CN112800348B - Tourism behavior identification method based on mobile phone signaling big data - Google Patents
Tourism behavior identification method based on mobile phone signaling big data Download PDFInfo
- Publication number
- CN112800348B CN112800348B CN202110103051.7A CN202110103051A CN112800348B CN 112800348 B CN112800348 B CN 112800348B CN 202110103051 A CN202110103051 A CN 202110103051A CN 112800348 B CN112800348 B CN 112800348B
- Authority
- CN
- China
- Prior art keywords
- travel
- user
- time
- mobile phone
- stay
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/9035—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/907—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/909—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using geographical or spatial information, e.g. location
-
- 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
Abstract
The invention discloses a tourism behavior identification method based on mobile big data, which is characterized in that a 'calibration general environment' mark is finished from a trip destination frequency based on a long-period track record of a mobile phone signaling, and a journey closed loop is divided by taking the 'calibration general environment' as an anchor point; calculating space-time thresholds of habitually nearby random trips in different town areas according to trip characteristics after stripping off relevant trips of the 'calibrated habitual environment', regarding the part exceeding the thresholds as non-habitually targeted trips, and regarding the part exceeding the thresholds as habitually nearby random trips to be equivalent to the habitual environment otherwise; after the closed-loop travel of specific POI information of hospitals, industrial parks and the like is removed from non-routine objective travel, the identification and calibration of tourism behaviors can be realized.
Description
Technical Field
The invention relates to the technical field of big data mining, in particular to a tourism behavior identification method based on mobile phone signaling big data.
Background
According to the definition about travel in the international travel statistics recommendation of 2008: "tourist" refers to a traveler traveling to some major destination outside of their usual environment for any primary purpose (for business, leisure, or other personal purposes, rather than being hired to a residential entity in a visited country or location), for a duration of less than one year. These trips by the guest meet tourist trip criteria. Travel refers to the activities of a guest. The travel of the home country, inbound or outbound tourist is called home country, inbound and outbound tourist, respectively. "it has defined the tourism action on 3 layers from trip purpose, duration, destination requirement, but in the actual operation process, to domestic (being the home country) tourism action, because the existence of factors such as the difficult discernment of trip purpose, customary environment identification standard uncertainty, etc., the mode such as tourism investigation and scenic spot passenger flow statistics report is usually adopted to carry out on the actual statistics operation.
With the development of big data technology, the tourist track calibration method based on mobile phone signaling is gradually applied to tourism statistics. Such as:
1. shanghai Tongji City planning and design research institute, Inc., Wangde et al in "population identification method and system based on mobile phone signaling data" CN 109992605A: respectively setting target time periods of a travel slack season and a travel peak season aiming at a target area, and respectively selecting target users of which the stay time of the target area exceeds a first time threshold in the two target time periods through mobile phone signaling data of the users; counting the staying days of each target user in the target area, detecting the staying days, and if the staying days of the target users in one target time period in the target area do not exceed the first day threshold and are greater than 0, and the staying days in the other target time period in the target area are 0, determining that the target users are the tourist population of the target area.
2. In a patent "population identification method and system based on mobile phone signaling data" CN109992605A, wangde et al, of shanghai Tongji city planning and design research institute, beijing Zhi digital space-time technology, ltd: the invention provides a method and a system for carrying out country trip behavior statistics based on user mobile information messages. The statistical method distinguishes urban areas and rural areas through the mobile population density, and then obtains the user statistics of a new tourism mode of touring in the rural areas (not entering scenic spots) through user signaling messages.
3. The national holiday travel statistics survey system which is formulated by the Ministry of culture and travel and published by the national statistics bureau in 12 months of 2020 is specifically referred to in the specification: domestic tourists refer to Chinese (continent) residents who leave the usual living environment, travel distance exceeds 10 kilometers, travel time exceeds 6 hours, visit other places in China for sightseeing, leisure and vacation, visiting friends and visiting, health care and recuperation, shopping and entertainment, learning and communication, conference training or carrying out activities such as economy, culture, sports, religion and the like, and do not form employment relationship with destinations. In the scheme for measuring and calculating the holiday travel market in the appendix, more than 1 operator, large map provider or SDK service provider (users are not less than 2 hundred million people and are hereinafter referred to as location data providers) active users are definitely taken as samples, and the calculation range comprises tourists nationwide and domestic.
The existing domestic tourism behavior identification method comprises the following steps: the scheme of investigation and scenic spot reporting has the obvious problems of long sampling period, low sampling rate, high data acquisition cost and the like; in the existing scheme for carrying out travel statistics by adopting mobile phone signaling big data, a resident time length at a destination is mainly adopted to distinguish a standing user from a traveling user, and the scheme is a classification scheme adopting an association rule, and because the travel purpose and whether the destination is frequently subjected to environment change or not are not identified, and repeated statistics of a plurality of visiting places in the same trip can occur, the identification of the travel behavior cannot be well realized; although the survey system made by the ministry of travel and the like clearly defines the general environment, and divides travel distance and duration, and also mentions that the analysis can be carried out by adopting a sample of the position data provider, no relevant technical basis and implementation scheme are given.
Disclosure of Invention
In order to solve the defects of the prior art and realize the purpose of more accurate travel behavior identification, the invention adopts the following technical scheme:
a tourism behavior identification method based on mobile phone signaling big data comprises the following steps:
s1, calibrating the user, analyzing and identifying the user routinely, setting a routine screening threshold value according to a curvature analysis method by using the user stay in a specified place for a specified time as a screening index based on the user track of the mobile phone signaling, and calibrating the routine of the user;
s2, extracting a closed-loop journey taking the nominal anchor point as a standard, wherein the closed-loop journey means that the user passes through a section of track taking the nominal anchor point as a standard and then returns to the anchor point, and the section of track is taken as the closed-loop journey of the user going out;
s3, analyzing and filtering a space-time threshold value of a routine random trip nearby, removing and calibrating the routine trip for all closed-loop trips of users, counting the remaining trips, calculating to obtain a space-time threshold value of the random trip, and filtering the random trip through the space-time threshold value to obtain an effective trip of the tour;
s4, removing scene destinations, and removing travel of non-travel destinations of the user from effective travel;
and S5, outputting the result of the tourism behavior.
Further, the step S1 includes the following steps:
s11, summarizing track data of users not less than one year, and calculating the stay times and the stay months of each user at a specified position within one year, wherein the specified position refers to a grid with the range of 250 meters, the stay times take days as a calculation unit, the stay times in one day are recorded as one time, and the stay months refer to the number of months that the user has stayed at the specified position;
s12, counting the grid number of each stay time for all users, and averaging all users in each stay time;
s13, performing curve fitting on the distribution data by adopting a power function, calculating the curvature of the fitted curve, and taking the stay times D corresponding to the maximum curvature value as a customary screening threshold value;
s14, regarding the grids with the user staying times less than D, the long-interval periodic travel is also determined as a usual travel, the travel for more than or equal to 3 months is determined as a formed periodic travel habit, and the position with the staying month number M of more than or equal to 3 is also used as the usual travel of the user;
s15, screening the data obtained in the step S11 through the calculated habitually screening threshold values D and M to obtain the habitually corresponding to the user, and calibrating the grid staying most at night as the habitual residence place and calibrating the grid staying most at day as the habitual working place.
Further, the step S13 includes the following steps:
s131, the curve fitting basis function is as follows: y ═ a × xbWhere a and b are constants to be estimated, x is the number of dwells and the range is [1,365 ]]Y is the number of grids;
s132, calculating the curvature of the power function, wherein the curvature calculation formula is as follows: k ═ y "|/[ (1+ (y')2)](2/3)Wherein y' is a power function oneOrder derivatives, i.e. y' ═ ab x(b-1)Y "is a second derivative of the power function, i.e. y ═ ab (b-1) × x(b-2)(ii) a And drawing curvature trends at intervals of 0.1, taking the point corresponding days D with the maximum curvature as a user habitually times screening threshold, and when the user stays on a certain grid for more than D, considering that the place is habitually used by the user.
Further, the step S2 includes the following steps:
s21, taking the residence convention as an anchor point, calibrating in a mobile phone signaling track, taking the moment when the anchor point departs from the range with the peripheral radius R as the departure time and the return time as the arrival time;
s22, if the user comes from outside the analysis area, the user takes the access boundary as the departure time and the arrival time;
s23, forming a closed-loop travel of the user by all the tracks in the departure time and the arrival time;
and S24, calculating the travel distance and the travel time length of the closed-loop travel.
Further, in step S3, after all the user closed-loop trips are eliminated and calibrated routine trips, the joint distribution of the maximum distance and the trip duration of the remaining trips is counted: counting by taking 1km and 1 hour as data intervals respectively, and when the journey is concentrated in the vicinity of the habitually, and the journey time is short, the number of the journeys is more, and the journey is considered to belong to random trips in the vicinity of the habitually; when the travel time length and the maximum distance of the travel are gradually increased at the same time, the user is considered to be probably travelling; obtaining a space-time threshold value of random travel by a gradient descent mode:
wherein n _ (t, d) represents the value of time t and distance d; and when the delta is less than or equal to 1, calculating the maximum value of the delta, taking the travel time T and the travel distance D 'corresponding to the maximum value as time and distance dividing points of the effective travel, and when the travel time is more than or equal to T hours and the travel distance D' is kilometers, considering the travel as non-random travel, namely the travel is taken as the effective travel for travel.
Further, the step S4 includes the following steps:
s41, taking the position with the longest stay time in the travel as the travel destination;
s42, eliminating the journey with the destination being the non-travel scene type.
Further, conventionally, in step S15, considering the spatial error, a region with a radius r and a center point of the grid as a circle center is conventionally used, and r is determined by the positioning error, so that the calculation speed can be increased.
Further, in the closed-loop route of the step S2, the residence convention and the working convention occur at the same time, and the residence convention is taken as the standard.
Further, the size of R in step S21 depends on the positioning accuracy, and 1km is selected according to the mobile phone signaling positioning.
Further, the travel distance in step S24 is a weighted sum of linear distances of the trajectory stopping points, and the weight is a non-linear coefficient of the city, where the non-linear coefficient is a ratio of an actual traffic distance between a road start-destination point and a road end-destination point to a spatial linear distance between two points, and is used to evaluate the convenience of the relationship between different road network types and the passenger/cargo flow line collecting and distributing points.
The invention has the advantages and beneficial effects that:
the method is based on big data mining technology, starts from tourism definition, takes a mobile phone signaling track as a data base, and is combined with a data analysis method to identify the tourism behavior of the user. Compared with the prior art, the method increases the schemes of habitually identifying the user, identifying the closed-loop journey taking the habitually as an anchor point, habitually randomly going out around and filtering the destination, is more accurate and scientific than the traditional rule identification, and has popularization.
Drawings
Fig. 1 is a schematic structural diagram of functional modules corresponding to steps in the present invention.
FIG. 2 is a graph of the mean of the number of dwells and the number of grids in the present invention.
FIG. 3 is a graph of a profile curve fit according to the present invention.
FIG. 4 is a graph of the power function curvature of the distribution plot of the present invention.
Fig. 5 is a conventional closed loop stroke schematic of the present invention.
Fig. 6 is a combined distribution diagram of maximum distance and travel time length in the present invention.
FIG. 7 is a gradient profile of maximum distance and travel duration in the present invention.
FIG. 8 is a chart showing the result of tourism behavior in the present invention.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating the present invention, are given by way of illustration and explanation only, not limitation.
The invention provides a tourism behavior identification method based on mobile big data, which mainly adopts the steps that a 'calibration general environment' mark is finished from a trip destination frequency based on a long-period track record of mobile phone signaling, and a journey closed loop is divided by taking the 'calibration general environment' as an anchor point; calculating space-time threshold values of habitually nearby random trips in different town areas according to trip characteristics after stripping of relevant trips of 'calibrating habitual environment', and regarding the parts exceeding the threshold values as non-habitually targeted trips (otherwise, regarding as habitually nearby random trips, and equivalent habitual environment); after the closed-loop travel of specific POI information of hospitals, industrial parks and the like is removed from non-routine objective travel, the identification and calibration of tourism behaviors can be realized.
As shown in fig. 1, the method for identifying a tourism behavior based on big data of a mobile phone signaling is an operation performed on an extracted and obtained standardized grid residence track (250 meters by 250 meters, 30-minute residence time) after a user track based on the mobile phone signaling is extracted, and includes the following steps:
1. user calibration, general environment analysis and identification;
2. extracting a closed loop journey by taking a calibrated usual environment as an anchor point;
3. analyzing and filtering a space-time threshold value of a conventional nearby random trip;
4. removing scene destinations;
5. and (5) outputting the result to a system.
Step 1: user-calibrated routine environment analysis recognition
The conventional identification of the user mainly limits the user to stay in a certain place for a certain time or form regular travel behaviors, and the device uses the number of days and months the user stays in the certain place as a screening index and sets a reasonable threshold value according to a curvature analysis method.
(1) Summarizing track data of users not less than 1 year (the example adopts data of 1 year in a certain province), and calculating the number of stay times and the number of stay months of each user at a certain position in one year, wherein the position refers to a grid with the range of 250 meters, the number of stay times in days is calculated as one time, and the number of stay months refers to how many months the user stays at the position. Data samples are as follows:
| subscriber number | Grid ID | Grid central latitude | Grid center longitude | Days of stay (D) | Number of stay months (M) |
| 138****4466 | 23887 | 29.44778 | 120.33778 | 6 | 4 |
| 138****7788 | 23669 | 29.43778 | 120.31778 | 5 | 3 |
(2) Counting the number of grids under each stay frequency for all users, and averaging all users under each stay frequency; the average distribution is shown in fig. 2, and it can be seen that the number of grids where the user stays for one day is the largest, and the number of grids has a significantly decreasing tendency (conventionally concentrated in a partial area) in a range where the number of stays is small.
(3) Performing curve fitting on the distribution by adopting a power function, calculating the curvature of the fitted curve, and taking the stay times corresponding to the maximum curvature as a customarily screening threshold;
the curve fitting basis function is: y ═ a × xbWhere a and b are constants to be estimated, x is the number of dwells and the range is [1,365 ]]Y is the number of grids; after fitting, as shown in fig. 3, where the points are the original distribution and the lines are the fitted power functions.
Calculating the curvature of the power function, wherein the curvature calculation formula is as follows: k ═ y "|/[ (1+ (y')2)](2/3)Where y 'is the first derivative of the power function, i.e. y' ab x(b-1)Y "is a second derivative of the power function, i.e. y ═ ab (b-1) × x(b-2)(ii) a And drawing curvature trends at intervals of 0.1, taking the point corresponding days D with the maximum curvature as a user habitually times screening threshold, and when the user stays on a certain grid for more than D, considering that the place is habitually used by the user. This calculation yields D ═ 6 (actual calculation 5.3, rounded up, as shown in fig. 4).
(4) For grids with the user stay times smaller than D, the long-interval periodic trip can also be considered as a usual trip; it is generally considered that a periodic travel habit is formed by traveling for 3 months or more, and therefore, a place where the number of staying months is 3 or more is also used as a user's habit.
(5) The original data in (1) is filtered through the calculated D and M thresholds (D > 6or D <6& M >3 in the example data), and the corresponding habitual ground of the user is obtained (usually, a region with a radius r around the selected grid is calculated considering the spatial error, and r is usually determined by the positioning error, and is usually 1000 meters). The grid with the most night stays is designated as usual for living and the most day stays is usual for working.
Step 2: conventionally closed loop trip extraction
The closed-loop journey means that the user passes through a section of track from the residence habitually or the work habitually serving as an anchor point, and then returns to the anchor point, and the section of track is used as the closed-loop journey of the user (the residence habitually and the work habitually occur in the closed-loop journey at the same time, so that the residence habitually is the standard), as shown in fig. 5. The specific implementation method comprises the following steps:
(1) and (2) with the residence as an anchor point (identified in step 1), calibrating the range time departing from the periphery of the mobile phone in the mobile phone signaling track with the radius of R (the size of R depends on positioning accuracy, and the mobile phone signaling positioning is selected as 1km according to experience) as the departure time and the return time to the arrival time.
(2) If the user comes from outside the analysis area (if the analysis is Z province, and the tourist comes from outside the Z province), the user takes the entrance and exit boundary as the departure and arrival time.
(3) All the tracks in the departure time and the arrival time form a travel closed-loop travel of the system.
(4) And calculating the travel distance (the weighted sum of the linear distances of the track stopping points is adopted, and the weight is the nonlinear coefficient of the city) and the travel duration of the closed-loop travel.
Note: the non-linear coefficient refers to the ratio of the actual traffic distance between the origin-destination points to the spatial linear distance between the two points. The convenience degree of the relation between different road network types and the gathering and scattering points of the passenger and cargo flow lines can be evaluated. Nonlinear coefficients: chessboard type road network
② the radioactive road network is generally about 2.6. ③ the ring-shaped and radioactive road network is generally
And step 3: routine nearby random trip spatiotemporal threshold analysis and filtering
The spatiotemporal threshold of the usual nearby random trips is calculated and filtered. For all the user closed-loop trips, after relevant trips containing more than 2 'calibration routine environments' (including anchor points and intermediate points) are eliminated, the remaining trips are analyzed, and the intermediate points refer to other calibration routine except living routine and working routine, such as a common supermarket and the like. And (3) counting the combined distribution of the maximum distance of the remaining travel and the travel time length: statistics were performed with 1km and 1 hour data intervals, respectively. Data for a city as an example is shown in FIG. 6:
generally, when the travel is more concentrated near the habitually used residence or work and the time is shorter, the more the travel is, the more random travel in the habitually used vicinity is considered; when the time and the distance are gradually increased at the same time, the user is considered to be probably travelling; the scheme adopts a gradient descending mode to obtain the space-time threshold value of random travel. Wherein, the time distance interval gradient is calculated according to the following formula:
wherein n _ (t, d) represents the value of time t and distance d; and when the delta is less than or equal to 1, calculating the maximum value of the delta, and setting T and D corresponding to the maximum value as the time-distance dividing point of the effective stroke. As shown in fig. 7, 2- >3km away and 2- >3 hours in time; the threshold is therefore chosen to be t-2 and d-3. When the travel time is more than or equal to 3 hours and the travel distance is more than or equal to 3 kilometers, the travel is considered as a non-random trip and can be used as an effective travel for tourism.
And 4, step 4: scenarioized destination culling
And (3) removing the travel of the user without the purpose of activities such as entertainment and sightseeing from the result of the step 3, wherein the main steps are as follows:
(1) taking the position with the longest stay time in the travel as a travel destination;
(2) the elimination destination is the travel with the scene type being hospital, office, industrial park, etc.
And 5: result output
According to the closed-loop journey output in the step 4, the travel behavior is identified, and meanwhile, the travel behavior can be further divided into cross-provincial travel, overnight travel and the like according to relevant characteristics, as shown in fig. 8.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. A tourism behavior identification method based on mobile phone signaling big data is characterized by comprising the following steps:
s1, calibrating the user, analyzing and identifying the user routinely, setting a routine screening threshold value according to a curvature analysis method by using the user stay in a specified place for a specified time as a screening index based on the user track of the mobile phone signaling, and calibrating the routine of the user; the curvature analysis method adopts a power function to perform curve fitting on the distribution data, calculates the curvature of the fitted curve, and takes the stay times D corresponding to the maximum curvature value as a customary screening threshold value;
s2, extracting a closed-loop journey taking the nominal anchor point as a standard, wherein the closed-loop journey means that the user passes through a section of track taking the nominal anchor point as a standard and then returns to the anchor point, and the section of track is taken as the closed-loop journey of the user going out;
s3, analyzing and filtering a space-time threshold value of a routine random trip nearby, removing and calibrating the routine trip for all closed-loop trips of users, counting the remaining trips, calculating to obtain a space-time threshold value of the random trip, and filtering the random trip through the space-time threshold value to obtain an effective trip of the tour;
s4, removing scene destinations, and removing travel of non-travel destinations of the user from effective travel;
and S5, outputting the result of the tourism behavior.
2. The method for identifying tourism behavior based on the big data of mobile phone signaling as claimed in claim 1, wherein said step S1 comprises the steps of:
s11, summarizing track data of users not less than one year, and calculating the stay times and the stay months of each user at a specified position within one year, wherein the specified position refers to a grid with the range of 250 meters, the stay times take days as a calculation unit, the stay times in one day are recorded as one time, and the stay months refer to the number of months that the user has stayed at the specified position;
s12, counting the grid number of each stay time for all users, and averaging all users in each stay time;
s13, performing curve fitting on the distribution data by adopting a power function, calculating the curvature of the fitted curve, and taking the stay times D corresponding to the maximum curvature value as a customary screening threshold value;
s14, regarding the grids with the user staying times less than D, the long-interval periodic travel is also determined as a usual travel, the travel for more than or equal to 3 months is determined as a formed periodic travel habit, and the position with the staying month number M of more than or equal to 3 is also used as the usual travel of the user;
s15, screening the data obtained in the step S11 through the calculated habitually screening threshold values D and M to obtain the habitually corresponding to the user, and calibrating the grid staying most at night as the habitual residence place and calibrating the grid staying most at day as the habitual working place.
3. The method for identifying tourism behavior based on the mobile phone signaling big data as claimed in claim 2, wherein said step S13 comprises the steps of:
s131, the curve fitting basis function is as follows: y = a xbWhere a and b are constants to be estimated, x is the number of dwells and the range is [1,365 ]]Y is the number of grids;
s132, calculating the curvature of the power function, wherein the curvature calculation formula is as follows: k = | y '|/[ (1+ (y')2)](2⁄3)Wherein y 'is the first derivative of the power function, i.e. y' = ab x(b-1)Y' is a second derivative of the power function, i.e., y = ab (b-1) × x(b-2)(ii) a And drawing curvature trends at intervals of 0.1, taking the point corresponding days D with the maximum curvature as a user habitually times screening threshold, and when the user stays on a certain grid for more than D, considering that the place is habitually used by the user.
4. The method for identifying tourism behavior based on the mobile phone signaling big data as claimed in claim 2, wherein said step S2 comprises the steps of:
s21, taking the residence convention as an anchor point, calibrating in a mobile phone signaling track, taking the moment when the anchor point departs from the range with the peripheral radius R as the departure time and the return time as the arrival time;
s22, if the user comes from outside the analysis area, the user takes the access boundary as the departure time and the arrival time;
s23, forming a closed-loop travel of the user by all the tracks in the departure time and the arrival time;
and S24, calculating the travel distance and the travel time length of the closed-loop travel.
5. The method as claimed in claim 1, wherein in step S3, after removing and calibrating routine trips for all user closed-loop trips, the joint distribution of the maximum distance and the trip duration of the remaining trips is counted: counting by taking 1km and 1 hour as data intervals respectively, and when the journey is concentrated in the vicinity of the habitually, and the journey time is short, the number of the journeys is more, and the journey is considered to belong to random trips in the vicinity of the habitually; when the travel time length and the maximum distance of the travel are gradually increased at the same time, the user is considered to be probably travelling; obtaining a space-time threshold value of random travel by a gradient descent mode:
wherein n ist,dA value representing time t and distance d; and when the delta is less than or equal to 1, calculating the maximum value of the delta, taking the travel time T and the travel distance D 'corresponding to the maximum value as time and distance dividing points of the effective travel, and when the travel time is more than or equal to T hours and the travel distance D' is kilometers, considering the travel as non-random travel, namely the travel is taken as the effective travel for travel.
6. The method for identifying tourism behavior based on the big data of mobile phone signaling as claimed in claim 1, wherein said step S4 comprises the steps of:
s41, taking the position with the longest stay time in the travel as the travel destination;
s42, eliminating the journey with the destination being the non-travel scene type.
7. The method as claimed in claim 2, wherein the area of the step S15 is conventionally centered at the center point of the grid and has a radius r, and r is determined by the positioning error.
8. The method as claimed in claim 4, wherein in the closed-loop journey of step S2, the residence convention and the working convention occur simultaneously, based on the residence convention.
9. The method as claimed in claim 4, wherein the size of R in step S21 depends on the positioning accuracy, and is selected to be 1km according to the mobile phone signaling positioning.
10. The method as claimed in claim 4, wherein the travel distance in step S24 is calculated by weighted sum of linear distances of the track stop points, wherein the weight is a non-linear coefficient of a city, and the non-linear coefficient is a ratio of an actual traffic distance between the road start-end point and the road end-end point to a spatial linear distance between the two points.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110103051.7A CN112800348B (en) | 2021-01-26 | 2021-01-26 | Tourism behavior identification method based on mobile phone signaling big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110103051.7A CN112800348B (en) | 2021-01-26 | 2021-01-26 | Tourism behavior identification method based on mobile phone signaling big data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112800348A CN112800348A (en) | 2021-05-14 |
| CN112800348B true CN112800348B (en) | 2022-03-25 |
Family
ID=75811763
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110103051.7A Active CN112800348B (en) | 2021-01-26 | 2021-01-26 | Tourism behavior identification method based on mobile phone signaling big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112800348B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113963536A (en) * | 2021-10-19 | 2022-01-21 | 海南波罗密信息科技有限公司 | Method for identifying resident general environment and trip purpose through mobile phone signaling data |
| CN115034524A (en) * | 2022-08-11 | 2022-09-09 | 北京融信数联科技有限公司 | Method, system and storage medium for predicting working population based on mobile phone signaling |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105142106A (en) * | 2015-07-29 | 2015-12-09 | 西南交通大学 | Traveler home-work location identification and trip chain depicting method based on mobile phone signaling data |
| CN106570184A (en) * | 2016-11-11 | 2017-04-19 | 同济大学 | Method of extracting recreation-dwelling connection data set from mobile-phone signaling data |
| US10248912B1 (en) * | 2015-03-06 | 2019-04-02 | Google Llc | Systems and methods for displaying travel itineraries |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8073615B2 (en) * | 2007-05-02 | 2011-12-06 | Panasonic Corporation | Destination-prediction apparatus, destination-prediction method, and navigation apparatus |
| US20110301835A1 (en) * | 2010-06-07 | 2011-12-08 | James Bongiorno | Portable vacation/travel planner, and family tour guide device |
| US9435654B2 (en) * | 2013-06-01 | 2016-09-06 | Savari, Inc. | System and method for creating, storing, and updating local dynamic MAP database with safety attribute |
| US10510105B2 (en) * | 2016-06-10 | 2019-12-17 | Oath Inc. | Traveler recommendations |
-
2021
- 2021-01-26 CN CN202110103051.7A patent/CN112800348B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10248912B1 (en) * | 2015-03-06 | 2019-04-02 | Google Llc | Systems and methods for displaying travel itineraries |
| CN105142106A (en) * | 2015-07-29 | 2015-12-09 | 西南交通大学 | Traveler home-work location identification and trip chain depicting method based on mobile phone signaling data |
| CN106570184A (en) * | 2016-11-11 | 2017-04-19 | 同济大学 | Method of extracting recreation-dwelling connection data set from mobile-phone signaling data |
Non-Patent Citations (1)
| Title |
|---|
| 基于移动数据的旅游行为模式挖掘;孙天悦;《中国优秀博硕士学位论文全文数据库（硕士） 信息科技辑》;20181115;第I138-295页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112800348A (en) | 2021-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106096631B (en) | A kind of floating population's Classification and Identification analysis method based on mobile phone big data | |
| Wang et al. | Estimating dynamic origin-destination data and travel demand using cell phone network data | |
| Caceres et al. | Deriving origin–destination data from a mobile phone network | |
| Bonnel et al. | Passive mobile phone dataset to construct origin-destination matrix: potentials and limitations | |
| CN106600960B (en) | Travel origin-destination point identification method based on space-time clustering analysis algorithm | |
| CN109583640A (en) | A kind of Urban Traffic passenger flow attribute recognition approach based on multi-source location data | |
| CN112800348B (en) | Tourism behavior identification method based on mobile phone signaling big data | |
| CN102708680B (en) | Commute travel mode identification method based on AGPS technology | |
| US8185131B2 (en) | Method of providing location-based information from portable devices | |
| Qian et al. | Using mobile phone data to determine spatial correlations between tourism facilities | |
| CN107194525A (en) | A kind of down town appraisal procedure based on mobile phone signaling | |
| CN109299438A (en) | A kind of public transit facility supply level evaluation method based on net about car data | |
| CN107529135A (en) | User Activity type identification method based on smart machine data | |
| CN104303013B (en) | Location-based numerical data method and product | |
| CN111651529A (en) | Airport aviation passenger classification identification method based on mobile phone signaling data | |
| CN114501336B (en) | Road traffic volume measuring and calculating method and device, electronic equipment and storage medium | |
| Yang et al. | Detecting home and work locations from mobile phone cellular signaling data | |
| CN109978746B (en) | Population exchange amount estimation method for judging travel effectiveness based on signaling data and combined with dynamic threshold | |
| Duan et al. | MobilePulse: Dynamic profiling of land use pattern and OD matrix estimation from 10 million individual cell phone records in Shanghai | |
| CN113256978A (en) | Method and system for diagnosing urban congestion area and storage medium | |
| CN111508220B (en) | Method for accurately performing tail end connection based on public transport population distribution | |
| CN112734216A (en) | Resident trip purpose identification method based on communication positioning data | |
| CN115988422A (en) | Multilevel rail transit transfer passenger flow extraction method based on travel chain analysis | |
| CN110135679A (en) | A kind of appraisal procedure based on mobile phone signaling data Living Center space performance | |
| Batran et al. | Urban travel time estimation in greater maputo using mobile phone big data |
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 |