CN118433649A - Permanent population identification method based on mobile phone signaling data - Google Patents

Abstract

The invention provides a resident population identification method based on mobile phone signaling data, which has the characteristics that the method comprises the following steps: step S1, acquiring mobile phone signaling data of each user in a continuous K-day target area; step S2, preprocessing the mobile phone signaling data to obtain preprocessed mobile phone signaling data; step S3, for each user, calculating and obtaining the residence of the user according to the preprocessed mobile phone signaling data; s4, marking each user as a stable living user or an unstable living user according to living places and preprocessing mobile phone signaling data; step S5, for each stable living user, calculating to obtain the effective residence days of the stable living user according to the preprocessed mobile phone signaling data and the single-day residence time threshold M; and S6, obtaining the number of resident population according to the effective residence days, the resident threshold N and the stable resident users. In a word, the method can accurately obtain the number of resident population in the target area.

Description

Permanent population identification method based on mobile phone signaling data

Technical Field

The present invention relates to the technical field of mobile phone signaling data analysis and urban planning, and in particular to a method for identifying permanent residents based on mobile phone signaling data.

Background technique

Permanent population refers to the population that has lived in a certain area for more than six months. It not only reflects the population size and population density of the area, but also is an important basis for studying social and economic development and allocating public service facilities. It is also an important supporting data for urban planning and refined urban governance. At present, the main way to obtain the number of permanent residents in a city is through a population census conducted every ten years. However, the cycle of conducting a population census is relatively long, and it is only conducted once every ten years. The speed of data update lags far behind urban development, and cannot provide timely population data for urban planning and urban governance.

With the development of information technology, mobile phone signaling data has been used more and more widely. Mobile phone signaling data has high spatial accuracy, rich spatiotemporal information, and comprehensive sample coverage. It can record user behavior, identify population numbers, and provide a basis for urban planning decisions. However, since it is expensive to purchase and obtain mobile phone signaling data, planners can often only obtain mobile phone signaling data for one month instead of six months or longer, and cannot determine whether a user has lived in the area for more than six months.

Therefore, planners can only use a simple threshold setting method to identify mobile phone users who meet the requirements as permanent residents. This method lacks scientific basis and has a low accuracy rate, making it difficult to rely on mobile phone signaling data to identify permanent residents.

Summary of the invention

The present invention is made to solve the above-mentioned problem, and aims to provide a method for identifying permanent residents based on mobile phone signaling data.

The present invention provides a method for identifying permanent residents based on mobile phone signaling data, which is used to obtain the number of permanent residents in a target area and has the following characteristics: step S1, collecting mobile phone signaling data of each user in the target area for K consecutive days; step S2, preprocessing the mobile phone signaling data to obtain preprocessed mobile phone signaling data; step S3, for each user, calculating the corresponding place of residence of the user according to the corresponding preprocessed mobile phone signaling data; step S4, marking each corresponding user as a stable resident user or an unstable resident user according to the place of residence and the preprocessed mobile phone signaling data; step S5, for each stable resident user, calculating the effective number of days of stay of the stable resident user within K days according to the corresponding preprocessed mobile phone signaling data and a preset single-day residence time threshold M; step S6, taking the total number of all stable resident users whose effective number of days of stay is greater than or equal to the preset permanent threshold N as the number of permanent residents, wherein when the single-day residence time of the stable resident user in the target area is greater than the single-day residence time threshold M, the day is the effective residence day of the stable resident user, and the sum of all effective residence days within K days is the effective residence day.

The method for identifying permanent residents based on mobile phone signaling data provided by the present invention may also have the following features: wherein the mobile phone signaling data includes age and gender, and in step S2, preprocessing includes eliminating mobile phone signaling data with abnormal age or gender.

In the permanent resident population identification method based on mobile phone signaling data provided by the present invention, it can also have the following characteristics: wherein, step S3 includes the following sub-steps: step S3-1, based on the user's pre-processed mobile phone signaling data, obtaining the various activity locations and corresponding stay durations of the user in the target area within a specified time period of each day; step S3-2, accumulating the corresponding stay durations of each activity location within K days to obtain the corresponding total stay duration; step S3-3, taking the activity location with the largest total stay duration as the place of residence.

The method for identifying permanent residents based on mobile phone signaling data provided by the present invention may also have the following feature: wherein the designated time period is from 9:00 p.m. on the first day to 8:00 a.m. on the second day.

The method for identifying permanent residents based on mobile phone signaling data provided by the present invention may also have the following features: wherein, in step S4, it is determined whether the user meets the determination condition. If so, the user is a stable resident user; if not, the user is an unstable resident user. The determination condition is expressed as follows: Where a is the total number of days the user appears in his/her residence within K days, b is the total number of days the user appears in the target area within K days, c is the total length of stay at the user's corresponding residence, d is the total length of stay at all activity locations corresponding to the user, and m and n are both preset thresholds.

The method for identifying permanent residents based on mobile phone signaling data provided by the present invention may also have the following features: wherein m=80, n=50.

The method for identifying permanent residents based on mobile phone signaling data provided by the present invention may also have the following features: wherein, according to the existing population census data corresponding to the target area and the corresponding historical mobile phone signaling data for K consecutive days, a single-day residence time threshold M and a permanent residence threshold N are calculated, comprising the following steps: step T1, setting an initial single-day residence time threshold, a first step length, a maximum time threshold, an initial permanent residence threshold, a second step length and a maximum permanent residence threshold; step T2, pre-processing the historical mobile phone signaling data to obtain pre-processed historical mobile phone signaling data of each historical user; step T3, for each historical user, calculating the historical user according to the corresponding pre-processed historical mobile phone signaling data The residence corresponding to the household; step T4, according to the residence and the pre-processed historical mobile phone signaling data, the corresponding historical users are marked as historical stable resident users or historical unstable resident users; step T5, the initial single-day residence time threshold is used as the debugging single-day residence time threshold, and the initial permanent residence threshold is used as the debugging permanent residence threshold; step T6, for each historical stable resident user, the day on which the single-day residence time of the historical stable resident user in the target area is greater than the debugging single-day residence time threshold is used as the historical effective residence day; step T7, for each historical stable resident user, the total number of all corresponding historical effective residence days is used as the historical effective residence days; step T8, the historical effective residence days The total number of all historical stable resident users greater than or equal to the debugging permanent threshold is taken as the historical permanent population; step T9, the historical permanent population and the census data are subjected to Pearson correlation analysis to obtain the correlation P; step T10, based on the historical permanent population and the census data, the ratio o of the historical permanent population to the total population of the target area in the census data is calculated; step T11, whether the debugging single-day residence time threshold is greater than the maximum time threshold, if so, execute step T12, if not, the sum of the debugging single-day residence time threshold and the first step length is taken as the new debugging single-day residence time threshold, and execute step T6; step T12, determine Whether the debugging permanent threshold is greater than the maximum permanent threshold, if so, execute step T13, if not, use the initial single-day residence time threshold as the debugging single-day residence time threshold, use the sum of the debugging permanent threshold and the second step length as the new debugging permanent threshold, and execute step T6; step T13, determine whether the maximum correlation P is unique, if so, use the maximum correlation P as the optimal correlation P, if not, use the correlation P with the largest ratio o among all the maximum correlations P as the optimal correlation P; step T14, use the debugging single-day residence time threshold corresponding to the optimal correlation P as the single-day residence time threshold M, and use the debugging permanent threshold corresponding to the optimal correlation P as the permanent threshold N.

In the permanent resident population identification method based on mobile phone signaling data provided by the present invention, it can also have the following characteristics: wherein, the initial single-day residence time threshold is 0, the first step length is 3, the maximum time threshold is 12, the initial permanent threshold is 10, the second step length is 1, and the maximum permanent threshold is K.

Functions and Effects of the Invention

According to the method for identifying permanent residents based on mobile phone signaling data involved in the present invention, on the one hand, the residence of each user is calculated through mobile phone signaling data and a designated time period that meets the daily work and rest needs of people, and whether the user is a stable resident user in the target area is determined based on the residence time data of each user in the residence; on the other hand, the single-day residence time threshold M and the permanent threshold N that meet the target area are calculated based on the census data and the corresponding historical mobile phone signaling data, and then the activity data of the stable resident user in the target area is analyzed based on the single-day residence time threshold M and the permanent threshold N, so as to determine whether the stable resident user is the permanent population of the target area. Therefore, the method for identifying permanent residents based on mobile phone signaling data of the present invention can accurately obtain the number of permanent residents in the target area.

BRIEF DESCRIPTION OF THE DRAWINGS

1 is a schematic flow chart of a method for identifying permanent residents based on mobile phone signaling data in an embodiment of the present invention;

2 is a schematic diagram of a process for calculating a place of residence based on pre-processed mobile phone signaling data in an embodiment of the present invention;

FIG3 is a schematic diagram of a flow chart of calculating a single-day residence time threshold M and a permanent residence threshold N in an embodiment of the present invention.

Detailed ways

In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the following embodiments and the accompanying drawings specifically illustrate the method for identifying permanent residents based on mobile phone signaling data of the present invention.

This embodiment provides a method for identifying permanent residents based on mobile phone signaling data, which is used to obtain the number of permanent residents in a target area.

FIG1 is a flow chart of a method for identifying permanent residents based on mobile phone signaling data in an embodiment of the present invention.

As shown in FIG1 , the method for identifying permanent residents based on mobile phone signaling data of this embodiment includes the following steps:

Step S1, collect the mobile phone signaling data of each user in the target area for K consecutive days. The existing mobile phone signaling data procurement unit is month, that is, 28-31 days, and in this embodiment, K=30.

Step S2, preprocessing the mobile phone signaling data to obtain preprocessed mobile phone signaling data.

The mobile phone signaling data includes age and gender. The preprocessing includes removing the mobile phone signaling data with abnormal age or gender. In this embodiment, the abnormal values include missing values and values that do not conform to the normal range.

Step S3: For each user, the residence corresponding to the user is calculated based on the corresponding pre-processed mobile phone signaling data.

FIG. 2 is a schematic diagram of a flow chart of calculating a place of residence based on pre-processed mobile phone signaling data in an embodiment of the present invention.

As shown in FIG. 2 , step S3 includes the following sub-steps:

Step S3-1, based on the user's pre-processed mobile phone signaling data, obtain the various activity locations and corresponding stay durations of the user in the target area during the specified time period of each day. The specified time period is set according to the person's daily rest time. In this embodiment, the specified time period is from 9 pm on the first day to 8 am on the second day. The sum of the user's stay time at activity location A during the time period from 9 pm on the first day to 8 am on the second day is the stay duration of the user at activity location A.

Step S3-2, accumulating the corresponding stay durations at each activity location within K days to obtain the corresponding total stay duration.

Step S3-3, taking the activity location with the largest total stay time as the residence.

Step S4, marking each corresponding user as a stable resident user or an unstable resident user according to the place of residence and the pre-processed mobile phone signaling data.

Among them, in step S4, it is determined whether the user meets the determination conditions. If so, the user is a stable resident user, and if not, the user is an unstable resident user.

The expression of the judgment condition is:

Where a is the total number of days that the user appears in the residence within K days, b is the total number of days that the user appears in the target area within K days, c is the total stay time corresponding to the residence corresponding to the user, d is the total sum of the stay time corresponding to all activity locations corresponding to the user, and m and n are both preset thresholds. In this embodiment, m=80, n=50. In this embodiment, the total stay time c is the total time that the user stays in the residence during each specified time period of K days, that is, from 9 pm on the first day to 8 am on the second day. The total d is the total time that the user stays in each activity location during each specified time period of K days, that is, from 9 pm on the first day to 8 am on the second day.

Step S5, for each stable resident user, the effective stay days of the stable resident user within K days are calculated according to the corresponding pre-processed mobile phone signaling data and the preset single-day residence time threshold M.

When the single-day residence time of a stable resident user in the target area is greater than the single-day residence time threshold M, the day is the effective residence day of the stable resident user, and the sum of all effective residence days within K days is the effective residence days. In this embodiment, the unit of the single-day residence time threshold M is hours. In this embodiment, the single-day residence time is the sum of the residence time of the stable resident user in the target area in one day, that is, the whole day.

Step S6: The total number of all stable resident users whose effective stay days are greater than or equal to the preset permanent threshold N is taken as the permanent population. In this embodiment, the unit of the permanent threshold N is day.

The method for identifying permanent residents based on mobile phone signaling data of this embodiment obtains permanent residents from mobile phone signaling data by screening out permanent residents through the set single-day residence time threshold M and permanent residence threshold N. Therefore, the setting method of the single-day residence time threshold M and the permanent residence threshold N has a great influence on the accuracy of permanent resident identification.

In this embodiment, the single-day residence time threshold M and the permanent residence threshold N are calculated through the existing census data corresponding to the target area and the historical mobile phone signaling data of K consecutive days corresponding to the census data, thereby improving the rationality of setting the single-day residence time threshold M and the permanent residence threshold N.

FIG3 is a schematic diagram of a flow chart of calculating a single-day residence time threshold M and a permanent residence threshold N in an embodiment of the present invention.

As shown in FIG3 , calculating the single-day residence time threshold M and the permanent residence threshold N includes the following steps:

Step T1, setting the initial single-day residence time threshold, the first step length, the maximum time threshold, the initial permanent threshold, the second step length and the maximum permanent threshold.

In this embodiment, the initial single-day residence time threshold is 0, the first step length is 3, the maximum time threshold is 12, the initial permanent threshold is 10, the second step length is 1, and the maximum permanent threshold is K.

Step T2, pre-processing the historical mobile phone signaling data to obtain the pre-processed historical mobile phone signaling data of each historical user.

Step T3: for each historical user, the residence corresponding to the historical user is calculated based on the corresponding pre-processed historical mobile phone signaling data.

Step T4, according to the place of residence and the pre-processed historical mobile phone signaling data, the corresponding historical users are marked as historical stable resident users or historical unstable resident users.

Step T5, using the initial single-day residence time threshold as the debugging single-day residence time threshold, and using the initial permanent threshold as the debugging permanent threshold.

Step T6: for each historical stable resident user, the day on which the single-day stay duration of the historical stable resident user in the target area is greater than the debugged single-day stay time threshold is taken as a historical effective stay day.

Step T7: for each historical stable resident user, the total number of all corresponding historical valid stay days is taken as the historical valid stay days.

Step T8, taking the total number of all historical stable resident users whose historical effective stay days are greater than or equal to the debugged permanent resident threshold as the historical permanent resident population.

Step T9, performing Pearson correlation analysis on the historical permanent population and the census data to obtain the correlation P.

Step T10, based on the historical permanent population and the census data, calculate the ratio o of the historical permanent population to the total population of the target area in the census data.

Step T11, determine whether the debugging single-day residence time threshold is greater than the maximum time threshold. If so, execute step T12. If not, take the sum of the debugging single-day residence time threshold and the first step as the new debugging single-day residence time threshold, and execute step T6.

Step T12, determine whether the debugging permanent threshold is greater than the maximum permanent threshold. If so, execute step T13. If not, use the initial single-day residence time threshold as the debugging single-day residence time threshold, and use the sum of the debugging permanent threshold and the second step length as the new debugging permanent threshold, and execute step T6.

Step T13, determine whether the maximum correlation P is unique, if so, take the maximum correlation P as the optimal correlation P, if not, take the correlation P with the largest ratio o among all the maximum correlations P as the optimal correlation P.

Step T14: the debugging single-day residence time threshold corresponding to the optimal correlation P is used as the single-day residence time threshold M, and the debugging permanent threshold corresponding to the optimal correlation P is used as the permanent threshold N.

In this embodiment, Shanghai is taken as the target area. According to the census data obtained from the survey in Shanghai in November 2020, namely the seventh national census data of Shanghai, and the China Unicom mobile phone signaling data in Shanghai in November 2020 are used as the historical mobile phone signaling data corresponding to the seventh national census data of Shanghai, the value of the single-day residence time threshold M and the permanent residence threshold N are calculated through the above-mentioned method. The value of the single-day residence time threshold M is 0, and the value of the permanent residence threshold N is 22.

In this embodiment, when the value of the single-day residence time threshold M is 0 and the value of the permanent threshold N is 22, the correlation between the permanent population distribution calculated by the permanent population identification method based on mobile phone signaling data in this embodiment and the permanent population distribution obtained by the corresponding census reaches 0.926, which is significantly higher than the correlation of 0.918 obtained by the existing simple threshold setting method. The existing simple threshold setting method identifies mobile phone users who appear in the target area for more than 10 days in a month as permanent residents.

Functions and Effects of the Embodiments

According to the method for identifying permanent residents based on mobile phone signaling data involved in this embodiment, on the one hand, the residence of each user is calculated through mobile phone signaling data and a specified time period that meets the daily needs of people, and whether the user is a stable resident user in the target area is determined based on the residence time data of each user in the residence; on the other hand, the single-day residence time threshold M and the permanent threshold N that meet the target area are calculated based on the census data and the corresponding historical mobile phone signaling data, and then the activity data of the stable resident users in the target area are analyzed based on the single-day residence time threshold M and the permanent threshold N, so as to determine whether the stable resident users are the permanent residents of the target area. In short, this method can accurately obtain the number of permanent residents in the target area.

Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims (8)

1. A method for identifying permanent residents based on mobile phone signaling data, for obtaining the number of permanent residents in a target area, characterized in that it comprises the following steps: Step S1, collecting mobile phone signaling data of each user in the target area for K consecutive days; Step S2, preprocessing the mobile phone signaling data to obtain preprocessed mobile phone signaling data; Step S3, for each of the users, calculating the residence of the user according to the corresponding pre-processed mobile phone signaling data; Step S4, marking each corresponding user as a stable resident user or an unstable resident user according to the residence and the pre-processed mobile phone signaling data; Step S5, for each of the stable resident users, the effective number of stay days of the stable resident user within the K days is calculated according to the corresponding pre-processed mobile phone signaling data and the preset single-day residence time threshold M; Step S6, taking the total number of all the stable resident users whose effective stay days are greater than or equal to the preset permanent resident threshold N as the permanent resident population, When the single-day residence time of the stable resident user in the target area is greater than the single-day residence time threshold M, the current day is the effective residence day of the stable resident user. The sum of all the valid stay days within the K days is the valid stay days. 2. The method for identifying permanent residents based on mobile phone signaling data according to claim 1, characterized in that: Wherein, the mobile phone signaling data includes age and gender, In the step S2, the preprocessing includes eliminating the mobile phone signaling data whose age or gender is an abnormal value. 3. The method for identifying permanent residents based on mobile phone signaling data according to claim 1, characterized in that: Wherein, the step S3 includes the following sub-steps: Step S3-1, obtaining the various activity locations and corresponding stay durations of the user in the target area during a specified time period every day according to the pre-processed mobile phone signaling data of the user; Step S3-2, accumulating the stay durations corresponding to the various activity locations within the K days to obtain the corresponding sum of stay durations; Step S3-3, taking the activity location with the largest total stay time as the residence. 4. The method for identifying permanent residents based on mobile phone signaling data according to claim 3 is characterized in that: The designated time period is from 9:00 p.m. on the first day to 8:00 a.m. on the second day. 5. The method for identifying permanent residents based on mobile phone signaling data according to claim 3 is characterized in that: Wherein, in the step S4, it is determined whether the user meets the determination condition. If so, the user is the stable resident user; if not, the user is the unstable resident user. The expression of the judgment condition is: In the formula, a is the total number of days that the user appears in the residence within the K days, b is the total number of days that the user appears in the target area within the K days, c is the total stay time corresponding to the residence of the user, d is the total sum of the stay time corresponding to all the activity locations corresponding to the user, and m and n are both preset thresholds. 6. The method for identifying permanent residents based on mobile phone signaling data according to claim 5, characterized in that: Among them, m=80, n=50. 7. The method for identifying permanent residents based on mobile phone signaling data according to claim 1, characterized in that: The single-day residence time threshold M and the permanent residence threshold N are calculated based on the existing population census data corresponding to the target area and the corresponding historical mobile phone signaling data for K consecutive days, including the following steps: Step T1, setting the initial single-day residence time threshold, the first step length, the maximum time threshold, the initial permanent threshold, the second step length and the maximum permanent threshold; Step T2, preprocessing the historical mobile phone signaling data to obtain preprocessed historical mobile phone signaling data of each historical user; Step T3, for each of the historical users, calculating the corresponding residence of the historical user according to the corresponding pre-processed historical mobile phone signaling data; Step T4, marking each corresponding historical user as a historical stable resident user or a historical unstable resident user according to the residence and the pre-processed historical mobile phone signaling data; Step T5, using the initial single-day residence time threshold as the debugging single-day residence time threshold, and using the initial permanent threshold as the debugging permanent threshold; Step T6, for each of the historical stable resident users, the day on which the single-day stay time of the historical stable resident user in the target area is greater than the debugged single-day residence time threshold is taken as a historical effective stay day; Step T7, for each of the historical stable resident users, the total number of all corresponding historical valid stay days is taken as the historical valid stay days; Step T8, taking the total number of all the historical stable resident users whose historical effective stay days are greater than or equal to the debugging permanent resident threshold as the historical permanent resident population; Step T9, performing Pearson correlation analysis on the historical resident population and the census data to obtain the correlation P; Step T10, calculating the ratio o of the historical resident population to the total population of the target area in the census data based on the historical resident population and the census data; Step T11, determining whether the debugging single-day residence time threshold is greater than the maximum time threshold, if so, executing step T12, if not, taking the sum of the debugging single-day residence time threshold and the first step as the new debugging single-day residence time threshold, and executing step T6; Step T12, determining whether the debugging permanent threshold is greater than the maximum permanent threshold, if so, executing step T13, if not, taking the initial single-day residence time threshold as the debugging single-day residence time threshold, taking the sum of the debugging permanent threshold and the second step length as the new debugging permanent threshold, and executing step T6; Step T13, determining whether the largest correlation P is unique, if so, taking the largest correlation P as the optimal correlation P, if not, taking the correlation P with the largest ratio o among all the largest correlations P as the optimal correlation P; Step T14: using the debugging single-day residence time threshold corresponding to the optimal correlation P as the single-day residence time threshold M, and using the debugging permanent threshold corresponding to the optimal correlation P as the permanent threshold N. 8. The method for identifying permanent residents based on mobile phone signaling data according to claim 7, characterized in that: The initial single-day residence time threshold is 0. The first step length is 3, The maximum time threshold is 12, The initial resident threshold is 10, The second step length is 1, The maximum resident threshold is K.