CN112511971B - Travel mode identification method based on mobile phone signaling data - Google Patents
Travel mode identification method based on mobile phone signaling data Download PDFInfo
- Publication number
- CN112511971B CN112511971B CN202011366779.0A CN202011366779A CN112511971B CN 112511971 B CN112511971 B CN 112511971B CN 202011366779 A CN202011366779 A CN 202011366779A CN 112511971 B CN112511971 B CN 112511971B
- Authority
- CN
- China
- Prior art keywords
- travel
- trip
- track
- bus
- range
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/025—Services making use of location information using location based information parameters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/025—Services making use of location information using location based information parameters
- H04W4/027—Services making use of location information using location based information parameters using movement velocity, acceleration information
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/029—Location-based management or tracking services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W8/00—Network data management
- H04W8/18—Processing of user or subscriber data, e.g. subscribed services, user preferences or user profiles; Transfer of user or subscriber data
- H04W8/183—Processing at user equipment or user record carrier
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Remote Sensing (AREA)
- Mobile Radio Communication Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a travel mode identification method based on mobile phone signaling data, which solves the problems that multi-source data are difficult to obtain, mixed data of multiple travel modes at a time are difficult to identify, data are wasted and the like in the traditional machine learning method. The invention adopts multi-source data such as mobile phone signaling, public transport line network identification, planning time-interval speed threshold interval establishment and the like to identify the user travel mode for many times. Calculating the time-interval traffic running speed of the city where the data are located, taking the degree of fit of the data and a bus line network where the city is located as a reference, and performing primary identification on the data by using different threshold conditions according to the time interval of the data for each trip; and carrying out secondary identification on the unidentified data by referring to the OD distance of the single trip and the above conditions. The method effectively utilizes a large amount of unidentifiable data, greatly reduces the interference of multiple traffic modes on the travel speed selection interval value, and improves the identification accuracy.
Description
Technical Field
The invention belongs to the technical field of traffic planning, and particularly relates to a travel mode identification method based on mobile phone signaling data.
Background
The travel mode identification plays a fundamental role in the academic fields of traffic planning, city planning and the like. In the current mobile phone signaling data travel mode identification, identification methods such as threshold value distinguishing, machine learning and neural network exist; however, the disadvantages of the handset signaling data itself include: the data granularity is coarse, and the short-distance trip data in a short time is inaccurate; the one-time travel data covers various travel modes; only the total time and not the segment time makes it impossible to calculate the acceleration and to slice the data. The method of machine learning and deep learning has the problems that a large amount of multi-source data are needed for learning, the learning cost is high and the learning is difficult to obtain; due to the defects of the signaling data, the identification degree of the walking and the bus is not ideal, the process is complex, and the applicability is low. In addition, the method for distinguishing the threshold has the problems that the value of the key characteristic threshold is not accurate, the elasticity is lacked, and the data utilization rate is not high.
In summary, a new method for identifying a travel mode based on mobile phone signaling data is needed.
Disclosure of Invention
The invention aims to provide a travel mode identification method based on mobile phone signaling data, so as to solve one or more technical problems. The method effectively utilizes a large amount of unidentifiable data, greatly reduces the interference of a plurality of traffic modes on the travel speed selection interval value, and can improve the identification accuracy to a certain extent.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention relates to a travel mode identification method based on mobile phone signaling data, which specifically comprises the following steps:
step 1, acquiring mobile phone signaling data; based on the longitude and latitude and time slice information of the user in the mobile phone signaling data, POI point elements of a starting point and a terminal point of single trip of the user are obtained, and a middle section track of the single trip of the user is obtained; acquiring the position information of a city where a user is located;
step 2, obtaining a fitted travel track based on POI point elements of a starting point and a finishing point of a user and a middle section track; based on the fitted travel track, obtaining the single travel distance and travel time of the user, and calculating to obtain the average speed of the single travel of the user;
step 3, acquiring a speed threshold range and a distance threshold range of each travel mode in a peak time period and a peak-smoothing time period of a city where a user is located; generating a buffer area based on the urban public transport operation line network, wherein the buffer area is used for acquiring track points of the fitted travel track on the urban public transport operation line; acquiring the coincidence rate of the travel track and the urban bus operation line network based on the fitted travel track and the buffer area; judging whether the travel time of the user is in a peak time period or a flat time period based on the single travel distance and travel time of the user and the position information of the city where the user is located;
step 4, judging the user trip mode for one time based on a preset threshold condition, finishing trip mode identification if a judgment result is obtained for one time, and otherwise, skipping to execute step 5; the threshold conditions of the first judgment are the travel distance, the average speed, the travel time period and the coincidence rate of the travel track and the urban public transport operation line network of the user;
and 5, carrying out secondary judgment on the problem data out of the threshold condition of the primary judgment, wherein the secondary judgment comprises the following steps: obtaining a travel OD distance of the problem data based on the OD point longitude and latitude of the fitted travel track obtained in the step 2; performing secondary discrimination on the problem data to obtain a secondary discrimination result, and finishing travel mode identification; the threshold conditions of the secondary judgment are user travel distance, average speed, OD distance, and coincidence rate of travel track and urban public transport operation line network.
Further, in step 1, the mobile phone signaling data comes from the unicom mobile phone signaling data of the smart footprint smart platform, and the method includes: the longitude and latitude of the starting and ending point of the single trip, the longitude and latitude, the sequence and the number of each path node in the single trip track, the trip starting time and the trip ending time.
Further, in step 2, in the mobile phone signaling data, a subway trip mode has an individual trip field; and screening out single subway trip samples of the user to obtain a database of trip modes to be identified.
Further, in step 2, the specific step of obtaining the fitted travel trajectory based on the POI point elements of the start point and the end point of the user and the middle section trajectory includes:
fitting and connecting the first section track and the tail section track of the user with the middle section track to obtain a fitted travel track;
the middle section track is a result of fitting the urban road network by the position of a base station when a user uses a mobile phone and signals are interacted with the base station; the first-segment track is a result of the connection of a POI point element of a starting point of the user and a POI point element of a starting point of the middle-segment track, and the last-segment track is a result of the connection of a POI point element of a finishing point of the user and a POI point element of an ending point of the middle-segment track; and cleaning the data with the sum of the distances between the first section of track and the tail section of track being more than or equal to 1 km.
Further, in step 3, according to the light-weight route planning data of the Baidu open platform, obtaining the speed threshold value ranges of all travel modes in the urban peak time period and the urban peak-off time period; and for the travel distance, classifying by using K-means unsupervised learning, and obtaining the distance threshold range of the travel of each traffic mode in the city by taking the classification result as reference.
Further, in step 3, the specific steps of obtaining the coincidence rate of the travel track and the urban public transport operation line network based on the fitted travel track and the buffer area comprise:
(1) selecting track point data of the single trip track in the buffer area by using the formed buffer area to form coincident track data of the single trip track and the urban public transport operation line network;
(2) and comparing the obtained coincidence track data with the single trip track data to obtain the coincidence rate of the single trip track and the urban public transport operation line network.
Further, in step 4, the specific step of primary discrimination includes:
the bus trip mode is distinguished and includes: the threshold conditions comprise a bus speed threshold range, a bus distance threshold range, a trip time period, a track and urban bus operation line network coincidence rate range; wherein, the bus speed threshold value includes: the bus speed threshold value range is obtained when the trip time period is a peak-off time period, and the bus speed threshold value range is obtained when the trip time period is a peak time period; when the bus trip is within the threshold value ranges of the bus, the bus trip is a bus trip mode;
the judging of the walking travel mode comprises the following steps: the threshold condition comprises a walking speed threshold range and a walking distance threshold range; if the current trip falls within the range of each threshold value of walking, the trip is a walking trip mode;
the bicycle trip mode is distinguished and includes: the threshold condition comprises a bicycle speed threshold range and a bicycle distance threshold range; falling within the range of each threshold value of the bicycle is a bicycle trip mode.
Further, in step 4, the specific step of primary discrimination further includes:
according to the judgment result of the bus trip mode, private cars in a bus trip threshold interval are judged, and the threshold conditions comprise a private car speed threshold range, a private car distance threshold range, a trip time period range and a coincidence rate range of a track and a city bus operation line network; wherein the private car speed threshold range comprises: a private car speed threshold range when the travel time period is a peak-off time period and a private car speed threshold range when the travel time period is a peak-off time period; the mode of the private car going out is determined when the vehicle falls within the threshold value range of the private car;
judging private cars outside the bus trip threshold interval, wherein the threshold conditions comprise a private car speed threshold range and a private car distance threshold range; wherein the private car speed threshold range comprises: a private car speed threshold range when the travel time period is a flat peak time period and a private car speed threshold range when the travel time period is a peak time period; the mode of going out of the private car is determined when the signal is within the range of each threshold value of the private car;
the method for judging the travel mode of the motorcycle comprises the following steps: the threshold condition comprises a motorcycle speed threshold range and a motorcycle distance threshold range; and if the signal value falls within the range of each threshold value of the motorcycle, the mode of motorcycle travel is adopted.
Further, in step 5, the specific step of secondary discrimination includes:
performing secondary judgment on walking, wherein the threshold conditions comprise a walking distance threshold range and a walking OD distance threshold range; if the current trip is within the range of each secondary discrimination threshold value of walking, the trip is a walking trip mode;
performing secondary judgment on the bicycle, wherein the threshold conditions comprise a bicycle distance threshold range and a bicycle OD distance threshold range; if the bicycle falls within the range of each secondary discrimination threshold value of the bicycle, the bicycle is taken as a bicycle travel mode;
carrying out secondary judgment on the bus, wherein the threshold conditions comprise a bus speed threshold range, a bus distance threshold range, a bus OD distance threshold range and a coincidence rate range of a track and an urban bus operation line network; if the bus trip is within the range of each secondary judging threshold value of the bus, the bus trip is a bus trip mode;
the secondary judgment of the private car comprises the following steps: according to the secondary judgment result of the bus, judging private cars in a bus trip threshold range, wherein the threshold conditions comprise a private car speed threshold range, a private car distance threshold range, a private car OD distance threshold range and a coincidence rate of a track and a city bus operation line network; judging private cars outside the bus trip threshold interval, wherein the threshold conditions comprise a private car OD distance threshold range and a private car distance threshold range; and if the traveling mode falls within the range of each secondary judgment threshold value of the private car, the traveling mode is the private car traveling mode.
Compared with the prior art, the invention has the following beneficial effects:
the method provided by the invention avoids the problems that the traditional machine learning method is difficult to obtain multi-source data, the mixed data of multiple trip modes at a time is difficult to identify, the data is wasted and the like. Aiming at the problems in the prior art, the method adopts multi-source data such as mobile phone signaling, bus line network identification, establishment of speed threshold intervals in planning time-sharing periods (light-weight route rule division based on a Baidu map open platform) and the like to identify the user travel modes for multiple times. Calculating the time-interval traffic running speed of the city where the data are located, taking the degree of fit of the data and a bus line network where the city is located as a reference, and performing primary identification on the data by using different threshold conditions according to the time interval of the data for each trip; and (4) carrying out secondary identification on the unidentified data by referring to the OD distance of the single trip and the above conditions. The method effectively utilizes a large amount of unidentifiable data, greatly reduces the interference of multiple traffic modes on the travel speed selection interval value, and improves the identification accuracy.
In the invention, because the data particles of the signaling data are relatively coarse, the first section of path from the starting point and the last section of path to the end point are already in the field and can not be fitted on the road network, so that the three sections of track sub-segments of one trip are spliced, the fitted moving track is arranged in the middle, and the identification accuracy can be improved.
In the invention, the mobile phone signaling data come from the intelligent footprint, a modeling calculation request is submitted to the intelligent footprint DaaS calculation cluster, demographic data set results with different dimensional combinations of positions, time, attributes and the like are generated, and the travel data of residents are formed. The speed threshold data is obtained by analyzing the equal time circle data acquired by the API (application program interface) of the Baidu open platform, and the distance threshold is obtained by analyzing the mobile phone signaling data sample by using K-means clustering. The invention has low difficulty in obtaining multi-source data, utilizes the data to the maximum extent and ensures the data quantity required by the experiment. The method for identifying the mobile phone signaling data travel mode has the advantages of easily-obtained multi-source data, simplified operation process and higher utilization rate of big data.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art are briefly introduced below; it is obvious that the drawings in the following description are some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
Fig. 1 is a schematic flow chart of a travel mode identification method based on mobile phone signaling data according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a three-segment travel trajectory principle according to an embodiment of the present invention;
FIG. 3 is a diagram illustrating a result of the primary determination according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating a second-order discrimination correction result according to an embodiment of the present invention.
Detailed Description
In order to make the purpose, technical effect and technical solution of the embodiments of the present invention clearer, the following clearly and completely describes the technical solution of the embodiments of the present invention with reference to the drawings in the embodiments of the present invention; it is to be understood that the described embodiments are part of the present invention. Other embodiments, which can be derived by one of ordinary skill in the art from the disclosed embodiments without inventive faculty, are intended to be within the scope of the invention.
Referring to fig. 1, a travel mode identification method based on mobile phone signaling data according to an embodiment of the present invention specifically includes the following steps:
step 1, mobile phone signaling data come from the communicated mobile phone signaling data of the intelligent footprint extremely intelligent platform, and a single trip track of a user is extracted according to longitude and latitude and time slice information of the user in the data; due to the relatively coarse data particles of the signaling data, the first section of path from the starting point and the last section of path to the end point are already inside the field and may not be fitted on the road network, so that three sections of track sub-segments of one trip are spliced, and the middle is a fitted moving track.
Step 1 of the embodiment of the present invention specifically includes:
step 1-1, the moving track of the middle section is the result of the user using a mobile phone, the signal is interacted with a base station, and finally the result is fitted with an urban road network through the position of the base station;
step 1-2, connecting a POI point element for recording a starting point of the user trip and a POI point element for recording a starting point of the user trip track in the first segment;
and 1-3, connecting the POI point element of which the tail section is recorded as the end point of the travel track of the user with the POI point element of which the travel starting point of the user is recorded.
Step 2, connecting the track of the head and tail sections with the middle fitting movement track to obtain an approximately real travel track; and cleaning the data with the sum of the distances between the head and the tail sections being more than 1km, and cleaning the trip data of the user with too long distance.
And 3, generating a buffer area by the urban public transport operation line network.
And 4, the subway has an independent signal base station, and a single subway trip sample of the user is screened out according to the judged trip mode to form a subway trip database.
And step 5, obtaining the average speed avg _ speed of the single trip of the user and the coincidence rate busoline _ ratio of the trip track and the bus line network according to the distance and the time of the single trip of the user obtained in the step 2 and the buffer zone in the step 3.
In the embodiment of the present invention, step 5 specifically includes:
step 5-1, comparing the coincidence rate busline _ ratio of the public traffic line network with the user intermediate fitting track mid _ dis according to the coincidence distance ch _ dis of the user intermediate fitting track mid _ dis and the public traffic line network;
step 5-2, the fitting track of the user is formed by connecting multiple sections of tracks, and longitude and latitude information of small sections of track coordinate points contained in the buffer area formed in the step 3 is connected into track information, so that a part of distance is slightly lost or increased when the superposition length is calculated;
and 5-3, selecting the track points according to the principle that the starting points of the small sections of tracks in the buffer area are selected as the starting points and the ending points of the ch _ dis.
And 6, judging whether the trip is in a peak time period or a peak-off time period according to the starting time still and the ending time etime of the single trip of the user in the step 2.
In the embodiment of the present invention, step 6 specifically includes:
step 6-1, judging the value of the time interval sd: if sd is judged to be a peak, sd is 1, and if sd is judged to be a flat peak, sd is 0;
step 6-2, the suggested value of the peak time is 07:30: 00-09: 30:00, 17:30: 00-19: 30: 00;
step 6-3, judging that the peak value removing principle is still (09: 30: 00), etime (07: 30: 00) or still (19: 30: 00), and etime (17: 30: 00);
and 7, obtaining the speed threshold range speed of each travel mode in the urban peak and peak leveling period according to the light-weight route planning data of the Baidu open platform.
In the embodiment of the present invention, step 7 specifically includes:
step 7-1, establishing each travel mode route planning starting point X by using FME (FeatureMeanipulateEngine);
7-1-1, creating a data base point X by using a Creator module;
step 7-1-2, using AttributeManager to endow the point element X in step 7-1-1 with space coordinate X (X _ lon, X _ lat) and make it have space attribute;
7-1-3, spatializing the point coordinate X in the step 7-1-2 by using a VertexCrarator;
step 7-1-4, using CoordinateSystemSetter to define the point coordinate X spatialized in step 7-1-3 as the WGS84 coordinate;
7-1-5, projecting the WGS84 coordinates of the point element X in the step 7-1-4 into UTM coordinates by using Reprojector;
7-2, creating a buffer area, creating a certain number of route end points Yk in the buffer area, and endowing the route end points Yk with space attributes;
step 7-2-1, creating a buffer area of a required area by using buffer;
step 7-2-2, creating a certain number of uniformly distributed point elements Yk in the buffer created in step 7-2-1 by using 2 DGridACCUMULATOR;
step 7-2-3, re-projecting the UTM84 coordinates of point element Yk in step 7-2-2 into WGS84 coordinates using Reprojector;
step 7-2-4, using CoordinateExtractor to add coordinate values (Yk _ lon, Yk _ lat) to the point Yk with spatial attribute in step 7-2-3;
step 7-3, using FeatureMerger to associate the point element Yk (Yk _ lon, Yk _ lat) in the buffer created by 7-2-4 with the data base point X (X _ lon, X _ lat) created by 7-1-4, and generating a planned path point with OD point X (X _ lon, X _ lat), Yk (Yk _ lon, Yk _ lat), k being (1, 2, 3, 4 … … k);
7-4, acquiring data by using an HTTPCaller according to a lightweight route planning interface of a WEB service API of the Baidu map open platform;
7-4-1, acquiring distance and time data of each travel mode route planning in a peak time period and a peak leveling time period by using a developer key;
7-4-2, formatting the JSON file by using a JSONFragmenter;
7-5, calculating the acquired distance and time data of the route planning generated by each pair of OD points by using an AttributeManager to obtain the average travelling speed of each pair of OD points in the range of the buffer area;
7-6, calculating a speed threshold value of each travel mode in each time period according to the average speed of each pair of OD point travel obtained in the step 7-5;
step 7-6-1, when the sd is calculated to be 0, taking an average value max _ avg _ speed 5% before the maximum value of each travel mode avg _ speed and an average value min _ avg _ speed 5% before the minimum value as a peak-leveling speed threshold speed (max _ avg _ speed to min _ avg _ speed) of each travel mode;
and 7-6-2, when the sd is 1, calculating an average value max _ avg _ speed 5% before the maximum value of each trip mode avg _ speed and an average value min _ avg _ speed 5% before the minimum value as peak speed threshold speeds (max _ avg _ speed to min _ avg _ speed) of each trip mode avg _ speed.
And 8, classifying the distance of the obtained data sample by using K-means unsupervised learning, and obtaining the distance threshold range of the travel of each traffic mode of the city by taking the classification result as a reference.
And 9, preliminarily judging the single trip track of the user obtained in the step 6, wherein the threshold conditions are distance, average speed, time interval and coincidence rate of bus routes.
In the embodiment of the present invention, step 9 specifically includes:
and 9-1, judging the buses, wherein the threshold conditions comprise a bus speed threshold range bus _ speed, a bus distance threshold range bus _ distance, sd and a busline _ ratio. According to step 7, when sd is 0, bus _ speed takes a value of S1, and when sd is 1, bus _ speed takes a value of S2. According to step 8, the bus _ distance value is D1. busline _ ratio > -R;
and 9-2, judging walking, wherein the threshold conditions comprise a walking speed threshold range walk _ speed and a walking distance threshold range walk _ distance. According to step 7, walk _ speed takes the value of S3. According to step 8, the walk _ distance value is D2;
and 9-3, judging the bicycle, wherein the threshold conditions comprise a bicycle speed threshold range bike _ speed and a walking distance threshold range bike _ distance. According to step 7, the value of bike _ speed is S4. According to step 8, the value of bike _ distance is D3;
step 9-4, distinguishing the car;
and 9-4-1, judging the car according to the judgment result in the step 9-1, wherein the threshold conditions comprise a car speed threshold range car1_ speed, a car distance threshold range car1_ distance and sd. According to step 7, when sd is 0, car1_ speed takes a value of S5, and when sd is 1, car1_ speed takes a value of S6. According to step 8, the bus _ distance value is D4. busline _ ratio < R;
and 9-4-2, judging the car, wherein the threshold conditions comprise a car speed threshold range car2_ speed and a car distance threshold range car2_ distance. According to step 7, when sd is 0, car1_ speed takes a value of S6, and when sd is 1, car1_ speed takes a value of S7. According to the step 8, the bus _ distance value is D5;
and 9-5, judging the motorcycle, wherein the threshold conditions comprise a motorcycle speed threshold range motor _ speed and a motorcycle distance threshold range motor _ distance. According to step 7, the motor _ speed value is S8. According to step 8, the value of motor _ distance is D6.
And 10, adding the data successfully judged in the step 9 into a database.
And 11, performing secondary judgment on the problem data out of the threshold condition, and correcting the result of the step 10.
And step 12, obtaining the travel OD distance OD _ dis of the problem data according to the OD point longitude and latitude of the user single travel track in the step 2.
And step 13, carrying out secondary judgment on the problem data, wherein the judgment condition threshold values are the user travel distance, the speed range, the OD distance and the bus coverage rate.
In the embodiment of the present invention, step 13 specifically includes:
step 13-1, because the data with the too large distance is cleaned in the step 2, the problem that the data is not judged is mainly focused on that the avg _ speed caused by the time field is too small or too large, and the avg _ speed caused by the fact that the user does not exceed the range of the signal base station within 30min of the activity range is too small;
and step 13-2, performing secondary judgment on walking, wherein the threshold conditions comprise a walking distance threshold range walk _ distance and a walking OD distance threshold range walk _ distance. The walkmod _ distance value is 0, and the walkjdistance value is D7;
and step 13-3, performing secondary judgment on the bicycle, wherein the threshold conditions comprise a bicycle distance threshold range bike _ distance and a bicycle OD distance threshold range bikeod _ distance. The bikeod _ distance value is Dod1, and the bike _ distance value is D8; and step 13-4, carrying out secondary judgment on the bus, wherein the threshold conditions comprise a bus speed threshold range bus _ speed, a bus distance threshold range bus _ distance, a bus OD distance threshold range busod _ distance and a busline _ ratio. According to step 13-1, bus _ speed takes the value Sw. According to step 8, the value of the busod _ distance is Dod2, and the value of the bus _ distance is D9. The value of the busline _ ratio is suggested to be 0.8;
step 13-5, carrying out secondary judgment on the car;
and step 13-5-1, judging the car according to the judgment result in the step 9-1, wherein the threshold conditions comprise a car speed threshold range car1_ speed, a car distance threshold range car1_ distance, a car OD distance threshold range car1OD _ distance and a busline _ ratio. And (4) suggesting values: according to step 13-1, car1_ speed takes the value Sw. According to step 7, car1od _ distance takes the value Dod3, and car1_ distance takes the value D10. The busline _ ratio suggested value is busline _ ratio < 0.8;
and step 13-5-2, judging the car, wherein the threshold conditions comprise car OD distance threshold range car2OD _ distance and car distance threshold range car2_ distance. According to step 8, the value of car2od _ distance is Dod4, and the value of car2_ distance is D11;
and step 14, adding the data successfully judged in the step 13 into the database in the step 10, and correcting the data.
Referring to fig. 1 to 4, in an embodiment, a travel mode identification method based on mobile phone signaling data includes the following steps:
step 1, grouping original data of mobile phone signaling of users in the Unicom of the Seisan city of the extremely intelligent platform according to a monthly travel number moi _ id and a daily travel number move _ id of the users, sequencing the data according to time, connecting the data end to form a single travel track, spatially connecting a poi number start _ poi corresponding to a starting point with a track starting point, spatially connecting a poi number end _ poi corresponding to an end point with a track end point, and finally cleaning the data. And calculating the obtained track distance data dis and the total time of the single trip to obtain the average speed avg _ speed of the single trip, as shown in table 1.
TABLE 1 average speed calculation results of one-time complete travel trajectory of user
And 2, uploading an shp file of a public transportation line network in Xian city to an extremely intelligent platform, and calculating the coincidence rate busline _ ratio of the single trip track of the user and the public transportation line network. According to the travel starting time and the travel ending time, the travel time period sd of the user is judged, as shown in table 2.
Table 2 time interval judgment of one-time complete travel track of user, and bus route contact ratio result
And 3, obtaining the route planning condition of the real-time road condition information of the city of the Xian by using the FME and Baidu open platform route planning, counting the average speed in a peer-to-peer time circle, and calculating the speed threshold range as shown in the table 3. The distance threshold range is obtained by using the K-means clustering result and the survey report reference of the west ampere city trip, and is calculated according to the following formula as shown in table 4:
where k is the given number of clusters, i represents each given sample, n represents each sample element is an n-dimensional vector, and the centroid μ j Representing our guess of the center point of the samples belonging to the same class, r ij Represents the data point x (i) Is classified into mu j Is 1 when it is used, otherwise is 0. In the present invention, the proposed value of k is 5.
When sd is 0, the value of bus _ speed is 3.6-8 m/s, and when sd is 1, the value of bus _ speed is 2.6-6.5 m/sbus _ distance is 2000-15000 m. The value of the busline _ ratio is suggested to be 0.8; the walk _ speed value is 0.05-1.2 m/s. The value of walk _ distance is 0-3000 m; the value of bike _ speed is 1.2-3.6 m/s. The value of bike _ distance is 500-4000 m; when sd is 0, car1_ speed takes a value of 3.6 to 8m/s, and when sd is 1, car1_ speed takes a value of 2.6 to 6.5 m/s. The bus _ distance value is 2000-50000 m. The busline _ ratio suggested value is busline _ ratio < 0.8; when sd is 0, car1_ speed takes a value of 8-15 m/s, and when sd is 1, car1_ speed takes a value of 6.5-12 m/s. The bus _ distance value is 2000-50000 m; the motor _ speed value is 2.6-6 m/s. The value of motor _ distance is 500-10000 m.
The data is first discriminated according to the threshold condition, as shown in fig. 1.
TABLE 3 average speed within isochronous cycles
TABLE 4K-means Classification based on data samples
And fourthly, performing secondary judgment on the unsuccessfully judged data sample, calculating the OD distance, and judging according to a threshold condition, wherein the walkod _ distance value is 0, and the walk _ distance value is 0-2000 m. The value of bikeod _ distance is 0-2000 m, and the value of bike _ distance is 2000-4000 m. The bus _ speed value is 0-2 m/s or more than or equal to 25 m/s. The value of the bus _ distance is 2000-12000 m, and the value of the bus _ distance is 4000-15000 m. The value of busline _ ratio is suggested to be 0.8. The value of car1_ speed is 0-2 m/s or more than or equal to 25 m/s. The value of car1od _ distance is 2000-12000 m, and the value of car1_ distance is 4000-15000 m. The busline _ ratio suggested value is busline _ ratio < 0.8. The value of car2od _ distance is 8000-50000 m, and the value of car2_ distance is 15000-50000 m, as shown in FIG. 2 and Table 5.
Table 5. secondary discrimination result of primary complete travel locus of unsuccessfully discriminated user
Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art can make modifications and equivalents to the embodiments of the present invention without departing from the spirit and scope of the present invention, and such modifications and equivalents are within the scope of the claims of the present invention as hereinafter claimed.
Claims (7)
1. A travel mode identification method based on mobile phone signaling data is characterized by specifically comprising the following steps:
step 1, acquiring mobile phone signaling data; based on the longitude and latitude and time slice information of the user in the mobile phone signaling data, POI point elements of a starting point and a terminal point of single trip of the user are obtained, and a middle section track of the single trip of the user is obtained; acquiring the position information of a city where a user is located;
step 2, obtaining a fitted travel track based on POI point elements of a starting point and a finishing point of a user and a middle section track; based on the fitted travel track, obtaining the single travel distance and travel time of the user, and calculating to obtain the average speed of the single travel of the user;
step 3, acquiring a speed threshold range and a distance threshold range of each travel mode in a peak time period and a peak-smoothing time period of a city where a user is located; generating a buffer area based on the urban bus operation line network, wherein the buffer area is used for acquiring track points of the fitted travel track on the urban bus operation line; acquiring the coincidence rate of the travel track and the urban bus operation line network based on the fitted travel track and the buffer area; judging whether the travel time of the user is in a peak time period or a flat time period based on the single travel distance and travel time of the user and the position information of the city where the user is located;
step 4, judging the user trip mode for one time based on a preset threshold condition, finishing trip mode identification if a judgment result is obtained for one time, and otherwise, skipping to execute step 5; the threshold conditions of the first judgment are the travel distance, the average speed, the travel time period and the coincidence rate of the travel track and the urban public transport operation line network of the user;
and 5, carrying out secondary judgment on the problem data outside the threshold condition of the primary judgment, wherein the secondary judgment comprises the following steps: obtaining a travel OD distance of the problem data based on the OD point longitude and latitude of the fitted travel track obtained in the step 2; performing secondary discrimination on the problem data to obtain a secondary discrimination result, and finishing travel mode identification; the threshold conditions of the secondary judgment are user travel distance, average speed, OD distance and coincidence rate of travel track and urban public transport operation line network;
in step 4, the specific steps of primary discrimination include:
the bus trip mode is distinguished and includes: the threshold conditions comprise a bus speed threshold range, a bus distance threshold range, a trip time period, a track and urban bus operation line network coincidence rate range; wherein, the bus speed threshold value includes: the bus speed threshold value range is obtained when the trip time period is a peak-off time period, and the bus speed threshold value range is obtained when the trip time period is a peak time period; when the bus trip is within the threshold value ranges of the bus, the bus trip is a bus trip mode;
the judging of the walking travel mode comprises the following steps: the threshold condition comprises a walking speed threshold range and a walking distance threshold range; if the current trip falls within the range of each threshold value of walking, the trip is a walking trip mode;
the bicycle trip mode is distinguished and includes: the threshold condition comprises a bicycle speed threshold range and a bicycle distance threshold range; the bicycle trip mode is determined when the bicycle falls within the range of each threshold value of the bicycle;
according to the judgment result of the bus trip mode, private cars in a bus trip threshold interval are judged, and the threshold conditions comprise a private car speed threshold range, a private car distance threshold range, a trip time period range and a coincidence rate range of a track and a city bus operation line network; wherein the private car speed threshold range comprises: a private car speed threshold range when the travel time period is a peak-off time period and a private car speed threshold range when the travel time period is a peak-off time period; the mode of the private car going out is determined when the vehicle falls within the threshold value range of the private car;
judging private cars outside the bus trip threshold interval, wherein the threshold conditions comprise a private car speed threshold range and a private car distance threshold range; wherein the private car speed threshold range comprises: a private car speed threshold range when the travel time period is a peak-off time period and a private car speed threshold range when the travel time period is a peak-off time period; the private car trip mode is determined when the vehicle falls within each threshold range of the private car;
the method for judging the travel mode of the motorcycle comprises the following steps: the threshold condition comprises a motorcycle speed threshold range and a motorcycle distance threshold range; and if the signal value falls within the range of each threshold value of the motorcycle, the mode of motorcycle travel is adopted.
2. A travel mode identification method based on mobile phone signaling data according to claim 1, wherein in step 1, the mobile phone signaling data comes from unicom mobile phone signaling data of smart footprint smart platform, and the method comprises: the longitude and latitude of the starting and ending point of the single trip, the longitude and latitude, the sequence and the number of each path node in the single trip track, the trip starting time and the trip ending time.
3. A travel mode identification method based on mobile phone signaling data according to claim 2, characterized in that in step 2, in the mobile phone signaling data, the subway travel mode has an individual travel field; and screening out single subway trip samples of the user to obtain a database of trip modes to be identified.
4. A travel mode identification method based on mobile phone signaling data according to claim 1, wherein in step 2, the specific step of obtaining the fitted travel trajectory based on the POI point elements of the start point and the end point of the user and the middle section trajectory includes:
fitting and connecting the first section track and the tail section track of the user with the middle section track to obtain a fitted travel track;
the middle section track is a result of fitting the urban road network by the position of the base station when a user uses a mobile phone and signals interact with the base station; the first-segment track is a result of the connection of a POI point element of a starting point of the user and a POI point element of a starting point of the middle-segment track, and the last-segment track is a result of the connection of a POI point element of a finishing point of the user and a POI point element of an ending point of the middle-segment track; and cleaning the data with the sum of the distances between the first section of track and the tail section of track being more than or equal to 1 km.
5. A travel mode identification method based on mobile phone signaling data according to claim 1, characterized in that in step 3, according to Baidu open platform lightweight route planning data, a speed threshold range of each travel mode in urban peak time and peak-off time is obtained; and for the travel distance, classifying by using K-means unsupervised learning, and obtaining the distance threshold range of the travel of each traffic mode in the city by taking the classification result as reference.
6. A travel mode identification method based on mobile phone signaling data according to claim 1, characterized in that in step 3, the specific step of obtaining the coincidence rate of the travel trajectory and the urban public transportation operation line network based on the fitted travel trajectory and the buffer zone comprises:
(1) selecting track point data of the single trip track in the buffer area by using the formed buffer area to form coincident track data of the single trip track and the urban bus operation line network;
(2) and comparing the obtained coincidence track data with the single trip track data to obtain the coincidence rate of the single trip track and the urban public transport operation line network.
7. A travel mode identification method based on mobile phone signaling data according to claim 1, wherein in step 5, the specific step of secondary discrimination includes:
performing secondary judgment on walking, wherein the threshold conditions comprise a walking distance threshold range and a walking OD distance threshold range; if the current trip falls within the range of each secondary judgment threshold value of walking, the trip is a walking trip mode;
performing secondary judgment on the bicycle, wherein the threshold conditions comprise a bicycle distance threshold range and a bicycle OD distance threshold range; if the bicycle falls within the range of each secondary discrimination threshold value of the bicycle, the bicycle is taken as a bicycle travel mode;
carrying out secondary judgment on the bus, wherein the threshold conditions comprise a bus speed threshold range, a bus distance threshold range, a bus OD distance threshold range and a coincidence rate range of a track and a city bus operation line network; if the bus trip is within the range of each secondary judging threshold value of the bus, the bus trip is a bus trip mode;
the secondary judgment of the private car comprises the following steps: according to the secondary judgment result of the bus, judging private cars in a bus trip threshold range, wherein the threshold conditions comprise a private car speed threshold range, a private car distance threshold range, a private car OD distance threshold range and a coincidence rate of a track and a city bus operation line network; judging private cars outside the bus trip threshold interval, wherein the threshold conditions comprise a private car OD distance threshold range and a private car distance threshold range; and if the traveling mode falls within the range of each secondary judgment threshold value of the private car, the traveling mode is the private car traveling mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011366779.0A CN112511971B (en) | 2020-11-26 | 2020-11-26 | Travel mode identification method based on mobile phone signaling data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011366779.0A CN112511971B (en) | 2020-11-26 | 2020-11-26 | Travel mode identification method based on mobile phone signaling data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112511971A CN112511971A (en) | 2021-03-16 |
| CN112511971B true CN112511971B (en) | 2022-08-16 |
Family
ID=74968007
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011366779.0A Active CN112511971B (en) | 2020-11-26 | 2020-11-26 | Travel mode identification method based on mobile phone signaling data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112511971B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114396962A (en) * | 2022-02-25 | 2022-04-26 | 北京世纪高通科技有限公司 | Reachable domain generation method, device, equipment and storage medium |
| CN116681507B (en) * | 2023-05-18 | 2024-07-09 | 北京大也智慧数据科技服务有限公司 | Payment index calculation method, device, storage medium and equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105117789A (en) * | 2015-07-29 | 2015-12-02 | 西南交通大学 | Resident trip mode comprehensive judging method based on handset signaling data |
| CN105608505A (en) * | 2015-12-22 | 2016-05-25 | 重庆邮电大学 | Cellular signaling data based track traffic travel mode identification method for resident |
| CN111653094A (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 and containing road network correction |
| 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 |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101562247B1 (en) * | 2013-06-25 | 2015-10-23 | 아주대학교산학협력단 | System and method for lifestyle analysis |
| CN108320501B (en) * | 2017-12-21 | 2021-01-12 | 江苏欣网视讯软件技术有限公司 | Bus route identification method based on user mobile phone signaling |
| CN110505583B (en) * | 2019-07-23 | 2021-01-22 | 中山大学 | Trajectory matching method based on bayonet data and signaling data |
-
2020
- 2020-11-26 CN CN202011366779.0A patent/CN112511971B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105117789A (en) * | 2015-07-29 | 2015-12-02 | 西南交通大学 | Resident trip mode comprehensive judging method based on handset signaling data |
| CN105608505A (en) * | 2015-12-22 | 2016-05-25 | 重庆邮电大学 | Cellular signaling data based track traffic travel mode identification method for resident |
| CN111653094A (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 and containing road network correction |
| 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 |
Non-Patent Citations (3)
| Title |
|---|
| 基于GPS轨迹数据的城市交叉路口识别;高原等;《数据分析与知识发现》(第11期);全文 * |
| 基于手机信令数据的居民出行特征研究;周剑明;《信息与电脑(理论版)》;20191210(第23期);全文 * |
| 基于手机信令的城市群通道出行特征提取方法研究;章玉;《综合运输》;20171120(第11期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112511971A (en) | 2021-03-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111653097B (en) | Urban trip mode comprehensive identification method based on mobile phone signaling data and containing personal attribute correction | |
| CN111681421B (en) | Mobile phone signaling data-based external passenger transport hub centralized-sparse space distribution analysis method | |
| CN108362293B (en) | Vehicle track matching method based on key point technology | |
| CN112511971B (en) | Travel mode identification method based on mobile phone signaling data | |
| CN108848460B (en) | Man-vehicle association method based on RFID and GPS data | |
| CN112150796B (en) | Customized bus stop layout method based on mobile phone signaling data | |
| CN109035787B (en) | Method for identifying vehicle category by using mobile communication data | |
| CN107919014B (en) | Taxi running route optimization method for multiple passenger mileage | |
| CN106197458A (en) | A kind of cellphone subscriber's trip mode recognition methods based on mobile phone signaling data and navigation route data | |
| CN106297280A (en) | A kind of information processing method and device | |
| CN109902899A (en) | Information generating method and device | |
| CN111653093B (en) | Urban trip mode comprehensive identification method based on mobile phone signaling data | |
| CN101964941A (en) | Intelligent navigation and position service system and method based on dynamic information | |
| CN109191883A (en) | A kind of intelligent transportation system based on intelligent perception | |
| CN110310477B (en) | Bus passenger flow detection method based on bus GPS and mobile phone signaling data | |
| CN110472999B (en) | Passenger flow mode analysis method and device based on subway and shared bicycle data | |
| CN109729518B (en) | Mobile phone signaling-based urban traffic early peak congestion source identification method | |
| CN110634290B (en) | Bicycle track data-based retrograde behavior identification method | |
| CN110414795B (en) | Newly-increased high-speed rail junction accessibility influence method based on improved two-step mobile search method | |
| CN105792134A (en) | Carpooling method and system | |
| CN107545318B (en) | Bus line priority determination and bus transfer line sequencing method and device | |
| CN111341135A (en) | Mobile phone signaling data travel mode identification method based on interest points and navigation data | |
| CN113079463A (en) | Tourist attraction tourist travel activity identification method based on mobile phone signaling data | |
| CN109489679A (en) | A kind of arrival time calculation method in guidance path | |
| CN113573238A (en) | Method for identifying trip passenger trip chain based on mobile phone signaling |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |