Intercity low-frequency commuting crowd identification method based on mobile phone signaling data
By analyzing mobile phone signaling data, identifying the behavioral patterns of commuters from the short term, the problem of difficult to identify this group in the existing technology is solved, and technical support for the co-construction of urban agglomerations and urbanized regional facilities is provided, which has achieved the goal of realizing the problem of difficult to identify this group in the existing technology. Accurate identification and analysis of complex behavior patterns.
Patent Information
- Application Number
- CN202411184034.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-27
- Publication Date
- 2024-11-12
AI Technical Summary
The existing technology is difficult to identify commuters who travel between the two cities for a short period of time and multiple days, and lacks accurate identification of this complex behavior pattern, which affects the co-construction and sharing of facilities and public service configuration in urban circles, urbanized and integrated areas.
By obtaining mobile phone signaling data, we can judge the user's activity point data, distinguish daytime stayers and night stayers, calculate the number of monthly occurrences of daytime employed people and night residents and the number of different places of residence, and filter out short-term multi-day round trip users. Identify their workplace and residence and identify the commuter population.
It has achieved accurate identification of the short-term multi-day commuter group, provided technical support for the co-construction and sharing of facilities and public service configuration in urban circles, urbanized and integrated areas, and improved the ability to analyze complex behavior patterns.
Smart Images

Figure CN118945607A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of identification and analysis, and in particular to a method for identifying inter-city low-frequency commuters based on mobile phone signaling data. Background Art
[0002] The application of mobile phone signaling data in population activity monitoring has become a general trend. Compared with population census and population sampling survey data, population monitoring based on mobile phone signaling data has the advantages of lower cost, more comprehensive caliber, and stronger dynamism. Therefore, in recent years, mobile phone signaling data has been widely used in the field of spatial planning, and the application technology has basically matured. With the high-quality development of urban agglomerations, metropolitan areas, and urbanization, different cities in the region rely on convenient transportation networks, and cross-city commuters have gradually emerged.
[0003] Among them, according to the frequency of commuting, some people commute between their city of residence and their city of work every day, some people are in their city of work on weekdays and return to their city of residence on weekends, and some people are in their city of work for multiple days (two or three days, four or five days, half a month) and return to their city of residence on weekends. At present, there have been relevant studies on the first two types of cross-city commuters with fixed commuting time periods between the two places, but there is no identification technology for the third type of commuters who travel between the two cities for short periods of time. Identifying the third type of commuters who travel between the two cities for short periods of time is of great significance to the co-construction and sharing of facilities and the allocation of public services in urban agglomerations, urbanized and integrated areas.
[0004] Therefore, in order to identify commuters who travel between two cities for a short period of time or for multiple days, a method for identifying inter-city low-frequency commuters based on mobile phone signaling data is needed. This method can accurately identify target users, facilitate the analysis of users' complex behavior patterns, and provide technical support for the co-construction and sharing of facilities in urban agglomerations, urbanized areas, and integrated regions. Summary of the invention
[0005] In view of this, the purpose of the present invention is to overcome the defects in the prior art and provide a method for identifying inter-city low-frequency commuters based on mobile phone signaling data, which can accurately identify target users, facilitate the analysis of users' complex behavior patterns, and provide technical support for the co-construction and sharing of facilities in urban agglomerations, urbanized areas and integrated regions.
[0006] The method for identifying inter-city low-frequency commuters based on mobile phone signaling data of the present invention comprises:
[0007] Obtain the user's activity point data, add the area information to the user's activity point data, and obtain the added activity point data;
[0008] According to the added activity point data, determine whether the user is a daytime stayer or a nighttime stayer;
[0009] Determine whether the user is a daytime employee or a nighttime resident; calculate the number of days per month that daytime employees and nighttime residents live in the same place of residence and the number of different places of residence in a month;
[0010] The users whose number of different residences in a month is greater than a are taken as the first target users, and the users whose number of days in the same residence in a month is greater than b days and less than c days are taken as the second target users;
[0011] The number of cycles that the second target user travels to and from each residence in a month is calculated, and the second target users whose corresponding number of cycles is greater than d are regarded as identified users.
[0012] Furthermore, the activity point data is data of all target mobile phone users staying at a certain place at any time and the stay time exceeds 30 minutes.
[0013] Further, judging whether the user is a daytime stayer or a nighttime stayer includes:
[0014] Users who enter or leave between 00:00 and 06:00 are considered night stayers, and their city is considered a night stay location;
[0015] Users who enter or leave the system between 08:00 and 17:00 are daytime visitors, and their city is a daytime destination.
[0016] Furthermore, it is determined whether the user is a daytime employee or a nighttime resident, specifically including:
[0017] According to the four fields of user unique number, stay date, entry time, and city, extract the earliest entry time and the latest exit time of night stayers and day stayers in the same city every day;
[0018] Calculate the time TIME_day that daytime stayers spend in the same city from 08:00 to 17:00 every day, that is, TIMEE_day = latest departure time - earliest entry time;
[0019] Calculate the time TIME_night that night stayers spend in the same city from 00:00 to 06:00 every day, that is, TIME_night = latest departure time - earliest entry time;
[0020] When TIME_day of daytime stayers is greater than or equal to 5 hours, they are daytime employed people, and the daytime employment place is the work place; when TIME_night of nighttime stayers is greater than or equal to 4 hours, they are nighttime residents, and the nighttime residence place is the residence place.
[0021] Furthermore, traveling from residence A to residence B and then back to residence A is considered as one cycle.
[0022] Furthermore, it also includes: determining the workplace and multiple residences of commuters who travel between two or more cities for a short period of time, including:
[0023] The day in a month when a user appears at the same workplace the most times is the user's long-term workplace;
[0024] The place of residence in the same city as the place of work is the place of residence at work;
[0025] A place of residence that is not in the same city as the place of work is a leisure residence, and there can be one or more leisure residences.
[0026] The beneficial effects of the present invention are as follows: the present invention discloses a method for identifying inter-city low-frequency commuters based on mobile phone signaling data, by judging whether the user is a daytime employment group or a nighttime residence group; calculating the number of days per month that daytime employment groups and nighttime residence groups appear in the same residence and the number of different residences in a month; gradually screening users based on the number of days per month and the number of different residences in a month, and finally identifying target users. The present invention realizes accurate identification of target users, and provides technical support for the co-construction and sharing of facilities and public service configuration in metropolitan areas, intra-city and integrated regions. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] The present invention will be further described below in conjunction with the accompanying drawings and embodiments:
[0028] Figure 1 Schematic diagram of the crowd identification method of the present invention. DETAILED DESCRIPTION
[0029] The present invention is further described below in conjunction with the accompanying drawings, as shown in the drawings:
[0030] This embodiment discloses a method for identifying inter-city low-frequency commuters based on mobile phone signaling data, including the following steps:
[0031] Obtain the user's activity point data, add the area information to the user's activity point data, and obtain the added activity point data;
[0032] According to the added activity point data, determine whether the user is a daytime stayer or a nighttime stayer;
[0033] Determine whether the user is a daytime employee or a nighttime resident; calculate the number of days per month that daytime employees and nighttime residents live in the same place of residence and the number of different places of residence in a month;
[0034] The users whose number of different residences in a month is greater than a are taken as the first target users, and the users whose number of days in the same residence in a month is greater than b days and less than c days are taken as the second target users;
[0035] Calculate the number of cycles that the second target user travels to and from each residence in a month, and identify the second target users whose corresponding number of cycles is greater than d. Where a is 1, b is 4, c is 24, and d is 1.
[0036] In this embodiment, the activity point data of the mobile phone user is obtained. The activity point data refers to the data of all target mobile phone users staying in a certain place at any time and staying for more than 30 minutes; wherein, the update frequency is 2 seconds to 3 seconds, and the data format includes the user's unique number (MSID), the longitude (X) of the stay, the latitude (Y) of the stay, the date (DAY) of the stay, the entry time (INTIME), and the departure time (OUTTIME). The specific data format is shown in Table 1.
[0037] Table 1
[0038] User unique ID (MSID) Abcd … Stop longitude (X) 136.0112 … Stop latitude (Y) 29.8738 … Date of stay(DAY) 20230201 … INTIME 20230201081000 … Departure time (OUTTIME) 20230201085500 …
[0039] Add the region information to the user's activity point data. Based on the activity point data's longitude (X) and latitude (Y) fields, determine the city and administrative district where the user stayed, and add two fields, city (CITY) and administrative district (XZQ), to each activity point data. The format of the added data is shown in Table 2.
[0040] Table 2
[0041]
[0042]
[0043] In this embodiment, it is determined whether the user is a daytime stayer or a nighttime stayer. That is, if the user's entry time (INTIME) or exit time (OUTTIME) is in the time period of 00:00-06:00, the user is a nighttime stayer, and the city (CITY) where the user is located is a nighttime stay; if the user's entry time (INTIME) or exit time (OUTTIME) is in the time period of 08:00-17:00, the user is a daytime stayer, and the city (CITY) where the user is located is a daytime stay; the data format for setting the stay area is shown in Table 3.
[0044] Table 3
[0045] User unique ID (MSID) Abcd … Stop longitude (X) 136.0112 … Stop latitude (Y) 29.8738 … Date of stay(DAY) 20230201 … INTIME 20230201081000 … Departure time (OUTTIME) 20230201085500 … CITY Bishan District … Administrative region (XZQ) Bicheng Street Stay area Daytime stop …
[0046] In this embodiment, judging whether the user is a daytime employee or a nighttime resident based on the length of stay in a specific time period specifically includes:
[0047] 1) According to the four fields of user unique ID (MSID), stay date (DAY), entry time (INTIME), and city (CITY), extract the earliest entry time (INTIMEMIN) and the latest exit time (OUTIMEMAX) of night stayers and day stayers in the same city every day;
[0048] 2) Calculate the time consumption TIME_day of the daytime stayer in the same city during the period of 08:00-17:00 every day, that is, TIMEE_day = the latest departure time (OUTIMEMAX) - the earliest entry time (INTIMEMIN);
[0049] Calculate the time TIME_night that night stayers spend in the same city from 00:00 to 06:00 every day, that is, TIME_night = latest departure time (OUTIMEMAX) - earliest entry time (INTIMEMIN);
[0050] 3) When TIME_day of daytime stayers is greater than or equal to 5 hours, they are daytime employed people, and their daytime employment place is their work place (WORKi), where i represents the date; when TIME_night of nighttime stayers is greater than or equal to 4 hours, they are nighttime residents, and their nighttime residence place is their residence place (HOMEi); users who do not meet the requirements are deleted.
[0051] As shown in Table 4, it is determined that the user's TIME_day is 9.2 hours, which means that the user is employed during the day.
[0052] Table 4
[0053]
[0054] In this embodiment, the number of days per month and the number of different places of residence of daytime employed people and nighttime resident people are counted, specifically including:
[0055] 1) According to the above data analysis to determine whether the user is a daytime employed person or a nighttime resident person, the daily data of each user is processed to obtain the daytime employed person and nighttime resident person for each day of the month. The fields retained for each data are the user's unique ID (MSID), stay date (DAY), work place (WORKi) or residence place (HOMEi), and long-term stay time TIME_day or TIME_night;
[0056] 2) Count the number of days DAY_home that the user appears in the same residence (HOMEi) in a month and the number of different residences CNT_home in a month.
[0057] In this embodiment, it is determined whether the night-time residents live across cities, that is, whether they have multiple different residences. If the number of residences of the same user in a month CNT_home>1, there are multiple residences, and users who do not meet the requirements are deleted. It is determined whether the night-time residents who live across cities have a long-term and stable residence. If the number of days DAY_home that the same residence (HOMEi) appears is greater than 4 days and less than 24 days, it means that the probability of the user living in different cities for a long time within a month is greater, and the user is retained. Otherwise, the user who does not meet the requirements is deleted. Among them, DAY_home less than 4 days may be a residence for visiting relatives and friends, and greater than 24 days may be a long-term residence in one city, and occasional residence in another city, and neither belongs to the identification target population.
[0058] In this embodiment, users who travel back and forth to different residences for many times are judged, and users who live in different cities due to business trips, tourism, visiting relatives and friends, etc. are excluded. That is, the number of cycles (PER) of users traveling back and forth to each residence (HOMEi) in a month is counted, where going from residence A to residence B and then back to residence A is considered as one cycle.
[0059] If the number of cycles (PER) of a residence (HOMEi) is greater than 1, it means that the user moved to another place after living there, and then returned to the original place of residence to continue living there. In this way, this type of commuter who travels between two or more cities for a short period of time for multiple days is identified.
[0060] In this embodiment, the crowd identification method of the present invention further includes:
[0061] Determine the workplace and multiple residences of commuters who travel between two or more cities for a short period of time, including:
[0062] 1) The user's long-term workplace is the one with the most days (DAY_work) at the same work location (WORKi) in a month. Except for jobs with high mobility such as drivers, there is usually only one workplace;
[0063] 2) The place of residence in the same city as the place of work is the place of residence at work;
[0064] 3) A place of residence that is not in the same city as the place of work is a leisure residence, and there can be one or more leisure residences.
[0065] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.
Claims
1. A method for identifying inter-city low-frequency commuters based on mobile phone signaling data, characterized in that: include: Obtain the user's activity point data, add the area information to the user's activity point data, and obtain the added activity point data; According to the added activity point data, determine whether the user is a daytime stayer or a nighttime stayer; Determine whether the user is a daytime employee or a nighttime resident; Calculate the number of days per month that daytime employed people and nighttime resident people appear in the same place of residence and the number of different places of residence in a month; The users whose number of different residences in a month is greater than a are taken as the first target users, and the users whose number of days in the same residence in a month is greater than b days and less than c days are taken as the second target users; The number of cycles that the second target user travels to and from each residence in a month is calculated, and the second target users whose corresponding number of cycles is greater than d are regarded as identified users.
2. The method for identifying inter-city low-frequency commuters based on mobile phone signaling data according to claim 1 is characterized in that: The activity point data refers to the data of all target mobile phone users staying at a certain place at any time and the stay time exceeds 30 minutes.
3. The method for identifying inter-city low-frequency commuters based on mobile phone signaling data according to claim 1 is characterized in that: Determine whether the user is a daytime or nighttime visitor, including: Users who enter or leave between 00:00 and 06:00 are considered night stayers, and their city is considered a night stay location; Users who enter or leave the system between 08:00 and 17:00 are daytime visitors, and their city is a daytime destination.
4. The method for identifying inter-city low-frequency commuters based on mobile phone signaling data according to claim 1 is characterized in that: Determine whether the user is a daytime employee or a nighttime resident, including: According to the four fields of user unique number, stay date, entry time, and city, extract the earliest entry time and the latest exit time of night stayers and day stayers in the same city every day; Calculate the time TIME_day that daytime stayers spend in the same city from 08:00 to 17:00 every day, that is, TIMEE_day = latest departure time - earliest entry time; Calculate the time TIME_night that night stayers spend in the same city from 00:00 to 06:00 every day, that is, TIME_night = latest departure time - earliest entry time; When TIME_day of daytime stayers is greater than or equal to 5 hours, they are daytime employed people, and the daytime employment place is the work place; when TIME_night of nighttime stayers is greater than or equal to 4 hours, they are nighttime residents, and the nighttime residence place is the residence place.
5. The method for identifying inter-city low-frequency commuters based on mobile phone signaling data according to claim 1 is characterized in that: Traveling from residence A to residence B and then back to residence A is considered one cycle.
6. The method for identifying inter-city low-frequency commuters based on mobile phone signaling data according to claim 1 is characterized by: Also includes: Determine the workplace and multiple residences of commuters who travel between two or more cities for a short period of time, including: The day in a month when a user appears at the same workplace the most times is the user's long-term workplace; The place of residence in the same city as the place of work is the place of residence at work; A place of residence that is not in the same city as the place of work is a leisure residence, and there can be one or more leisure residences.
