Method and apparatus for extracting work and residence locations based on signaling data

Abstract

The application provides a job and residence place extraction method and device based on signaling data. The method comprises the following steps: dividing a user work time period and a user home time period based on user signaling data; screening a first work place and a first residence place according to user stay point data and stay point stay duration data in the user work time period and in the user home time period; and performing distribution inspection on the first work place and the first residence place to obtain a final work place and a final residence place. The job and residence place extraction method based on signaling data provided by the application embodiment effectively guarantees the accuracy of extracting the user job and residence place, reduces the complexity of data processing, and makes the job and residence place extraction method and device based on signaling data provided by the application embodiment be widely applied.

Description

Technical Field

[0001] This application relates to the field of data analysis and processing technology, specifically to a method, apparatus, electronic device, and computer program product for extracting workplace and residence information based on signaling data. Background Technology

[0002] User work-residence location extraction, or work-residence location analysis, involves identifying a user's workplace and residence. Accurately obtaining user work-residence location information can be beneficial in various fields such as epidemic prevention and control, targeted marketing, and urban planning.

[0003] Some existing methods for extracting users' work and residence locations rely on clustering analysis of each user's location data within a fixed time period based on their mobile phone signaling data. This involves a huge amount of computation, making it unsuitable for big data analysis. Moreover, over long periods, users may not only be at work or at home, but may also frequently visit places of interest for entertainment. This introduces a lot of noise into the data and leads to poor clustering results, making the extracted user work and residence locations inaccurate. Summary of the Invention

[0004] This application provides a method for extracting the work and residence address based on signaling data, in order to solve the technical problem that the extraction of the user's work and residence address is not accurate enough.

[0005] In a first aspect, embodiments of this application provide a method for extracting the workplace and residence location based on signaling data, including:

[0006] Based on user signaling data, user work time periods and user home time periods are divided;

[0007] Based on the user's dwell time and dwell duration data during the user's work and home periods, the first workplace and the first residence are selected.

[0008] A distribution test is performed on the first workplace and the first residence to obtain the final workplace and the final residence.

[0009] In one embodiment, the step of dividing the user's work time period and the user's home time period based on user signaling data includes:

[0010] Based on user signaling data, the initial user work and home time periods are preset;

[0011] By adding or deleting time periods for the initial user's work and home time periods using a random algorithm, the changed user's work and home time periods are obtained.

[0012] Based on the user dwell time data and dwell time data during the changing user's work and home time periods, the time points for segmentation are filtered out. The changing user's work and home time periods are then segmented using the time points to obtain the final user's work time period and final user's home time period.

[0013] In one embodiment, the step of filtering out segmentation time points based on user dwell time point data and dwell time duration data within the changing user's work and home time periods, and using the segmentation time points to divide the changing user's work and home time periods to obtain the final user's work time period and final user's home time period includes:

[0014] Based on the user dwell time data and dwell time data during the changing user's work and home time periods, several candidate time points that meet the first preset dwell conditions are selected.

[0015] Among several candidate time points, the candidate time points that meet the second preset stay conditions are selected as the final time points.

[0016] The work and home time periods of the changing users are divided using the final division time point to obtain the final user's work time period and the final user's home time period.

[0017] In one embodiment, the first preset stay condition is that the stay duration at the stop point exceeds one hour; the second preset stay condition is that the stay duration at the stop point is the minimum.

[0018] In one embodiment, the step of filtering to obtain the first workplace and the first residence based on user dwell time data and dwell time data during the user's work hours and home hours includes:

[0019] Based on the user's dwell time data and dwell time data during the user's work hours and home hours in a day, filter out a number of work dwell time points and a number of home dwell time points that meet the third preset dwell conditions during the user's work hours and home hours in a day.

[0020] By merging several work stops and several home stops of a user during the user's work hours and home hours over multiple days, several merged work stops and several merged home stops are obtained.

[0021] Based on the dwell time and number of stay data of several combined work stay points and several combined home stay points, the first workplace and the first residence are selected.

[0022] In one embodiment, the step of filtering out the first workplace and the first residence based on the dwell time and dwell frequency data of the plurality of merged work stops and the plurality of merged home stops specifically involves:

[0023] The duration and number of stays at several of the combined work stops and several of the combined home stops are weighted and summed to obtain a score. The first workplace and the first residence are then selected based on the score results.

[0024] In one embodiment, performing a distribution test on the first workplace and the first residence to obtain the final workplace and the final residence includes:

[0025] Based on the rating results of the combined work stop and the combined home stop, the rating difference values ​​between the combined work stop and the combined home stop are obtained;

[0026] When the score difference between several combined work stops and the score difference between several combined home stops are both greater than a preset threshold, the highest score among several combined work stops is taken as the first work location, and the highest score among several combined home stops is taken as the first residence location.

[0027] When at least one of the rating differences between the combined work stop points and the rating differences between the combined home stop points is less than or equal to a preset threshold, the users corresponding to the combined work stop points and / or the combined home stop points are marked as abnormal users.

[0028] Secondly, embodiments of this application provide a work-residence location extraction device based on signaling data, comprising:

[0029] The time period segmentation module is used to: divide users' work time periods and users' home time periods based on user signaling data;

[0030] The first workplace and first residence filtering module is used to: filter out the first workplace and first residence based on the user's dwell time data and dwell time data during the user's working time and the user's home time.

[0031] The module for obtaining the final workplace and final residence is used to: perform a distribution test on the first workplace and the first residence to obtain the final workplace and final residence.

[0032] Thirdly, embodiments of this application provide an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the program to implement the steps of the workplace and residence extraction method based on signaling data described in the first aspect.

[0033] Fourthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the workplace and residence extraction method based on signaling data described in the first aspect.

[0034] The work-residence extraction method and apparatus based on signaling data provided in this application can divide different users into working time periods and home time periods. Then, based on the user's dwell time and dwell time data during the user's working time and home time periods, the first workplace and first residence are first filtered out. Then, the distribution test is performed on the first workplace and first residence to obtain the final workplace and final residence. This effectively ensures the accuracy of extracting the user's work-residence, while reducing the complexity of data processing, making the work-residence extraction method and apparatus based on signaling data provided in this application widely applicable. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a flowchart illustrating the method for extracting work and residence locations based on signaling data provided in an embodiment of this application;

[0037] Figure 2 This is a schematic diagram of the structure of the workplace and residence extraction device based on signaling data provided in the embodiments of this application;

[0038] Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0040] Figure 1 This is a flowchart illustrating a method for extracting the workplace and residence location based on signaling data, as provided in an embodiment of this application.

[0041] Reference Figure 1 This application provides a method for extracting the workplace and residence location based on signaling data, which may include:

[0042] S110. Based on user signaling data, divide users' working hours and home hours;

[0043] S120. Based on the user's dwell point data and dwell time data during the user's working hours and home hours, filter to obtain the first workplace and the first residence.

[0044] S130. Perform a distribution test on the first workplace and the first residence to obtain the final workplace and the final residence.

[0045] It should be noted that user signaling data, or mobile signaling data, is data captured and recorded by the operator's communication base stations when a mobile phone user makes a call, sends a text message, or moves their location. This data includes the user's current latitude and longitude, the time of accessing or leaving the base station, and other information. It is characterized by its high real-time performance, high accuracy, and wide coverage. Therefore, extracting a user's work and residence location through signaling data is an efficient and feasible method.

[0046] It should be noted that the execution entity of the method for extracting workplace and residence based on signaling data provided in this application embodiment can be a terminal-side device, such as a data processor, etc.

[0047] In step S110, the terminal device will divide the user's working time period and the user's home time period based on the user signaling data.

[0048] It should be noted that user signaling data can be obtained from the communication base stations that the user passes through based on the user's trajectory; alternatively, user signaling data can also be obtained from a database used to store user signaling data.

[0049] Since users have various types of work, such as some users having fixed working hours from 9 to 5, while others may have irregular working hours, and different users may also have different lifestyles, this application provides a method for extracting work and residence based on signaling data. It can divide users' working time periods and home time periods according to user signaling data, which is more humane and can improve the accuracy of subsequent extraction of users' work and residence.

[0050] In step S120, the terminal device will filter out the first workplace and the first residence based on the user's dwell time data and dwell time data during the user's work time and home time.

[0051] It should be noted that user dwell time data may include information such as the number of user dwell times, the location of user dwell times, and the number of times user dwell times.

[0052] Regarding the selection of the first workplace and the first residence, for example, the length of time a user stays at a location can be used to select the user's first workplace and the first residence during the user's working hours and the user's home hours, respectively. Alternatively, for example, the number of times a user stays at a location can be combined with the length of time a user stays at a location. Combining different data can result in a more accurate selection of the first workplace and the first residence.

[0053] In step S130, the terminal device performs a distribution check on the first workplace and the first residence to obtain the final workplace and the final residence.

[0054] The first workplace and first residence are selected from several stops along the user's trajectory, and their locational relationship should satisfy a certain distribution. It should be noted that the distribution could be a chi-square distribution, a T-distribution, an F-distribution, etc., and different testing methods can be used for different distributions, such as the chi-square test, the T-test, and the F-test. In this way, the accuracy of the final workplace and final residence obtained after distribution testing can be effectively guaranteed.

[0055] The work-residence extraction method based on signaling data provided in this application can divide different users into working time periods and home time periods. Then, based on the user's dwell time and dwell time data during the user's working time and home time periods, the first workplace and first residence are first filtered out. Then, the distribution test is performed on the first workplace and first residence to obtain the final workplace and final residence. This effectively ensures the accuracy of extracting the user's work-residence, while reducing the complexity of data processing, making the work-residence extraction method based on signaling data provided in this application widely applicable.

[0056] In one embodiment, step S110 may include:

[0057] Based on user signaling data, the initial user work and home time periods are preset;

[0058] By adding or deleting time periods for the initial user's work and home time periods using a random algorithm, the changed user's work and home time periods are obtained.

[0059] Based on the user dwell time data and dwell time data during the changing user's work and home time periods, the time points for segmentation are filtered out. The changing user's work and home time periods are then segmented using the time points to obtain the final user's work time period and final user's home time period.

[0060] Specifically, the step of filtering out time points based on user dwell time data and dwell time data within the changing user's work and home time periods, and using these time points to divide the changing user's work and home time periods to obtain the final user's work time period and final user's home time period, may include:

[0061] Based on the user dwell time data and dwell time data during the changing user's work and home time periods, several candidate time points that meet the first preset dwell conditions are selected.

[0062] Among several candidate time points, the candidate time points that meet the second preset stay conditions are selected as the final time points.

[0063] The work and home time periods of the changing users are divided using the final division time point to obtain the final user's work time period and the final user's home time period.

[0064] More specifically, the first preset stay condition is a stay duration exceeding one hour; the second preset stay condition is a stay duration of at least one hour. Furthermore, both the first and second preset stay conditions can be set according to actual needs.

[0065] To more clearly describe the detailed process of step S110, the following example is provided.

[0066] Assume the random algorithm performs k = 0 calculations, and the initial user work and home time periods are preset as follows:

[0067] day all =

[0068] {0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24}

[0069] The numbers in the expression represent 24-hour clocks, indicating the time period from 00:00 to 59:59 of that hour. The stall coefficient t = 0 for the random algorithm.

[0070] In this way, the initial user's work and home time periods can be pre-divided into the initial user's work time period and the initial user's home time period.

[0071] For example, let the initial user working time period be:

[0072] day k ={7,8,9,10,11,12,13,14,15,16,17,18},

[0073] The initial user home stay period is:

[0074] night k ={19,20,21,22,23,24,0,1,2,3,4,5,6}.

[0075] Subsequently, a random integer d is generated in the range [7,18] using a random algorithm, and a random integer n is generated in the ranges [19,24] and [0,6].

[0076] If d is in day k If it exists in, then day ′ k =day k Delete d, otherwise day ′ k =day k add d, if n is at night k If it exists in, then night ′ k =night k Delete n, otherwise night ′ k =night k add n.

[0077] Among them, day ′ k =day k `delete d` indicates the initial user working period (day). k Delete time period d to get the new initial user work time period day. ′ k ;day ′ k =day k add d indicates the initial user working time period day k Add a time period d to obtain the new initial user work time period day. ′ k ;night ′ k =night k delete n represents the initial user's home time period, night. kDelete time period n to obtain the new initial user home time period night. ′ k ;night ′ k =night k add n indicates the initial user's home time period, night. k Add a time period n to obtain a new initial user home time period night. ′ k .

[0078] Subsequently, locals were calculated separately. d locals ′ d locals n locals ′ n .

[0079] Among them, locals d Indicates that the user is on day k The number of user dwell times exceeding one hour within a given time period; locals ′ d Indicates that the user is on day ′ k The number of user dwell times exceeding one hour within a given time period; locals n Indicates that the user is at night k The number of user dwell times exceeding one hour within a given time period; locals ′ d Indicates that the user is at night ′ k The number of user stay points that last for more than one hour within a given time period.

[0080] If locals ′ d <locals d Or night ′ k <night k Then day k =day ′ k night k =night ′ k And k = k + 1, t = 0; otherwise, record t = t + 1.

[0081] If t > 24, then output day. k and night k , indicating dayk The time period represented is the end-user's working hours and night. k The time period represented is the end user's home time period; otherwise, random integers d and n are regenerated, and subsequent calculation steps continue until the output condition is met.

[0082] It should be noted that the final length of the end-user's time spent at home is often longer than the end-user's time spent at work. This is reasonable because, excluding commuting and activity time, most users spend more time at home than at work.

[0083] The work-residence extraction method based on signaling data provided in this application can accurately extract the division time points from the changed work-residence time periods obtained by adding or deleting time periods for the initial user's work and home time periods through a random algorithm. This can obtain more accurate final user work time periods and final user home time periods, effectively ensuring the accuracy of subsequent extraction of user work-residence locations.

[0084] In one embodiment, the step of filtering to obtain the first workplace and the first residence based on user dwell time data and dwell time data during the user's work hours and home hours includes:

[0085] Based on the user's dwell time data and dwell time data during the user's work hours and home hours in a day, filter out a number of work dwell time points and a number of home dwell time points that meet the third preset dwell conditions during the user's work hours and home hours in a day.

[0086] By merging several work stops and several home stops of a user during the user's work hours and home hours over multiple days, several merged work stops and several merged home stops are obtained.

[0087] Based on the dwell time and number of stay data of several combined work stay points and several combined home stay points, the first workplace and the first residence are selected.

[0088] Specifically, the third preset stay condition can be that the stay duration at the stop point is greater than a preset threshold.

[0089] Specifically, the step of filtering out the first workplace and the first residence based on the dwell time and number of visits data of the combined work and residence locations can be as follows:

[0090] The duration and number of stays at several of the combined work stops and several of the combined home stops are weighted and summed to obtain a score. The first workplace and the first residence are then selected based on the score results.

[0091] It should be noted that the terminal device can execute step S100 before executing step S120:

[0092] Clean user signaling data.

[0093] Because the coverage of communication base stations may overlap, and when a user is at the edge of two communication base stations, the user's access signal will repeatedly appear between different communication base stations. Therefore, before extracting the first workplace and first residence based on user signaling data, it is preferable to perform user signaling data cleaning.

[0094] Specifically, the process of cleaning user signaling data may include the following steps:

[0095] For each user's daily records, a locallist is generated, which represents the list of daily records for each user.

[0096] The locallist will contain information such as locallist.id, locallist.time, locallist.st, and locallist.et. Among them, locallist.id represents the ID (identity) of the communication base station appearing in the user's signaling data on that day, locallist.time represents the cumulative duration of the user's access to the corresponding communication base station (locallist.id) in hours, locallist.st represents the time point of access to the corresponding communication base station (locallist.id), and locallist.et represents the time point of departure from the corresponding communication base station (locallist.id).

[0097] Generate an empty list Tlocallist and calculate the Euclidean distance between every two communication base stations using latitude and longitude.

[0098] Similarly, Tlocallist will contain information such as Tlocallist.id, Tlocallist.time, Tlocallist.st, and Tlocallist.et.

[0099] If the Euclidean distance of the longitude and latitude between two communication base stations (such as Communication Base Station A and Communication Base Station B) is less than the threshold Ω, and the locallist.time of Communication Base Station A is more than twice that of Communication Base Station B, then Tlocallist.id is equal to the string formed by combining the accuracy and dimension of Communication Base Station A, and the corresponding Tlocallist.time is equal to the locallist.time of Communication Base Station A plus the locallist.time of Communication Base Station B.

[0100] If the Euclidean distance of the longitude and latitude between Communication Base Station A and Communication Base Station B is less than the threshold Ω, and the following conditions are met:

[0101] |locallist.time of Communication Base Station A minus locallist.time of Communication Base Station B| < pmin{locallist.time of Communication Base Station A, locallist.time of Communication Base Station B},

[0102] then Tlocallist.id = the string synthesized by the average value of the longitude and latitude of Communication Base Station A and the longitude and latitude of Communication Base Station B, and the corresponding Tlocallist.time is equal to the locallist.time of Communication Base Station A plus the locallist.time of Communication Base Station B. At the same time, Tlocallist.st is equal to min{locallist.st of Communication Base Station A, locallist.st of Communication Base Station B}, and Tlocallist.et is equal to max{locallist.et of Communication Base Station A, locallist.et of Communication Base Station B}.

[0103] Output the Tlocallist obtained in Step 2 as the cleaned user signaling data.

[0104] It should be noted that cleaning user signaling data mainly involves merging user signaling data from communication base stations. This merging generally takes two forms: First, when a user appears at both adjacent communication base stations A and B, and the duration of their stay at station A is significantly longer than their stay at station B, the user's destination is likely station A. Their visit to station B is likely just a stopover or brief stop. Therefore, the user signaling data from station B will be merged into the user signaling data from station A. Second, when a user appears at both adjacent communication base stations A and B, and the difference in their stay durations at stations A and B is minimal, the user is located between stations A and B. Therefore, the user signaling data at the midpoint of the latitude and longitude of stations A and B, after summing the times spent at both stations, is taken as the cleaned user signaling data.

[0105] However, some users' stopover points may span time ranges. For example, a user may stay at communication base station A from 8:30 to 9:30 and from 19:30 to 22:30, while the user's final working period is day = {9, 10, 11, 14, 15, 16} and the user's final home period is night = {20, 23, 24, 0, 1, 2, 3, 4, 5, 6}. In this case, the user's Tlocallist needs to be calculated as follows:

[0106] Iterate through Tlocallist and read each record.

[0107] If Tlocallist.st and Tlocallist.et are not included in day (or night), then delete the record.

[0108] If only one of `Tlocallist.st` and `Tlocallist.et` is in `day` (or `night`), a new record is created with the same ID as the original record. If `Tlocallist.st` is in `day` (or `night`), the new record's `Tlocallist.st` is the same as the original record's, and the new record's `Tlocallist.et` is equal to the minimum distance of the original record's `Tlocallist.et` from the adjacent boundary of `day` (or `night`). If `Tlocallist.et` is in `day` (or `night`), the new record's `Tlocallist.et` is the same as the original record's, and the new record's `Tlocallist.st` is equal to the maximum distance of the original record's `Tlocallist.st` from the adjacent boundary of `day` (or `night`). Simultaneously, the new record's `Tlocallist.time` is equal to the new record's `Tlocallist.st` minus the new record's `Tlocallist.et`. Then, the original record is deleted, and the process returns.

[0109] If `Tlocallist.st` and `Tlocallist.et` cover a time period of day (or night), a new record is generated. The ID of the new record is the same as the ID of the original record. The `Tlocallist.et` of the new record is equal to the minimum distance of the `Tlocallist.et` of the original record from the adjacent boundary of day (or night), and the `Tlocallist.st` of the new record is equal to the maximum distance of the `Tlocallist.st` of the original record from the adjacent boundary of day (or night). The `Tlocallist.time` of the new record is equal to the `Tlocallist.st` of the new record minus the `Tlocallist.et` of the new record. Then, the original record is deleted, and the process returns.

[0110] Following the steps above, in the example above, the user's working time at communication base station A is Tlocallist.st = 9, Tlocallist.et = 9:30, and Tlocallist.time = 0.5; the user's home time at communication base station A is Tlocallist.st = 20, Tlocallist.et = 20:59, and Tlocallist.time = 1.

[0111] Using cleaned user signaling data can reduce noise in subsequent extraction of user work and residence locations, thereby improving extraction accuracy.

[0112] After completing step S100, to more clearly describe the detailed process of step S120, we will continue to illustrate the execution of S120 using the example of step S100. Specifically, it includes the following steps:

[0113] An empty list Nlocallist is generated for each user. Nlocallist contains information such as Nlocallist.id, Nlocallist.time, Nlocallist.st, and Nlocallist.et. Nlocallist.id represents the user's ID (identity) composed of the latitude and longitude of their stop point; Nlocallist.time represents the duration of the stop point corresponding to Nlocallist.id; and Nlocallist.co represents the number of times that ID appears within the calculation time period.

[0114] Read in the user's Tlocallist for a specific day, and generate a new Tlocallist based on day (or night).

[0115] First, merge identical communication base station IDs in Tlocallist and sum their corresponding Tlocallist.time values. Then, find the communication base station IDs corresponding to the three largest Tlocallist.time values, such as C, D, and E. Check if C, D, and E appear in Nlocallist.id. If C appears in Nlocallist.id, then the corresponding Nlocallist.time is equal to Nlocallist.time plus C's Tlocallist.time, and Nlocallist.co is incremented by 1. If D and E do not appear in Nlocallist.id, then add D and E to Nlocallist.id, and their corresponding Nlocallist.time is equal to D's (or E's) Tlocallist.time, and Nlocallist.co is equal to 0 (step two).

[0116] Read the Tlocallist containing all the required dates, return to the previous step (step two), and repeat until all dates have been calculated.

[0117] Generate a new list Slocallist, where Slocallist.id represents the ID composed of the latitude and longitude of the user's stop point, and Slocallist.score represents the score of the corresponding communication base station ID.

[0118] Slocallist.score can be calculated using the following expression:

[0119] Slocallist.score=α×Nlocallist.co+β×Tlocallist.time (1),

[0120] α and β are adjustment parameters that can be determined based on the number of dates being calculated.

[0121] Filter the top three Slocallist.scores in the Slocallist list, denoted as Slocal. Slocal contains the corresponding Slocallist.id.

[0122] In Slocal, select the id (which is a string composed of latitude and longitude) corresponding to the highest score in Slocallist.score, split it into latitude and longitude, and use it as the user's first place of residence (or first place of work).

[0123] It should be noted that the steps for extracting the first workplace and the first residence are the same. The only difference is that in step two, records of communication base stations that do not fall within the end user's working hours or home hours are removed.

[0124] The work-residence extraction method based on signaling data provided in this application first obtains several work stops and several home stops that meet a third preset stay condition based on user stop point data and stay duration data during the user's work and home periods in a day. Then, it merges several work stops and several home stops during the user's work and home periods over multiple days to obtain several merged work stops and several merged home stops. Finally, it accurately filters out the first workplace and the first residence based on the stay duration data and stay frequency data of the merged work stops and several merged home stops. The calculation process is simple and has low system requirements for running the work-residence extraction method based on signaling data provided in this application, making it suitable for big data analysis and further ensuring the accuracy of user work-residence extraction.

[0125] In one embodiment, performing a distribution test on the first workplace and the first residence to obtain the final workplace and the final residence includes:

[0126] Based on the rating results of the combined work stop and the combined home stop, the rating difference values ​​between the combined work stop and the combined home stop are obtained;

[0127] When the score difference between several combined work stops and the score difference between several combined home stops are both greater than a preset threshold, the highest score among several combined work stops is taken as the first work location, and the highest score among several combined home stops is taken as the first residence location.

[0128] When at least one of the rating differences between the combined work stop points and the rating differences between the combined home stop points is less than or equal to a preset threshold, the users corresponding to the combined work stop points and / or the combined home stop points are marked as abnormal users.

[0129] Specifically, the score difference value can be calculated using the following expression:

[0130]

[0131] Where dscore represents the score difference value, max(Slocal) represents the score of the communication base station with the highest score in Slocal, and sum(Slocal) represents the sum of the scores of the communication base stations in Slocal.

[0132] For users marked as anomalous, steps S120 and S130 can be repeated until the proportion of anomalous users in the total number of users reaches an acceptable range (e.g., less than 10%). Generally, the range of dscore is [1 / 3, 1), and a preset threshold Ф = 1 / 2 can be used to re-examine anomalous users.

[0133] The work-residence extraction method based on signaling data provided in this application embodiment, after obtaining the first work location and the first residence location in step S120, will also perform a distribution test on the first work location and the first residence location in step S130 to obtain the final work location and the final residence location with sufficiently high accuracy, so as to ensure the quality of extracting the user's work-residence location.

[0134] The following describes the workplace / residence extraction device based on signaling data provided in the embodiments of this application. The workplace / residence extraction device based on signaling data described below can be referred to in correspondence with the workplace / residence extraction method based on signaling data described above.

[0135] Figure 2 This is a schematic diagram of a work and residence location extraction device based on signaling data, provided in an embodiment of this application.

[0136] Reference Figure 2 This application provides a device for extracting the workplace and residence location based on signaling data, which may include:

[0137] The time period segmentation module 210 is used to: segment users' work time periods and users' home time periods based on user signaling data;

[0138] The first workplace and first residence filtering module 220 is used to: filter out the first workplace and first residence based on user dwell time data and dwell time data during the user's working hours and home hours;

[0139] The final workplace and final residence module 230 is used to: perform a distribution test on the first workplace and the first residence to obtain the final workplace and final residence.

[0140] In one embodiment, the time period segmentation module 210 includes:

[0141] The initial user work and home time period preset submodule is used to: preset the initial user work and home time periods based on user signaling data;

[0142] The module for obtaining the changed user's work and home time periods is used to: add or delete time periods to the initial user's work and home time periods using a random algorithm to obtain the changed user's work and home time periods;

[0143] The sub-module for obtaining end-user work time period and end-user home time period is used to: filter out division time points based on user dwell point data and dwell time data within the changing user work and home time periods, and use the division time points to divide the changing user work and home time periods to obtain the end-user work time period and end-user home time period.

[0144] In one embodiment, the submodule for obtaining the end-user's working time period and end-user's home time period includes:

[0145] The candidate time point division submodule is used to: filter out several candidate time points that meet the first preset stay conditions based on user stay point data and stay duration data during the changing user's work and home time periods;

[0146] The final time point division sub-module is used to: select the candidate time point that meets the second preset stay condition from several candidate time points as the final time point division;

[0147] The sub-module for obtaining end-user work time period and end-user home time period is used to: divide the changing user's work and home time periods using the final division time point to obtain the end-user work time period and end-user home time period.

[0148] In one embodiment, the first preset stay condition is that the stay duration at the stop point exceeds one hour; the second preset stay condition is that the stay duration at the stop point is the minimum.

[0149] In one embodiment, the first workplace and first residence screening module 220 includes:

[0150] The work stop and home stop filtering submodule is used to: filter out a number of work stop and a number of home stop that meet the third preset stay conditions during the user's work period and home period respectively, based on the user's stop data and stay duration data during the user's work period and home period in a day.

[0151] The submodule that merges work stops and home stops is used to: merge several work stops and several home stops of a user over multiple days during the user's work period and during the user's home period to obtain several merged work stops and several merged home stops.

[0152] The first workplace and first residence submodule is used to: filter and obtain the first workplace and the first residence based on the dwell time and dwell frequency data of the several merged work stay points and the several merged home stay points.

[0153] In one embodiment, the first workplace and first residence submodule is specifically used for:

[0154] The duration and number of stays at several of the combined work stops and several of the combined home stops are weighted and summed to obtain a score. The first workplace and the first residence are then selected based on the score results.

[0155] In one embodiment, the final workplace and final residence determination module 230 includes:

[0156] The rating difference value acquisition submodule is used to: obtain the rating difference value between the several combined work stops and between the several combined home stops based on the rating results of the several combined work stops and the several combined home stops;

[0157] The first rating difference comparison submodule is used to: when the rating difference between several merged work stops and the rating difference between several merged home stops are both greater than a preset threshold, take the highest rating result among several merged work stops as the first work location, and take the highest rating result among several merged home stops as the first residence location.

[0158] The second rating difference comparison submodule is used to: mark the users corresponding to the several combined work stop points and / or the several combined home stop points as abnormal users when at least one of the rating difference values ​​between the several combined work stop points and the several combined home stop points is less than or equal to a preset threshold.

[0159] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3 As shown, the electronic device may include a processor 810, a communication interface 820, a memory 830, and a communication bus 840, wherein the processor 810, the communication interface 820, and the memory 830 communicate with each other via the communication bus 840. The processor 810 can call a computer program in the memory 830 to execute steps of the job location extraction method based on signaling data, such as:

[0160] Based on user signaling data, user work time periods and user home time periods are divided;

[0161] Based on the user's dwell time and dwell duration data during the user's work and home periods, the first workplace and the first residence are selected.

[0162] A distribution test is performed on the first workplace and the first residence to obtain the final workplace and the final residence.

[0163] Furthermore, the logical instructions in the aforementioned memory 830 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0164] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the steps of the workplace and residence extraction method based on signaling data provided in the above embodiments, such as including:

[0165] Based on user signaling data, user work time periods and user home time periods are divided;

[0166] Based on the user's dwell time and dwell duration data during the user's work and home periods, the first workplace and the first residence are selected.

[0167] A distribution test is performed on the first workplace and the first residence to obtain the final workplace and the final residence.

[0168] On the other hand, embodiments of this application also provide a processor-readable storage medium storing a computer program for causing a processor to execute the steps of the signaling data-based workplace and residence extraction method provided in the above embodiments, including, for example:

[0169] Based on user signaling data, user work time periods and user home time periods are divided;

[0170] Based on the user's dwell time and dwell duration data during the user's work and home periods, the first workplace and the first residence are selected.

[0171] A distribution test is performed on the first workplace and the first residence to obtain the final workplace and the final residence.

[0172] The processor-readable storage medium can be any available medium or data storage device that the processor can access, including but not limited to magnetic memory (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical memory (e.g., CD, DVD, BD, HVD), and semiconductor memory (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).

[0173] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0174] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0175] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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 this application.

Claims

1. A method for extracting the place of residence and work based on signaling data, characterized in that, include: Based on user signaling data, user work time periods and user home time periods are divided; Based on the user's dwell time and dwell duration data during the user's work and home periods, the first workplace and the first residence are selected. A distribution test is performed on the first workplace and the first residence to obtain the final workplace and the final residence. The process of dividing users' work time periods and home time periods based on user signaling data includes: Based on user signaling data, the initial user work and home time periods are preset; By adding or deleting time periods for the initial user's work and home time periods using a random algorithm, the changed user's work and home time periods are obtained. Based on the user dwell time data and dwell time data during the changing user's work and home time periods, the division time points are filtered out, and the changing user's work and home time periods are divided using the division time points to obtain the final user's work time period and final user's home time period. The step involves filtering out time points based on user dwell time data and dwell time data within the changing user's work and home time periods, and using these time points to divide the changing user's work and home time periods to obtain the final user's work time period and final user's home time period, including: Based on the user dwell time data and dwell time data during the changing user's work and home time periods, several candidate time points that meet the first preset dwell conditions are selected. Among several candidate time points, the candidate time points that meet the second preset stay conditions are selected as the final time points. The work and home time periods of the changing users are divided using the final division time point to obtain the final user's work time period and the final user's home time period.

2. The method for extracting work and residence locations based on signaling data according to claim 1, characterized in that, The first preset stay condition is that the stay time at the stop point exceeds one hour; the second preset stay condition is that the stay time at the stop point is the minimum.

3. The method for extracting work and residence locations based on signaling data according to any one of claims 1-2, characterized in that, The process of filtering out the first workplace and the first residence based on user dwell time data and dwell time data during the user's work and home periods includes: Based on the user's dwell time data and dwell time data during the user's work hours and home hours in a day, filter out a number of work dwell time points and a number of home dwell time points that meet the third preset dwell conditions during the user's work hours and home hours in a day. By merging several work stops and several home stops of a user during the user's work hours and home hours over multiple days, several merged work stops and several merged home stops are obtained. Based on the dwell time and number of stay data of several combined work stay points and several combined home stay points, the first workplace and the first residence are selected.

4. The method for extracting work and residence locations based on signaling data according to claim 3, characterized in that, The process of filtering out the first workplace and the first residence based on the dwell time and number of visits data of several merged work and residence locations specifically involves: The duration and number of stays at several of the combined work stops and several of the combined home stops are weighted and summed to obtain a score. The first workplace and the first residence are then selected based on the score results.

5. The method for extracting work and residence locations based on signaling data according to claim 4, characterized in that, The step of performing a distribution test on the first workplace and the first residence to obtain the final workplace and final residence includes: Based on the rating results of the combined work stop and the combined home stop, the rating difference values ​​between the combined work stop and the combined home stop are obtained; When the score difference between several combined work stops and the score difference between several combined home stops are both greater than a preset threshold, the highest score among several combined work stops is taken as the first work location, and the highest score among several combined home stops is taken as the first residence location. When at least one of the rating differences between the combined work stop points and the rating differences between the combined home stop points is less than or equal to a preset threshold, the users corresponding to the combined work stop points and / or the combined home stop points are marked as abnormal users.

6. A device for extracting the workplace and residence location based on signaling data, characterized in that, The method for extracting work and residence locations based on signaling data as described in claim 1 includes: The time period segmentation module is used to: divide users' work time periods and users' home time periods based on user signaling data; The first workplace and first residence filtering module is used to: filter out the first workplace and first residence based on the user's dwell time data and dwell time data during the user's working time and the user's home time. The module for obtaining the final workplace and final residence is used to: perform a distribution test on the first workplace and the first residence to obtain the final workplace and final residence.

7. An electronic device comprising a processor and a memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the workplace and residence extraction method based on signaling data as described in any one of claims 1 to 5.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the workplace and residence extraction method based on signaling data as described in any one of claims 1 to 5.

Images