House and job identification method and system based on mobile phone signaling

Abstract

The invention provides a job and house identification method and system based on mobile phone signaling. The method comprises the following steps: dividing daily user mobile phone signaling into a daytime data set and a night data set according to a job and house time window; mapping the coordinates of each base station into a first preset size grid, counting mobile phone signaling numbers corresponding to the base stations in each grid in the daytime and at night, and calculating the gravity center of each grid; constructing a neighborhood set in a preset distance range by taking the gravity center of each grid as a center, counting the number of mobile phone signaling of each neighborhood set, and taking the neighborhood set with the maximum number of signaling as a current-day job-housing candidate area; and mapping each base station coordinate to a second preset size grid, calculating a grid weight value through distribution weight accumulation according to the mobile phone signaling number of the job-house candidate area, comparing the grid weight value with a preset threshold value, and distinguishing a user work place grid and a residence place grid based on a comparison result. Through the scheme, distributed parallel computing based on SQL set operation can be realized, and the signaling data processing efficiency is improved.

Description

Technical Field

[0001] This invention belongs to the field of data mining technology, and in particular relates to a method and system for identifying the workplace and residence based on mobile phone signaling. Background Technology

[0002] With accelerating urbanization and population mobility, workplace-residence identification has become a core task in urban planning, traffic management, and demographic analysis. Mobile signaling data, as a passively collected large-scale location information source, has been widely used to detect an individual's home and workplace locations. This data records users' call, SMS, and data connection events, and performs location identification based on activity frequency, time patterns, and clustering techniques. These identification methods have been validated in real-world cases, supporting large-scale population mobility studies and resource allocation optimization.

[0003] Although existing research has widely utilized mobile signaling data for occupation-residence identification, and mature algorithms such as DBSCAN variants and random forest classification exist, traditional distributed DBSCAN implementations typically rely on iterative partitioning and merging when facing complete end-to-end verification and occupation-residence attribute identification of over 100 million active users and hundreds of billions of original signaling records across a province. Since it cannot be expressed as a pure declarative SQL set operation, data processing efficiency is usually low. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method and system for identifying work and residence based on mobile phone signaling, which solves the problem of low data processing efficiency in traditional work and residence identification based on the DBSCAN algorithm.

[0005] In a first aspect of the present invention, a method for identifying the workplace and residence based on mobile phone signaling is provided, comprising: Set a work-residence time window, and divide the daily user mobile phone signaling into daytime dataset and nighttime dataset according to the work-residence time window; Map the coordinates of each base station to a grid of the first preset size, count the number of mobile phone signaling in the daytime and nighttime datasets of each base station in each grid, and calculate the centroid of each grid. The centroid of each grid is used as the center to construct a neighborhood set within a predetermined distance. The number of mobile phone signaling in each neighborhood set is counted, and the neighborhood set with the most signaling is selected as the candidate work-residence area for the day. The coordinates of each base station are mapped to a grid of the second preset size. Based on the number of mobile phone signaling in the candidate work and residence areas and the distribution of monthly mobile phone signaling of users in each grid, the weight value of each grid under the second preset size is calculated by accumulating the distribution weights. The grid weight value is compared with a preset threshold, and the grid of the user's work location and the grid of the user's residence are distinguished based on the comparison results.

[0006] In a second aspect of the present invention, a workplace-residence identification system based on mobile phone signaling is provided, comprising: The time period segmentation module is used to set the work-residence time window and divide the daily user mobile phone signaling into daytime datasets and nighttime datasets according to the work-residence time window; The grid centroid calculation module is used to map the coordinates of each base station to a grid of the first preset size, count the number of mobile phone signaling in the daytime and nighttime datasets of each base station in each grid, and calculate the centroid of each grid. The candidate area identification module is used to construct a neighborhood set within a predetermined distance range with the centroid of each grid as the center, count the number of mobile phone signals in each neighborhood set, and select the neighborhood set with the most signaling signals as the candidate work-residence area for the day. The job-residence determination module is used to map the coordinates of each base station to a grid of a second preset size. Based on the number of mobile phone signaling in the job-residence candidate area and the distribution of the user's monthly mobile phone signaling in each grid, the weight value of each grid under the second preset size is calculated by accumulating the distribution weights. The grid weight value is compared with a preset threshold, and the user's work location grid and residence grid are distinguished based on the comparison result.

[0007] In a third aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable by the processor, wherein the processor executes the computer program to implement the steps of the method as described in the first aspect of the present invention.

[0008] In a fourth aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method provided in the first aspect of the present invention.

[0009] In this embodiment of the invention, the grid centroid is calculated, and a neighborhood set is constructed based on the grid centroid to determine the job-residence candidates. Combined with monthly signaling records, the job-residence attribute of the grid is determined through weight calculation. This avoids the point-by-point iterative traversal of the DBSCAN algorithm, realizing distributed computing based on SQL set operations, which can effectively improve the efficiency of signaling data processing and ensure the accuracy of job-residence identification. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1This is a flowchart illustrating a method for identifying the workplace and residence based on mobile phone signaling, provided as an embodiment of the present invention. Figure 2 This is a graphical diagram illustrating the workplace-residence recognition process according to an embodiment of the present invention. Figure 3 This is a schematic diagram of the recognition result based on the expansion of the neighborhood set, provided in one embodiment of the present invention; Figure 4 A schematic diagram of a workplace-residence identification system based on mobile phone signaling is provided as an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present invention. Detailed Implementation

[0012] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0013] It should be understood that the terms "comprising" and other similar expressions in the specification, claims, and accompanying drawings of this invention are intended to cover a non-exclusive inclusion, such as a process, method, system, or apparatus that includes a series of steps or units and is not limited to the listed steps or units. Furthermore, "first" and "second" are used to distinguish different objects and are not intended to describe a specific order.

[0014] Please see Figure 1 The present invention provides a flowchart illustrating a method for identifying a work-residence status based on mobile phone signaling, comprising: S101. Set the work-residence time window and divide the daily user mobile phone signaling into daytime dataset and nighttime dataset according to the work-residence time window. The work-residence time window refers to a user's work and residence time periods, which are generally fixed time ranges. For example, daytime work periods include 10:00–12:00 and 14:00–17:30, while nighttime periods include 22:00 of the previous day to 7:30 of the current day. Data for specific time periods can be filtered daily from all signaling data within a province / city / county. Daytime and nighttime datasets can be constructed separately for each day, with daily calculations independent of each other.

[0015] For example, the dataset filtered according to the definition of work and residence is: daytime dataset. The nighttime dataset includes records from 10:00–12:00 and 14:00–17:30. This includes records from 22:00 the previous day to 7:30 the current day. The dataset is formalized as follows: for each user u and date d, ,in, The base station's latitude and longitude coordinates. For time.

[0016] In this embodiment, daily daytime and nighttime datasets are constructed based on the definition of work and residence, which facilitates parallel processing using Hive and ensures that daily signaling data is calculated independently.

[0017] S102. Map the coordinates of each base station to a grid of the first preset size, count the number of mobile phone signaling in the daytime and nighttime datasets of each base station in each grid, and calculate the centroid of each grid. Base stations in a local area (such as a province, city, county, or district) are mapped to a grid of a predetermined size. The number of mobile phone signaling records of users in each grid during the day and night is counted (user signaling records are cached when users access base stations). Based on the number of mobile phone signaling records in the grid during the day and night, the centroid of the grid is calculated.

[0018] For example, the coordinates of base stations in Guangdong Province are mapped to 1 km. In a 1 km grid, the number of records in each grid is counted, which can be expressed as a number of records per user. coordinate set Define mesh function (A scaling factor of 100 corresponds to 0.01°≈1 km, which significantly reduces the computational load compared to projection calculations), and then the centroid of each grid is calculated.

[0019] Optionally, the centroid of each grid can be calculated based on the actual location of the base station and the number of mobile signaling records of the base station within the grid: ; In the formula, The number of user signaling entries within the grid, where p represents the base station coordinates. Indicates the center of gravity of the grid.

[0020] Unlike the core point search of the traditional DBSCAN algorithm, this embodiment performs pre-aggregation based on the centroid to adapt to the semantics of SQL sets, avoids global scanning, and optimizes data processing speed.

[0021] It is understandable that using grid clustering to perform statistical analysis by dividing the data into fixed grids and calculating the centroid of points within the grids can facilitate calculations beyond grid boundaries and meet the requirements of set operations. This avoids the need for traditional DBSCAN to perform traversal calculations for each point, thus effectively improving computational efficiency.

[0022] S103. Construct a neighborhood set within a predetermined distance range using the centroid of each grid as the center, count the number of mobile phone signaling in each neighborhood set, and select the neighborhood set with the most signaling as the candidate work-residence area for the day. Using the centroid of the grid as the center, construct neighborhood sets for each base station at predetermined distances, for example, constructing neighborhood sets with a radius of 600 meters. A base station may belong to multiple neighborhood sets. (Maximum 4) to form a multi-cluster overlapping set. Obtain the number of signaling records for each neighborhood set, and retain the cluster (i.e., the neighborhood set) with the most signaling records or the longest dwell time as the job / residence candidate for the day. The dwell time is generally determined by the time of entering and leaving the base station or by recording the user's location once every 5 minutes. The determination of the job / residence candidate area generally needs to meet the restrictions of dwell time (at least 1 hour) or number of signaling records (e.g., 12 records).

[0023] For example, with the centroid of each grid Centered on, construct Neighborhood set Get each collection The number of records, if If the threshold is 1 hour, then the cluster with the most signaling records will be retained as the candidate for work and residence for that day.

[0024] Centered on the centroid of each grid, a neighborhood set is further constructed according to the neighborhood radius, breaking through the constraints of the original boundary. At the same time, since the grid length, width, and neighborhood radius set in the early stage basically cover the range of signaling drift, a single cluster expansion can cover all possible drifting base stations of the fixed point, thereby optimizing SQL aggregation operations and ensuring the accuracy of job location judgment.

[0025] S104. Map the coordinates of each base station to a grid of the second preset size. Based on the number of mobile phone signaling in the candidate work and residence areas and the distribution of monthly mobile phone signaling of users in each grid, calculate the weight value of each grid under the second preset size by accumulating the distribution weight. Compare the grid weight value with a preset threshold and distinguish the grid of the user's work location and the grid of the user's residence based on the comparison result.

[0026] After projecting the reference coordinates onto the second preset size grid, based on the number of mobile phone signaling records in the candidate work-residence area and the monthly distribution of user mobile phone signaling records in each grid, the grid weight value is obtained by setting the signaling record weight and accumulating the monthly distribution weight. The work-residence attribute of the grid is determined according to the size of the grid weight value.

[0027] The second preset size grid is larger than the first preset size grid.

[0028] For example, the 30-day dataset is aggregated into a single collection, categorized by 1.5 km. Weighted dwell time is calculated using a 1.5 km grid (scaling factor 67 corresponds to approximately 0.015° ≈ 1.5 km). The total weight for each day is 1, and a weight is set for each individual signaling record. The grid weight value is obtained by summing the weights of the signaling records within the grid.

[0029] Optionally, if the grid weight value is greater than the first preset threshold, it is marked as a residential grid; if the grid weight value is greater than the second preset threshold, it is marked as a work grid. Both the first preset threshold and the second preset threshold are positive integers, and the first preset threshold is greater than the second preset threshold.

[0030] For example, if the grid weight value is greater than 18, it is marked as a residential grid; if the grid weight value is greater than 15, it is marked as a work grid.

[0031] In this embodiment, based on Hive's set-based programming paradigm and combined with the spatiotemporal sparsity of mobile signaling data and the coverage range of base stations (approximately 600 m to 1500 m), large-scale provincial-level workplace-residence identification can be achieved while ensuring data processing efficiency. Compared to the traditional DBSCAN algorithm, it does not pursue infinite expansion of clusters but focuses on identifying the fixed area (workplace / residence) where users spend the most time. Efficient distributed parallel computing is achieved through SQL set operations, avoiding the point-by-point iterative traversal of traditional DBSCAN and improving signaling data processing efficiency. Simultaneously, signaling drift is considered, and the core parameters of DBSCAN are mapped to the neighborhood radius and minimum sample size, ensuring the accuracy of workplace-residence identification.

[0032] It is understandable that existing distributed clustering technologies rely heavily on programming frameworks and UDFs, making it difficult to achieve efficient density clustering directly in a Hive environment. In this embodiment, no custom functions are required, and the core logic of DBSCAN can be completed with pure SQL set operations, meeting the job and residence identification needs of provincial-level data equivalent to hundreds of millions of users, and possessing production-grade scalability.

[0033] In one embodiment, step S104 includes: With a total daily weight of 1, the weight of a single signaling record within the grid is represented as... Then, the grid weights are calculated according to the formula:

[0034] In the formula, Indicates the first Japanese-style residential candidate area cluster Internal signaling record count Indicates the first Weight of a single signaling record per day denoted by , p represents the base station, and g represents the grid.

[0035] The total weight for each day is 1, and the weight of a single record is represented as: Then the grid weights can be expressed as ,according to The value is further used to determine the job-residence attribute of the grid. Based on the weighted accumulation calculation of signaling records in a single grid, combined with a fixed threshold, the job-residence attribute can be quickly determined.

[0036] In one embodiment, the base station weights in the user's work location grid and residence grid are calculated by weight accumulation, and the base station with the largest weight is selected as the grid marker point.

[0037] The base station with the largest weight within the grid is selected as the representative point. , This represents the base station weight. A representative point is selected as the geographical location where the user spends the most time during the day / night of a month.

[0038] This embodiment calculates representative base stations, not limited to regional judgment, which allows users to easily aggregate to different scales as needed, breaking through the original limitation of requiring a fixed surface.

[0039] In some embodiments, such as Figure 2 As shown, Figure 2 The diagram illustrates a graphical representation of workplace-residence identification. In "Grid Centroid Calculation," each grid cell represents approximately 1 km². The green cells are calculated to represent the maximum number of cells for the day. After calculating their centroids, surrounding base stations are searched. "Cluster Identification" indicates that points from adjacent grid cells are included in the day's cluster. Finally, a "Monthly Weight Calculation" is performed on all results from the past 30 days, dividing the area into 1.5 km² zones. The calculation is performed on a 1.5km grid, and the yellow grid represents the grid where the final identified work and residence location is located.

[0040] Based on the above solution, processing of 100 million users (not all of whom have a fixed place of residence and work) and hundreds of billions of line signaling data in a provincial-level environment was achieved (completed within 240 hours in actual testing), demonstrating usability in terms of production-grade SQL adaptation and large-scale robustness. Ultimately, using data from a certain month, 49.02 million people with a place of residence and a place of work were identified, 6.23 million people with a place of residence but no place of work were identified, and 16.58 million people with a place of work / fixed place of daytime residence but no place of residence were identified, for a total of 71.83 million people in the province.

[0041] Figure 3The identification results based on neighborhood set-based augmentation of street-level signaling data show a determination coefficient R² of 0.9887. Using a fixed-grid clustering algorithm with a 20-day threshold, the number of nighttime identifications reached 5.94 million, and adding daytime identifications brought the total to 14.07 million. The neighborhood set-based identification algorithm, using nighttime data with an 18-day threshold, achieved 13.17 million identifications, demonstrating superior performance compared to the fixed-grid method.

[0042] It is understood that the technical solutions related to public place image collection and personal identification involved in the embodiments strictly comply with the provisions of the "Personal Information Protection Law of the People's Republic of China" and Article 5 of the "Patent Law of the People's Republic of China," and do not violate the law, social morality, or harm the public interest. The application of this technical solution is limited to necessary scenarios for maintaining public safety, fully meeting the legal requirement of "necessary for maintaining public safety." Prominent warning signs will be set up during equipment installation to fully protect the public's right to know. The collected personal images and identification information are used only for public safety maintenance purposes and have no other uses. If, due to special circumstances, it is necessary to use them beyond this purpose, prior individual consent will be obtained. Furthermore, the design and implementation of the technical solution strictly comply with relevant national regulations, will not illegally collect or misuse personal information, and will not harm the legitimate rights and interests of the public or the normal social order, thus meeting the requirements of public order and good morals and the protection of the public interest.

[0043] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0044] Figure 4 This is a schematic diagram of a workplace-residence identification system based on mobile phone signaling, provided in an embodiment of the present invention. The system includes: The time period segmentation module 410 is used to set the work-residence time window and divide the daily user mobile phone signaling into daytime dataset and nighttime dataset according to the work-residence time window. The grid centroid calculation module 420 is used to map the coordinates of each base station to a grid of the first preset size, count the number of mobile phone signaling in the daytime and nighttime datasets of each base station in each grid, and calculate the centroid of each grid. Optionally, calculating the centroid of each grid cell includes: Calculate the centroid of each grid based on the actual location of the base station and the number of mobile signaling records of the base stations within the grid: ; In the formula, The number of user signaling entries within the grid, where p represents the base station coordinates. Indicates the center of gravity of the grid.

[0045] The candidate area identification module 430 is used to construct a neighborhood set within a predetermined distance range with the centroid of each grid as the center, count the number of mobile phone signals in each neighborhood set, and select the neighborhood set with the most signaling signals as the candidate work-residence area for the day. The job and residence determination module 440 is used to map the coordinates of each base station to a second preset size grid. Based on the number of mobile phone signaling in the job and residence candidate area and the distribution of the user's monthly mobile phone signaling in each grid, the weight value of each grid under the second preset size is calculated by accumulating the distribution weight. The grid weight value is compared with a preset threshold, and the user's work location grid and residence grid are distinguished based on the comparison result.

[0046] The second preset size grid is larger than the first preset size grid.

[0047] Optionally, if the grid weight value is greater than the first preset threshold, it is marked as a residential grid; if the grid weight value is greater than the second preset threshold, it is marked as a work grid. Both the first preset threshold and the second preset threshold are positive integers, and the first preset threshold is greater than the second preset threshold.

[0048] In one embodiment, the total daily weight is 1, and the weight of a single signaling record within the grid is represented as... Then, the grid weights are calculated according to the formula:

[0049] In the formula, Indicates the first Japanese-style residential candidate area cluster Internal signaling record count Indicates the first Weight of a single signaling record per day denoted by , p represents the base station, and g represents the grid.

[0050] In one embodiment, the work-residence determination module 440 includes: The base station marker identification unit is used to calculate the base station weights in the user's work area grid and residence grid by weight accumulation, and select the base station with the largest weight as the grid marker.

[0051] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working process of the system and modules described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0052] Figure 5 This is a schematic diagram of an electronic device according to an embodiment of the present invention. The electronic device is used for workplace / residence identification. Figure 5As shown, the electronic device 50 of this embodiment includes: a memory 510, a processor 520, and a system bus 530. The memory 510 includes an executable program 5101 stored thereon. As those skilled in the art will understand, Figure 5 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0053] The following is combined Figure 5 A detailed description of each component of the electronic device 50 is provided below: The memory 510 can be used to store software programs and modules. The processor 520 executes various functional applications and data processing of the electronic device by running the software programs and modules stored in the memory 510. The memory 510 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device (such as cached data), etc. In addition, the memory 510 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0054] The memory 510 contains an executable program 5101 for signaling processing methods. The executable program 5101 can be divided into one or more modules / units, which are stored in the memory 510 and executed by the processor 520 to perform functions such as job / residence identification. Each module / unit can be a series of computer program instruction segments capable of performing a specific function, describing the execution process of the executable program 5101 in the electronic device 50. For example, the executable program 5101 can be divided into functional modules such as a time period segmentation module, a grid centroid calculation module, a candidate area identification module, and a job / residence determination module.

[0055] The processor 520 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 510, and by calling data stored in the memory 510, it performs various functions and processes data, thereby monitoring the overall status of the electronic device. Optionally, the processor 520 may include one or more processing units; preferably, the processor 520 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, application programs, etc., and the modem processor mainly handles wireless communication. It is understood that the modem processor may also not be integrated into the processor 520.

[0056] The system bus 530 is used to connect various functional components inside the computer, transmitting data, address, and control information. Its type can be, for example, a PCI bus, an ISA bus, or a CAN bus. Instructions from the processor 520 are transmitted to the memory 510 via the bus, and the memory 510 sends data back to the processor 520. The system bus 530 is responsible for data and instruction exchange between the processor 520 and the memory 510. Of course, the system bus 530 can also connect to other devices, such as network interfaces and display devices.

[0057] In this embodiment of the invention, the executable program executed by the processor 520 included in the electronic device includes: Set a work-residence time window, and divide the daily user mobile phone signaling into daytime dataset and nighttime dataset according to the work-residence time window; Map the coordinates of each base station to a grid of the first preset size, count the number of mobile phone signaling in the daytime and nighttime datasets of each base station in each grid, and calculate the centroid of each grid. The centroid of each grid is used as the center to construct a neighborhood set within a predetermined distance. The number of mobile phone signaling in each neighborhood set is counted, and the neighborhood set with the most signaling is selected as the candidate work-residence area for the day. The coordinates of each base station are mapped to a grid of the second preset size. Based on the number of mobile phone signaling in the candidate work and residence areas and the distribution of monthly mobile phone signaling of users in each grid, the weight value of each grid under the second preset size is calculated by accumulating the distribution weights. The grid weight value is compared with a preset threshold, and the grid of the user's work location and the grid of the user's residence are distinguished based on the comparison results.

[0058] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0059] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0060] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying the workplace and residence based on mobile phone signaling, characterized in that, include: Set a work-residence time window, and divide the daily user mobile phone signaling into daytime dataset and nighttime dataset according to the work-residence time window; Map the coordinates of each base station to a grid of the first preset size, count the number of mobile phone signaling in the daytime and nighttime datasets of each base station in each grid, and calculate the centroid of each grid. The centroid of each grid is used as the center to construct a neighborhood set within a predetermined distance. The number of mobile phone signaling in each neighborhood set is counted, and the neighborhood set with the most signaling is selected as the candidate work-residence area for the day. The coordinates of each base station are mapped to a grid of the second preset size. Based on the number of mobile phone signaling in the candidate work and residence areas and the distribution of monthly mobile phone signaling of users in each grid, the weight value of each grid under the second preset size is calculated by accumulating the distribution weights. The grid weight value is compared with a preset threshold, and the grid of the user's work location and the grid of the user's residence are distinguished based on the comparison results.

2. The method according to claim 1, characterized in that, The second preset size grid is larger than the first preset size grid.

3. The method according to claim 1, characterized in that, The calculation of the centroid of each grid includes: Calculate the centroid of each grid based on the actual location of the base station and the number of mobile signaling records of the base stations within the grid: ； In the formula, The number of user signaling entries within the grid, where p represents the base station coordinates. Indicates the center of gravity of the grid.

4. The method according to claim 1, characterized in that, The step of calculating the weight value of each grid under the second preset size by accumulating the mobile phone signaling data of the candidate work-residence area and the distribution of users' monthly mobile phone signaling in each grid includes: With a total daily weight of 1, the weight of a single signaling record within the grid is represented as... Then, the grid weights are calculated according to the formula: In the formula, Indicates the first Japanese-style residential candidate area cluster Internal signaling record count Indicates the first Weight of a single signaling record per day denoted by , p represents the base station, and g represents the grid.

5. The method according to claim 1, characterized in that, The step of comparing the grid weight value with a preset threshold and distinguishing between the user's work location grid and residential location grid based on the comparison result includes: If the grid weight value is greater than the first preset threshold, it is marked as a residential grid. If the grid weight value is greater than the second preset threshold, it is marked as a work grid. Both the first preset threshold and the second preset threshold are positive integers, and the first preset threshold is greater than the second preset threshold.

6. The method according to claim 1, characterized in that, The method of distinguishing between user work location grids and residential location grids based on comparison results also includes: The weights of base stations in the user's work location grid and residence grid are calculated by weight accumulation, and the base station with the largest weight is selected as the grid marker point.

7. A workplace / residence identification system based on mobile phone signaling, characterized in that, include: The time period segmentation module is used to set the work-residence time window and divide the daily user mobile phone signaling into daytime datasets and nighttime datasets according to the work-residence time window; The grid centroid calculation module is used to map the coordinates of each base station to a grid of the first preset size, count the number of mobile phone signaling in the daytime and nighttime datasets of each base station in each grid, and calculate the centroid of each grid. The candidate area identification module is used to construct a neighborhood set within a predetermined distance range with the centroid of each grid as the center, count the number of mobile phone signals in each neighborhood set, and select the neighborhood set with the most signaling signals as the candidate work-residence area for the day. The job-residence determination module is used to map the coordinates of each base station to a grid of a second preset size. Based on the number of mobile phone signaling in the job-residence candidate area and the distribution of the user's monthly mobile phone signaling in each grid, the weight value of each grid under the second preset size is calculated by accumulating the distribution weights. The grid weight value is compared with a preset threshold, and the user's work location grid and residence grid are distinguished based on the comparison result.

8. The system according to claim 7, characterized in that, The step of comparing the grid weight value with a preset threshold and distinguishing between the user's work location grid and residential location grid based on the comparison result includes: If the grid weight value is greater than the first preset threshold, it is marked as a residential grid. If the grid weight value is greater than the second preset threshold, it is marked as a work grid. Both the first preset threshold and the second preset threshold are positive integers, and the first preset threshold is greater than the second preset threshold.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the job identification method based on mobile phone signaling as described in any one of claims 1 to 6.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed, it implements the steps of a job / residence identification method based on mobile phone signaling as described in any one of claims 1 to 6.

Images