CN114611622A - Method for identifying cross-city commuting crowd by utilizing mobile phone data - Google Patents

Method for identifying cross-city commuting crowd by utilizing mobile phone data Download PDF

Info

Publication number
CN114611622A
CN114611622A CN202210267171.5A CN202210267171A CN114611622A CN 114611622 A CN114611622 A CN 114611622A CN 202210267171 A CN202210267171 A CN 202210267171A CN 114611622 A CN114611622 A CN 114611622A
Authority
CN
China
Prior art keywords
user
city
travel
trips
ratio
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
Application number
CN202210267171.5A
Other languages
Chinese (zh)
Inventor
王春松
赖见辉
李金龙
陈艳艳
陈庚
龙莹洁
张建利
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing University of Technology
Original Assignee
Beijing University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing University of Technology filed Critical Beijing University of Technology
Priority to CN202210267171.5A priority Critical patent/CN114611622A/en
Publication of CN114611622A publication Critical patent/CN114611622A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/243—Classification techniques relating to the number of classes
    • G06F18/24323—Tree-organised classifiers
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • 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
    • H04W4/021—Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences

Abstract

The invention discloses a method for identifying cross-city commuting crowds by utilizing mobile phone data, which is divided into three parts, and the process is as follows: firstly, acquiring mobile phone data of a known user type, marking a road type label on each data set, preprocessing the mobile phone data, extracting a user travel OD, screening out users with cross-city travel behaviors, and acquiring relevant user travel indexes in the data sets as data model samples; and secondly, constructing a decision tree model. Training the decision tree by using k-fold cross validation, and determining a final crowd category identification model by comparing model index scores of each time; thirdly, mobile phone data of unknown user types are obtained, the mobile phone data are preprocessed, user travel related indexes are extracted, the travel related indexes are input into the identification model obtained in the last step, and corresponding user type labels are output, namely the user types are identified.

Description

Method for identifying cross-city commuting crowd by utilizing mobile phone data
Technical Field
The invention relates to the technical field of big data, in particular to a method for identifying a cross-city commuting population of a mobile phone user.
Background
Along with the enlargement of the urban scale and the improvement of the traffic conditions between cities, particularly the appearance of highways and railways, adjacent urban radiation areas are close to each other and partially overlapped, the cities are closely related to each other, the mutual influence is large, and a large number of city-crossing trips are induced. Basic activities such as living, working and recreation originally existing in the same city are diffused to adjacent cities of a city circle, and a mode that basic city functions such as living, working and recreation are dispersed among different cities is generated, namely cross-city commuting. Cross-city commuting has increased significantly in recent years within different intercity areas across the country as the primary mode of travel connecting habitation and workplace. However, conventional data resources and techniques have difficulty accurately locating cross-city commuter populations, identifying and tracking commuter travel over long periods of time. With the arrival of the big data era, data resources are increasingly abundant, the mobile phone signaling data has the characteristics of large sample amount and sustainable tracking, can be applied to quantitative analysis of traffic characteristics, and extracts possibility for accurately identifying cross-city commuting and analyzing cross-city commuting characteristics.
Disclosure of Invention
The invention provides a method for identifying city-crossing commuting crowds by utilizing mobile phone data, which lays a foundation for research of the mobile phone data in the city-crossing commuting and the development and construction of city groups, and the method is divided into three parts, and the process is as follows: firstly, acquiring mobile phone data of a known user type, marking a road type label on each data set, preprocessing the mobile phone data, extracting a user travel OD, screening out users with cross-city travel behaviors, and acquiring relevant user travel indexes in the data sets as data model samples; and secondly, constructing a decision tree model. Training the decision tree by using k-fold cross validation, and determining a final crowd category identification model by comparing model index scores of each time; thirdly, mobile phone data of unknown user types are obtained, the mobile phone data are preprocessed, user travel related indexes are extracted, the travel related indexes are input into the identification model obtained in the last step, and corresponding user type labels are output, namely the user types are identified.
The technical scheme adopted by the invention is a method for identifying the cross-city commute by utilizing mobile phone data, which specifically comprises the following steps:
step 1: and acquiring mobile phone signaling data of a known user type, wherein the mobile phone signaling data comprises a user number, a base station longitude, a base station latitude and a signaling record generation time.
Step 2: preprocessing the acquired mobile phone signaling data, identifying a user stop point, further extracting a user travel OD, and judging whether the travel OD is in the same city or not; if the cities are the same city, the city is on a trip, and if the cities are not the same city, the city is on a trip.
And step 3: and (3) according to the user travel OD extracted in the step (2), deleting all users without the cross-city travel behavior in the travel, namely deleting all users with travel ODs in the same city, and screening out the users with the cross-city behavior.
And 4, step 4: constructing a set of user categories for cross-city behavior, C ═ C1,c2,c3In which c is1Representing commuting groups across the city, c2The city-across-the-day commuter group, c1And c2Can be collectively referred to as city-crossing commuter group, c3Representing a non-cross-city commuter population. Defining class labels of all training set sample data, and determining peak travel time periods and peak travel dates of the cross-city commuter crowd.
And 5: according to the characteristics of commuting travel, obtaining travel indexes of each mobile phone user, wherein the travel indexes comprise the weekly average travel times N and the travel proportion P in peak periodsgCity-crossing trip ratio P of peak datedAnd the ratio of trip in off-peak period of working day is PfWorking sunrise-to-go ratio PwAnd the city-crossing travel ratio Pc. The specific calculation method of each index is as follows:
1) average weekly trip times N:refers to the average of the number of user trips per week.
Figure BDA0003552284090000031
Wherein QjThe total number of j week trips of the user is shown, and n is n weeks included in the study period.
2) Ratio of travel P in peak periodg: the ratio of the number of trips of the user in the peak period to the total number of trips is shown.
Figure BDA0003552284090000032
Wherein: qgRepresenting the number of trips of the user in the peak period, and Q representing the total number of trips of the user.
3) City-crossing trip ratio P of peak dated: which is the ratio of the number of trips of the user on the peak date to the total number of trips across the city.
Figure BDA0003552284090000033
Wherein: qdRepresenting the number of times a user travels across the city on a peak date, QcRepresenting the total number of times the user travels across the city.
4) Duty ratio P for off-peak hours of working dayf: refers to the ratio of the number of trips of the user during off-peak hours during the working day to the total number of trips.
Figure BDA0003552284090000034
Wherein: qfRepresenting the number of trips during off-peak hours during the user's workday.
5) Working sunrise ratio Pw: refers to the ratio of the number of trips of the user during the working day to the total number of trips.
Figure BDA0003552284090000035
Wherein: qwRepresenting the number of trips during the user's early work day.
6) Cross city trip ratio Pc: the ratio of the number of times of the user going out across the city to the total number of times of the user going out is shown.
Figure BDA0003552284090000036
Step 6: dividing the training set into k parts, wherein one part is used as a verification set, the other k-1 parts are used as the training set, inputting the selected verification set into a trained decision tree model, and scoring and storing the indexes of the verification model.
And 7: and repeating the step 6k times, wherein verification sets selected each time are different to obtain k model index scores, and taking the average value of the k model index scores to obtain the final score of the decision tree model.
And 8: and (3) carrying out data processing on the mobile phone data of the unknown user type according to the step (1) and the step (2), and calculating a user travel index. And (4) inputting the travel indexes into the decision tree recognition models obtained in the steps 6 and 7, and outputting corresponding user type labels.
Compared with the prior art, the method can identify the cross-city commute and analyze the cross-city commute characteristics by applying the quantitative analysis of the traffic characteristics; the mobile phone data is used for identifying the cross-city commuting crowd, so that accurate big data analysis is performed on the user types; meanwhile, a foundation is laid for the research of the mobile phone data in the cross-city commute and the development and construction of city groups.
Drawings
Fig. 1 is a flowchart of a method for identifying a cross-city commuter group using mobile phone data according to an embodiment of the present invention;
FIG. 2 is a flow chart of decision tree construction according to an embodiment of the present invention;
fig. 3 is a tree diagram of the classification result of the final decision tree algorithm according to the embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, the overall process of the method is divided into three parts, namely, mobile phone data of a known user type is obtained, each data set is marked with a road type label, the mobile phone data is preprocessed, a user travel OD is extracted, users with cross-city travel behaviors are screened out, and relevant user travel indexes in the data sets are obtained and used as data model samples; and secondly, constructing a decision tree model. Training the decision tree by using k-fold cross validation, and determining a final crowd category identification model by comparing model index scores of each time; thirdly, mobile phone data of unknown user types are obtained, the mobile phone data are preprocessed, user travel related indexes are extracted, the travel related indexes are input into the identification model obtained in the step 2, and corresponding user type labels are output, namely the user types are identified.
The following description takes a set of mobile phone data as an example:
step 1: and acquiring mobile phone signaling data of a known user type, wherein the mobile phone signaling data comprises a user number, a base station longitude, a base station latitude and a signaling record generation time. For example: the ith mobile phone data of the nth user is recorded as
Figure BDA0003552284090000051
Wherein UserIDnA unique identification code representing the nth user,
Figure BDA0003552284090000052
denotes the abscissa, Y, of the position of the ith entry of the n usersi nRepresents the vertical coordinate, T, of the position of the ith record of n usersi nIndicates the time points at which the ith entries of the n users occur, as shown in table 1.
Table 1: mobile phone signaling data
Figure BDA0003552284090000053
Figure BDA0003552284090000061
Step 2: by using"a method for judging user moving and staying state by using mobile phone positioning data" (patent number: CN106203505B), determining user staying point, obtaining user staying point information table, recording the ith staying point data of nth user as
Figure BDA0003552284090000062
Wherein UserIDnA unique identification code representing the nth user,
Figure BDA0003552284090000063
representing the abscissa, Y, of the position of the i-th stopping point of the n usersi nDenotes the position ordinate, ET, of the i-th stop point of the n usersi nIndicating the point in time, LT, at which the n users reach the ith stopi nIndicating the point in time when the n users left the ith dwell point as shown in table 2.
Further, sorting the user stop points according to time, wherein two adjacent stop points are a trip of the user, the former point is a starting point, the latter point is an end point, a trip OD information table of the user is obtained, and the ith trip OD data of the nth user is recorded as
Figure BDA0003552284090000064
Wherein UserIDnA unique identification code representing the nth user,
Figure BDA0003552284090000065
represents the starting position coordinates of the ith piece of OD data of n users,
Figure BDA0003552284090000066
coordinates of end points, OT, representing the i-th item OD data of n usersi nIndicating the time point, DT, at which the ith piece of OD data of n users starts from the starting pointi nThe time point when the ith OD data of n users reaches the end point is shown in table 3. And finally, determining whether the travel is the city-crossing travel or not according to whether the travel OD is in the same city or not.
Table 2: user dwell point data
Figure BDA0003552284090000071
Table 3: user travel OD data
Figure BDA0003552284090000072
And step 3: and deleting users who do not have the cross-city behavior in all the trips, namely deleting users who have all trips in the same city according to the user trips OD extracted in the step. And screening out users with the cross-city behavior.
And 4, step 4: constructing a crowd category set, C ═ C1,c2,c3In which c is1Representing commuting groups across the city, c2The city-across-the-day commuter group, c1And c2Can be collectively referred to as city-crossing commuter group, c3Representing a non-cross-city commuter population. Defining class labels of all training set sample data, and determining peak travel time periods of the cross-city commuter population and peak travel dates of the cross-city weekly commuter population, wherein through statistical analysis, the early peak time period of the cross-city commuter population is 6:00-8:00, the late peak travel time period is 5:00-7:00, the off-duty peak travel date of the cross-city weekly commuter population is friday, and the on-duty peak travel date is sunday and monday.
Table 4: user travel OD data
Figure BDA0003552284090000073
And 5: according to the characteristics of commuting travel, the travel indexes of each mobile phone user are obtained, wherein the travel indexes comprise the weekly average travel times N and the travel proportion P in peak periodsgCity-crossing trip ratio P of peak datedAnd the ratio of trip in off-peak period of working day is PfWorking sunrise-to-go ratio PwCity-crossing trip ratio Pc. The specific calculation method of each index is as follows:
1) average number of trips in weekN: refers to the average of the number of user trips per week.
Figure BDA0003552284090000081
Wherein QjThe total number of j week trips of the user is shown, and n is n weeks included in the study period. For example, the number of trips of a user in four weeks is 14,17,12 and 16 respectively, and the average number of trips in the week
Figure BDA0003552284090000082
2) Ratio of travel P in peak periodg: the ratio of the number of trips of the user in the peak period to the total number of trips is shown.
Figure BDA0003552284090000083
Wherein: qgRepresenting the number of trips of the user in the peak period, and Q representing the total number of trips of the user. For example, if a user has 18 trips in a week, and 12 trips in a peak period, the trip percentage in the peak period is
Figure BDA0003552284090000084
3) City-crossing trip ratio P of peak dated: the ratio of the number of trips of the user on the peak date to the total number of trips is shown.
Figure BDA0003552284090000085
Wherein: qdRepresenting the number of trips, Q, of the user crossing the city at peak datecRepresenting the total number of times the user travels across the city. For example, if the number of city-crossing trips of a user on the peak trip date (monday, friday, sunday) is 2, and the total number of city-crossing trips in one week is 4, the working day trip percentage is
Figure BDA0003552284090000086
4) Duty ratio P for off-peak hours of working dayf: refers to the ratio of the number of trips of the user during off-peak hours during the working day to the total number of trips.
Figure BDA0003552284090000087
Wherein: qfRepresenting the number of trips during the user's weekday during off-peak hours. For example, if a user has 18 total trips in a week and 4 trips in the off-peak hours during the weekdays from monday to friday, the duty ratio of trips in the off-peak hours of the weekdays is higher
Figure BDA0003552284090000091
5) Working day-to-day ratio Pw: refers to the ratio of the number of trips of the user during the working day to the total number of trips.
Figure BDA0003552284090000092
Wherein: qwRepresenting the number of trips during the user's early work day. For example, if a user has 18 trips in a week and 16 trips in the period from monday to friday, the duty ratio of the trips in the working day is higher
Figure BDA0003552284090000093
6) Cross city trip ratio Pc: the ratio of the number of times of the user going out across the city to the total number of times of the user going out is shown.
Figure BDA0003552284090000094
For example, if the total number of trips of a user in a week is 18, and the number of trips across cities is 10, the working sunrise duty ratio
Figure BDA0003552284090000095
Step 6: suppose that after data processing and screening, 1000 mobile phone data of users are used as training samples. And dividing the training set into 10 parts, wherein one part is used as a verification set, the other 9 parts are used as the training set, inputting the selected verification set into the trained decision tree model, and scoring and storing the indexes of the verification model.
The decision tree algorithm flow is as follows:
the key of the decision tree construction lies in the utilization of information gain confirmationAnd determining the purity of the classification result, setting the training data set as D, and expressing the sample capacity by | D |. Is provided with K classes Ck，k＝1,2,3,...,K，|CkIs of class CkThe number of samples. Let the characteristic value A have n different values { a }1,a2,...,anDividing D into n subsets D according to the value of the characteristic value A1,D2,...Dn，|DiL is DiThe number of samples. Memory set DiIn the class CkIs DikI.e. Dik＝Di∩Ck，|DikL is DikThe number of samples. The information entropy of the data set D is
Figure BDA0003552284090000096
Conditional entropy of eigenvalues A on data sets D
Figure BDA0003552284090000097
The information gain g (D, a) ═ H (D) -H (D | a) for feature a for dataset D.
Constructing a decision tree in the case, and calculating the information entropy of the sample set D in the first step
Figure BDA0003552284090000101
Second step calculates the current attribute set { N, P }g,Pd,Pf,Pw,Pc,FsComparing the information gain of each attribute in the data, dividing by using the attribute with the maximum information gain to obtain divided data subsets; and thirdly, continuously dividing each data subset according to the second step to obtain a final decision tree.
In addition, in order to verify the superiority and inferiority of the classification result, the kini Index (Gini Index) is selected for measurement, and the calculation formula is as follows:
Figure BDA0003552284090000102
wherein p iskRepresenting the probability that the selected sample belongs to the k class, the probability that this sample is misclassified is (1-p)k) With n classes in the sample setA randomly selected sample may belong to any of the n classes, and the probabilities for the classes are thus summed.
The idea of processing continuous values is to discretize the continuous features. Given a sample set D and a continuous attribute a, assuming that a has n different values on D, these values are ordered from small to large, denoted as { a }1,a2,...,an}. D can be divided into subsets based on the division point t
Figure BDA0003552284090000103
And
Figure BDA0003552284090000104
wherein
Figure BDA0003552284090000105
Including those samples that take on a value not greater than t on attribute a, and
Figure BDA0003552284090000106
then samples that take values greater than t on attribute a are included. Obviously, the values a are given to adjacent attributesiAnd ai+1In other words, t is in the interval [ a ]i,ai+1) The division results generated by taking any value are the same. Thus, for the continuous attribute a, consider a set of candidate partitions containing n-1 elements:
Figure BDA0003552284090000107
i.e. the handle interval [ ai,ai+1) Middle point of
Figure BDA0003552284090000108
As candidate division points. Then, the division points can be considered like discrete attribute values, and the optimal division point is selected for dividing the sample set.
And 7: repeating the step 6 for 10 times, wherein verification sets selected each time are different, and obtaining 10 model index scores in total [0.8352, 0.7932, 0.8716, 0.8925, 0.9623, 0.8159, 0.9399, 0.9015, 0.8367 and 0.8566], and obtaining the final score of the decision tree model by averaging: 0.87054. finally, a decision tree model is obtained, as shown in fig. 3.
And 8: and (3) carrying out data processing on the mobile phone data of the unknown user type according to the step (1) and the step (2), and calculating a user travel index. And (4) inputting the travel indexes into the decision tree recognition models obtained in the steps 6 and 7, and outputting corresponding user type labels. For example, the feature values of three users are input into the decision tree in table 4, and the corresponding user types can be obtained.
Figure BDA0003552284090000111

Claims (2)

1. A method for identifying cross-city commute by utilizing mobile phone data is characterized by comprising the following steps: the method comprises the following steps:
step 1: acquiring mobile phone signaling data of a known user type, wherein the mobile phone signaling data comprises a user number, a base station longitude, a base station latitude and signaling record generation time;
step 2: preprocessing the acquired mobile phone signaling data, identifying a user stop point, further extracting a user travel OD, and judging whether the travel OD is in the same city or not; if the cities are the same city, going out in the same city, and if the cities are not the same city, going out across cities;
and step 3: according to the user travel OD extracted in the step 2, deleting users without the cross-city travel behavior in all travels, namely deleting users with all travel ODs in the same city, and screening out users with the cross-city behavior;
and 4, step 4: constructing a set of user categories for cross-city behavior, C ═ C1,c2,c3In which c is1Representing commuting groups across the city, c2The city-across-the-day commuter group, c1And c2Can be collectively referred to as city-crossing commuter group, c3Representing non-cross-city commuter populations; defining class labels of all training set sample data, and determining peak travel time periods and peak travel dates of the cross-city commuter crowd;
and 5: according to characteristics of commuting tripAnd (3) acquiring travel indexes of each mobile phone user, wherein the travel indexes comprise the weekly average travel times N and the travel ratio P in peak periodgCity-crossing trip ratio P of peak datedAnd the ratio of trip in off-peak period of working day is PfWorking sunrise-to-go ratio PwCity-crossing trip ratio Pc；
And 6: dividing the training set into k parts, wherein one part is used as a verification set, the other k-1 parts are used as the training set, inputting the selected verification set into a trained decision tree model, and scoring and storing the indexes of the verification model;
and 7: repeating the step 6k times, wherein verification sets selected each time are different to obtain k model index scores, and taking the average value of the k model index scores to obtain the final score of the decision tree model;
and 8: carrying out data processing on mobile phone data of unknown user types according to the steps 1 and 2, and calculating user travel indexes; and (4) inputting the travel indexes into the decision tree recognition models obtained in the steps 6 and 7, and outputting corresponding user type labels.
2. The method of claim 1, wherein the method comprises the steps of: in step 5, the specific calculation method of each index is as follows:
1) average weekly trip times N: the average value of the weekly travel times of the user is referred to;
Figure FDA0003552284080000011
wherein QjThe total number of j week trips of the user is represented, and n represents n weeks included in the study period;
2) ratio of travel P in peak periodg: the ratio of the number of trips of the user in the peak period to the total number of trips is shown;
Figure FDA0003552284080000021
wherein: qgRepresenting the trip times of the user in the peak time period, and Q representing the total trip times of the user;
3) city-crossing trip ratio P of peak dated: the ratio of the number of trips of the user on a peak date to the total number of trips crossing the city is shown;
Figure FDA0003552284080000022
wherein: qdRepresenting the number of times a user travels across the city on a peak date, QcRepresenting the total number of times of travel of the user across cities;
4) duty ratio P for off-peak hours of working dayf: the ratio of the number of trips of the user in the off-peak period during the working day to the total number of trips is referred to;
Figure FDA0003552284080000023
wherein: qfRepresenting the number of trips during the off-peak period of the user's working day;
5) working sunrise ratio Pw: the ratio of the number of trips of the user in the working day period to the total number of trips is referred to;
Figure FDA0003552284080000024
wherein: qwRepresenting the number of trips of the user during the early working day;
6) cross city trip ratio Pc: the ratio of the number of times of the user going out across the city to the total number of times of the user going out is referred to;
Figure FDA0003552284080000025
CN202210267171.5A 2022-03-17 2022-03-17 Method for identifying cross-city commuting crowd by utilizing mobile phone data Pending CN114611622A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210267171.5A CN114611622A (en) 2022-03-17 2022-03-17 Method for identifying cross-city commuting crowd by utilizing mobile phone data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210267171.5A CN114611622A (en) 2022-03-17 2022-03-17 Method for identifying cross-city commuting crowd by utilizing mobile phone data

Publications (1)

Publication Number Publication Date
CN114611622A true CN114611622A (en) 2022-06-10

Family

ID=81864604

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210267171.5A Pending CN114611622A (en) 2022-03-17 2022-03-17 Method for identifying cross-city commuting crowd by utilizing mobile phone data

Country Status (1)

Country Link
CN (1) CN114611622A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115440040A (en) * 2022-09-02 2022-12-06 重庆大学 Commuting vehicle identification method based on highway traffic data

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115440040A (en) * 2022-09-02 2022-12-06 重庆大学 Commuting vehicle identification method based on highway traffic data
CN115440040B (en) * 2022-09-02 2023-09-22 重庆大学 Commuter vehicle identification method based on expressway traffic data

Similar Documents

Publication Publication Date Title
CN110245981B (en) Crowd type identification method based on mobile phone signaling data
CN111582948B (en) Individual behavior analysis method based on mobile phone signaling data and POI (Point of interest)
CN106931974B (en) Method for calculating personal commuting distance based on mobile terminal GPS positioning data record
CN106651424A (en) Electric power user figure establishment and analysis method based on big data technology
CN111985710A (en) Bus passenger trip station prediction method, storage medium and server
CN111737605A (en) Travel purpose identification method and device based on mobile phone signaling data
CN107194525A (en) A kind of down town appraisal procedure based on mobile phone signaling
CN112215666A (en) Characteristic identification method for different trip activities based on mobile phone positioning data
CN107016042B (en) Address information verification system based on user position log
Yao et al. Data-driven choice set generation and estimation of route choice models
CN115086880B (en) Travel characteristic identification method, device, equipment and storage medium
Mo et al. Individual mobility prediction in mass transit systems using smart card data: An interpretable activity-based hidden Markov approach
CN111510368B (en) Family group identification method, device, equipment and computer readable storage medium
CN112949784B (en) Resident trip chain model construction method and resident trip chain acquisition method
CN117056823A (en) Method and system for identifying occupation type of shared bicycle commuter user
Chen et al. An analysis of movement patterns between zones using taxi GPS data
CN114611622A (en) Method for identifying cross-city commuting crowd by utilizing mobile phone data
CN108596664A (en) A kind of unilateral tranaction costs of electronic ticket determine method, system and device
CN111104468A (en) Method for deducing user activity based on semantic track
CN113158084A (en) Method and device for processing movement track data, computer equipment and storage medium
Chen et al. A travel mode identification framework based on cellular signaling data
Yang et al. Mobility pattern identification based on mobile phone data
Xu et al. MM-UrbanFAC: Urban functional area classification model based on multimodal machine learning
Sari Aslam et al. Trip purpose identification using pairwise constraints based semi-supervised clustering
Afolabi et al. When and where? Proactively predicting traffic accident in South Africa: our machine learning competition winning approach

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