804539A (en) | A kind of track method for detecting abnormality under time and space double-visual angle | |
| CN110830915B (en) | Method and device for determining starting point position | |
| Jiang et al. | A framework of travel mode identification fusing deep learning and map-matching algorithm | |
| Chen et al. | A travel mode identification framework based on cellular signaling data | |
| Guo et al. | Vehicle travel path recognition in urban dense road network environments by using mobile phone data | |
| CN116952277A (en) | Potential position estimation method and system based on artificial intelligence and behavior characteristics | |
| CN116129643A (en) | Bus travel characteristic identification method, device, equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190125 |
|
| RJ01 | Rejection of invention patent application after publication |