Method and device for identifying job and residence locations based on mobile phone signaling and current land use data

Abstract

The application discloses a kind of based on mobile phone signaling and present situation land use data's job and live place identification method and equipment, belong to big data analysis field, steps include: one, processing mobile phone signaling data and base station data, merging the base station that each other is stable switching point, and dividing base station set cell;Two, through mobile phone signaling data, with 4 hours as a time window, according to 1 hour time interval, the stay duration of user in each cell in different time windows is calculated by rolling;Three, extract the active trigger event record of user in different time windows;Four, through present situation land use data, the proportion of residential land of user stay point in the cell in different time windows is calculated;Five, establish evaluation index, determine user rest period and its residential area;Six, select the stay point of user stay duration maximum value and more than 2 hours outside rest period as work place.The application can accurately identify the work place and residential area of resident, provide support for city management, traffic planning.

Description

Technical Field

[0001] This invention belongs to the field of big data analysis, specifically relating to a method and device for identifying workplaces and residences based on mobile phone signaling and current land use data. Background Technology

[0002] Accurately identifying residents' workplaces and residences, and analyzing the distribution of work and residence, is the most fundamental and important task in urban transportation planning and urban master planning. Existing methods for identifying work and residence mainly include the time threshold method, the relative dwell time method, and the information entropy method.

[0003] The time threshold method and the relative dwell time method assume that residents work during the day and rest at home at night. They select dwell points with dwell times exceeding a threshold or maximum value as their residences by statistically analyzing the dwell time during the nighttime rest period. This method cannot accurately identify the residences of people with irregular work or rest schedules, and may even misclassify workplaces as residences. The information entropy method calculates the information entropy value of users at different times, assuming that the lower the information entropy value, the lower the user's spatial activity intensity. It selects the dwell point where the user's information entropy value is at its minimum as their residence. This method does not consider the possibility that users have no spatial activity at either their residence or workplace, and therefore cannot accurately identify the work-residence area of ​​people with low activity intensity during work hours. Summary of the Invention

[0004] To address the aforementioned issues, this invention proposes a method for identifying the workplace and residence based on mobile phone signaling and current land use data. This method overcomes the problem of low accuracy in identifying the workplace and residence of people with irregular work and rest schedules, which is a common issue with traditional methods. It can provide support for urban management and transportation planning.

[0005] The technical solution of this invention is as follows: A method for identifying the workplace and residence based on mobile phone signaling and current land use data, the specific steps of which are as follows:

[0006] Step 1: Process mobile phone signaling data and base station data, merge base stations that are stable handover points, divide them into base station cluster cells, and use base station cluster cells as spatial units to identify the user's location, so as to eliminate the impact of frequent base station handover caused by home users being in the signal coverage area of ​​multiple base stations, which may lead to misjudgment of the user's travel.

[0007] Step 2: Using mobile signaling data, calculate the user's dwell time in each cell within a 4-hour time window, with 1-hour intervals. (User time window) Inside, in the community The duration of stay is ;

[0008] Step 3: Extract records of user-triggered events generated within different time windows;

[0009] Step 4: Calculate the proportion of residential land in the community where the user stays within different time windows using current land use data;

[0010] Step 5: Establish evaluation indicators. The smaller the evaluation indicator, the greater the probability that the time window is the user's rest period, and the greater the probability that the user's location is their residence, thereby determining the user's rest period and residence.

[0011] Step 6: Select the location where the user stays for the longest time outside of rest periods, which is more than 2 hours, as the workplace.

[0012] Furthermore, the specific steps in step one are as follows:

[0013] S1. Obtain mobile phone signaling data and base station data;

[0014] S2, Filtering valid base stations

[0015] The number of daily records and daily users generated by base stations are counted. Base stations whose daily record count and daily user count fluctuate within 10% within a week are considered valid base stations.

[0016] S3, Determine the stable switching base station on a single day

[0017] The effective base station set is The system sequentially determines the daily stable handover base stations for each base station and extracts all data that have passed through the system. The base station uses the mobile phone signaling data of users to identify when a user is stationary and switches to other base stations. The corresponding number of switching times is Select Base stations with a value greater than the threshold β As Stable daily switching of base stations;

[0018] S4. Determine if a base station has been stably switched over multiple days.

[0019] All data from multiple days A stable daily handover base station, then for Stable base station handover;

[0020] S5. Merge base stations that are stable handover points and divide them into base station clusters.

[0021] like and If each other's base stations have been stably switching for several days, then it is considered... and Locations can be merged and divided into the same base station set cell.

[0022] Furthermore, the specific steps for step three are as follows:

[0023] S1, Statistics on users within a time window The number of calling event records generated internally is The number of proactively sent text message events recorded is ;

[0024] S2, User in time window Within this scope, the number of actively triggered event records is:

[0025] .

[0026] Furthermore, the specific steps for step four are as follows:

[0027] S1, within the time window Within, calculate the area of ​​the cell containing the base station cluster. The area of ​​various land uses within the community is 1-8 represent: residential land, public management and public service land, commercial service facilities land, industrial land, logistics and warehousing land, road and transportation facilities land, public utility facilities land, and green space and square land, respectively.

[0028] S2, User in time window Within the base station cluster, the proportion of residential land is as follows:

[0029] .

[0030] Furthermore, the specific steps in step five are as follows:

[0031] S1, Calculation of Evaluation Indicators

[0032] User in time window The evaluation indicators within are:

[0033]

[0034] ——Time Window Internally, evaluation indicators for determining place of residence;

[0035] —User Time Window Inside, in the community Duration of stay;

[0036] n—User time window The number of stops within;

[0037] —Users in time windows Within the area, the proportion of residential land in the cell complex where the base station is located;

[0038] —Users in time windows Within, the number of actively triggered event records generated;

[0039] —Calibration parameters, The value range is [10, 25]. The value range is (0,1];

[0040] S2, Calibration Parameters ,

[0041] Collect survey data to obtain users' actual rest periods and residences, and set constraints: the evaluation index Q value is minimized when users are at their residences, and the calibration parameters are adjusted accordingly. , Perform calibration and value acquisition;

[0042] S3. Select the time window with the smallest evaluation index Q value as the user's rest period, and select the location corresponding to the user's maximum stay time during the rest period as the user's residence.

[0043] The present invention also provides a workplace and residence identification device based on mobile phone signaling and current land use data, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The processor processes and executes the computer program to implement the workplace and residence identification method described above, accurately identifying the resident's workplace and residence.

[0044] The present invention has the following advantages and beneficial effects: The present invention has good accuracy and can accurately identify the workplace and residence of residents, providing accurate and effective data for analyzing the distribution of work and residence, and can provide support for urban management, transportation planning and other purposes. Attached Figure Description

[0045] Figure 1 This is a technical flowchart.

[0046] Figure 2 A statistical time period chart for user A who rests during the night;

[0047] Figure 3 A statistical time period chart for user B who are concentrated in resting during the day. Detailed Implementation

[0048] The invention will be further described in detail below with reference to the accompanying drawings:

[0049] Example 1

[0050] like Figure 1 As shown, a method for identifying the workplace and residence based on mobile phone signaling and current land use data includes the following specific steps:

[0051] Step 1: Process mobile phone signaling data and base station data, merge base stations that are stable handover points, divide them into base station cluster cells, and use base station cluster cells as spatial units to identify the user's location. This eliminates the impact on home users' travel caused by frequent base station handovers due to being within the signal coverage area of ​​multiple base stations. The specific steps are as follows:

[0052] S1. Obtain mobile phone signaling data and base station data;

[0053] S2, Filtering valid base stations

[0054] The number of daily records and daily users generated by base stations are counted. Base stations whose daily record count and daily user count fluctuate within 10% within a week are considered valid base stations.

[0055] S3, Determine the stable switching base station on a single day

[0056] The effective base station set is The daily stable handover base stations of each base station are judged in sequence, so as to... For example, extract all processes. The base station uses the mobile phone signaling data of users to identify when a user is stationary and switches to other base stations. The corresponding number of switching times is Select Base stations with a value greater than the threshold β As Stable daily switching of base stations;

[0057] S4. Determine if a base station has been stably switched over multiple days.

[0058] All data from multiple days A stable daily handover base station, then for Stable base station handover;

[0059] S5. Merge base stations that are stable handover points and divide them into base station clusters.

[0060] like and If each other's base stations have been stably switching for several days, then it is considered... and Locations can be merged and divided into the same base station set cell.

[0061] Step 2: Using mobile signaling data, calculate the user's dwell time in each cell within a 4-hour time window, with 1-hour intervals. (User time window) Inside, in the community The duration of stay is ;

[0062] Step 3: Extract records of user-triggered events generated within different time windows; the specific steps are as follows:

[0063] S1, Statistics on users within a time window The number of calling event records generated internally is The number of proactively sent text message events recorded is ;

[0064] S2, User in time window Within this scope, the number of actively triggered event records is:

[0065] .

[0066] Step 4: Calculate the proportion of residential land in the community where the user's stop point is located within different time windows using current land use data; the specific steps are as follows:

[0067] S1, within the time window Within, calculate the area of ​​the cell containing the base station cluster. The area of ​​various land uses within the community is 1-8 represent: residential land, public management and public service land, commercial service facilities land, industrial land, logistics and warehousing land, road and transportation facilities land, public utility facilities land, and green space and square land, respectively.

[0068] S2, User in time window Within the base station cluster, the proportion of residential land is as follows:

[0069] .

[0070] Step 5: Establish evaluation indicators. The smaller the evaluation indicator, the greater the probability that the time window is a user's rest period, and the greater the probability that the user's location is their residence. This determines the user's rest period and their residence. The specific steps are as follows:

[0071] S1, Calculation of Evaluation Indicators

[0072] User in time window The evaluation indicators within are:

[0073]

[0074] ——Time Window Internally, evaluation indicators for determining place of residence;

[0075] —User Time Window Inside, in the community Duration of stay;

[0076] n—User time window The number of stops within;

[0077] —Users in time windows Within the area, the proportion of residential land in the cell complex where the base station is located;

[0078] —Users in time windows Within, the number of actively triggered event records generated;

[0079] —Calibration parameters, The value range is [10, 25]. The value range is (0,1];

[0080] S2, Calibration Parameters ,

[0081] Collect survey data to obtain users' actual rest periods and residences, and set constraints: the evaluation index Q value is minimized when users are at their residences, and the calibration parameters are adjusted accordingly. , Perform calibration and value acquisition;

[0082] S3. Select the time window with the smallest evaluation index Q value as the user's rest period, and select the location corresponding to the user's maximum stay time during the rest period as the user's residence.

[0083] Step 6: Select the location where the user stays for the longest time outside of rest periods, which is more than 2 hours, as the workplace.

[0084] like Figure 2-3 As shown, User A's activities are mainly concentrated during the day, with a concentrated rest period from 00:00 to 04:00; User B's activities are mainly concentrated at night, with a concentrated rest period from 14:00 to 18:00. It can be seen that the method in this embodiment can accurately identify users' rest periods, and thus accurately identify residents' workplaces and residences, which can provide support for urban management, traffic planning, etc.

[0085] Example 2

[0086] This embodiment provides a workplace and residence identification device based on mobile phone signaling and current land use data, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor. The processor processes and executes the computer program to implement the workplace and residence identification method provided in Embodiment 1, which can accurately identify the workplace and residence of residents.

Claims

1. A method for identifying job-housing locations based on mobile phone signaling and current land use data, characterized in that, The specific steps are as follows: Step one, process mobile phone signaling data and base station data, merge base stations that are stable switching points to each other, divide base station set cells, identify user stay locations by taking base station set cells as spatial units, eliminate the influence of frequent base station switching caused by home users in multiple base station signal coverage ranges, and thus misjudge the user to generate travel; Step two, through the mobile phone signaling data, taking 4 hours as a time window, according to 1 hour time interval, the user's stay time in each cell in different time windows is calculated, the user's stay time in each cell in the time window is ; and ; and ; Step three, extract user active trigger event records generated in different time windows; Step four, calculate the proportion of residential land in the cell where the user stays in different time windows through current land use data; Step five, establish evaluation indexes, the smaller the evaluation index, the greater the probability of the time window as a user rest period, and the greater the probability of the user stay location being a residential area, thereby determining the user rest period and the residential area; The specific steps are as follows: S1, evaluation index calculation: the evaluation index of the user in the time window is: ， - time window inside, judging the evaluation index of the residential location; - user time window within a cell of stay duration; n - user time window n - number of dwell points; - the user's residence land proportion of the cell of the base station set in the time window ; - the user generates a number of proactive trigger event records within a time window - the user generates a number of proactive trigger event records within a time window a calibration parameter, a value range of [10, 25], a value range of (0, 1]; S2, calibration parameter , : Collect survey data to obtain the real rest period and residence of the user, and set the constraint condition: the evaluation index Q value of the user at the residence is minimum, and the calibration parameter , is calibrated and valued; S3, select the time window with the smallest evaluation index Q value as the user rest period, and select the location corresponding to the maximum stay duration of the user in the rest period as the user's residential area; Step six, select the stay point where the user stays for the maximum duration and more than 2 hours outside the rest period as the work place. 2.The method of claim 1, wherein the method comprises: obtaining a mobile phone signaling data and a current land use data; and identifying the place of residence based on the mobile phone signaling data and the current land use data. The specific steps of step one are as follows: S1, obtain mobile phone signaling data and base station data; S2, screen effective base stations: count the number of daily records and daily users generated by the base station, and record the base station with the number of daily records and daily users in a week within 10% as an effective base station; S3. Determine the base stations with stable handover on a single day: The set of valid base stations is... The system sequentially determines the daily stable handover base stations for each base station and extracts all data that have passed through the system. The base station uses the mobile phone signaling data of users to identify when a user is stationary and switches to other base stations. The corresponding number of switching times is Select Base stations with a value greater than the threshold β As Stable daily switching of base stations; S4, judging multi-day stable switching base station: if all the multi-day data are single-day stable switching base stations, then the stable switching base station is ​ S5, merging the base stations which are stable switching points to each other, dividing the base station set cell: if and the base stations are stable switching points to each other, it is considered that and the positions can be merged, and divided into the same base station set cell. 3.The method of claim 2, wherein the method further comprises: determining a residence location of the mobile phone based on the mobile phone signaling data and the current land use data. The specific steps of step three are as follows: S1, count the number of call event records generated by the user in a time window ;​​ S2, the user generates a number of active trigger event records within a time window S2, the user generates a number of active trigger event records within a time window 。 4. The method of claim 3, wherein the method is characterized by, The specific steps of step four are as follows: S1, calculate the area of the cells of the set of base stations in which the user is located within a time window 1-8 represent residential land, public management and public service land, commercial service facility land, industrial land, logistics and warehousing land, road and transportation facility land, public facility land, and green land and square land, respectively.​​ S2, the user is in the time window The proportion of residential land in the cell of the base station set where the user is in the time window 。 5. A device for identifying job-housing locations based on mobile phone signaling and current land use data, comprising a memory, a processor, and a computer program stored on the memory and capable of running on the processor, characterized in that, The processor processes and executes the computer program to realize the job and residence place identification method according to any one of claims 1-4, which can accurately identify the work place and residential area of the residents.

Images