CN116756439B - Address selection method, device, server and computer readable storage medium - Google Patents

Abstract

This application discloses an address selection method, device, server and computer-readable storage medium, which belongs to the field of big data application technology. The method includes: dividing the target area into multiple grid areas; and determining the location of each grid area based on the user residence data and point of interest POI data corresponding to each grid area in the multiple grid areas. Site selection scoring: determine the grid area among the plurality of grid areas whose site selection scores meet the preset scoring conditions as the target grid area, and determine the corresponding target site selection location based on the target grid area. In this application, the target site selection location is determined in the target grid area, which greatly reduces the site selection range and improves the accuracy of the site selection location.

Description

Site selection method, apparatus, server and computer-readable storage medium

Technical field

This application belongs to the field of big data application technology, and specifically relates to an address selection method, device, server and computer-readable storage medium.

Background technique

In the process of selecting outlets for shops, hospitals and other institutions, the location of outlets is usually determined based only on the number of residents in the community, resulting in inaccuracies such as the location being far away from most users and the location being inconvenient to build outlets. Exact question.

Contents of the invention

The purpose of the embodiments of the present application is to provide an address selection method, device, server and computer-readable storage medium, which can solve the problem of inaccurate address selection locations in existing address selection methods.

In a first aspect, embodiments of the present application provide a site selection method, which method includes:

Divide the target area into multiple grid areas;

Based on the user residence data and point of interest POI data corresponding to each grid area in the plurality of grid areas, determine the site selection score of each grid area;

A grid area whose site selection score satisfies a preset scoring condition among the plurality of grid areas is determined as a target grid area, and a corresponding target site selection location is determined based on the target grid area.

Optionally, determining the site selection score of each grid area based on the user residence data and point of interest POI data corresponding to each grid area in the plurality of grid areas includes:

Obtain user residency data corresponding to each grid area in the plurality of grid areas;

Obtain POI data corresponding to each grid area in the plurality of grid areas;

Based on the user residence data and POI data corresponding to each grid area, the site selection score of each grid area is determined.

Optionally, the user residence data includes residence time data and user portrait data; the location selection score of each grid area is determined based on the user residence data and POI data corresponding to each grid area, include:

Based on the residence time data corresponding to each grid area, determine the corresponding first scoring result;

Based on the user portrait data corresponding to each grid area, determine the corresponding second rating result;

Based on the POI data corresponding to each grid area, determine the corresponding third scoring result;

The first scoring result, the second scoring result and the third scoring result corresponding to each grid area are input into the preset scoring model, and the site selection score corresponding to each grid area is output.

Optionally, the obtaining user residency data corresponding to each grid area in the plurality of grid areas includes:

Obtain the base station data corresponding to each grid area;

Based on the corresponding relationship between the base station data and the user, determine the user portrait data corresponding to each grid area, the time information when the user enters the base station, and the time information when the user leaves the base station;

Based on the time information of the user entering the base station and the time information of the user leaving the base station corresponding to each grid area, the residence time data corresponding to each grid area is determined.

Optionally, after determining the site selection score of each grid area and before determining the grid area whose site selection score satisfies the preset scoring conditions among the plurality of grid areas as the target grid area, The method also includes:

Generate a grid score map of the target area based on the site selection score of each grid area;

Determining a grid area whose site selection scores meet preset scoring conditions among the plurality of grid areas as a target grid area includes:

The grid area that satisfies the preset scoring conditions is determined based on the grid score map and is determined as the target grid area.

In a second aspect, embodiments of the present application provide an address selection device. The address selection device includes:

The division module is used to divide the target area into multiple grid areas;

A determination module configured to determine the site selection score of each grid area based on the user residence data and point of interest POI data corresponding to each grid area in the plurality of grid areas;

A site selection module, configured to determine a grid area among the plurality of grid areas whose site selection scores meet preset scoring conditions as a target grid area, and determine a corresponding target site location based on the target grid area. .

Optionally, the determining module includes:

The first acquisition sub-module is used to acquire user residency data corresponding to each grid area in the plurality of grid areas;

The second acquisition sub-module is used to acquire POI data corresponding to each grid area in the plurality of grid areas;

The determination sub-module is used to determine the site selection score of each grid area based on the user residence data and POI data corresponding to each grid area.

Optionally, the user residence data includes residence time data and user portrait data; the determination sub-module includes:

The first scoring unit is used to determine the corresponding first scoring result based on the residence time data corresponding to each grid area;

The second scoring unit is used to determine the corresponding second scoring result based on the user portrait data corresponding to each grid area;

The third scoring unit is used to determine the corresponding third scoring result based on the POI data corresponding to each grid area;

An output unit is used to input the first scoring result, the second scoring result and the third scoring result corresponding to each grid area into the preset scoring model, and output the selection results corresponding to each grid area. Site rating.

Optionally, the first acquisition sub-module includes:

The acquisition unit is used to obtain the base station data corresponding to each grid area;

The first determination unit is configured to determine the user portrait data corresponding to each grid area, the time information of the user entering the base station, and the time information of the user leaving the base station based on the corresponding relationship between the base station data and the user;

The second determination unit is configured to determine the residence time data corresponding to each grid area based on the time information of the user entering the base station and the time information of the user leaving the base station corresponding to each grid area.

Optionally, the device further includes a generation module configured to generate a grid score map of the target area based on the site selection score of each grid area;

The site selection module includes a site selection sub-module, configured to determine a grid area that satisfies preset scoring conditions based on the grid score map as a target grid area.

In a third aspect, embodiments of the present application provide a server. The server includes a processor, a memory, and a program stored on the memory and executable on the processor. The program is executed by the processor. When implementing the steps of the site selection method described in the first aspect.

In a fourth aspect, embodiments of the present application provide a server. The server includes a transceiver and a processor. The processor is configured to perform the steps of the address selection method described in the first aspect.

In a fifth aspect, embodiments of the present application provide a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the address selection as described in the first aspect is implemented. Method steps.

In the embodiment of this application, the target area is divided into multiple smaller grid areas, and the location selection score corresponding to each grid area is determined based on the user residence data and POI data corresponding to each grid area. The grid area whose site score meets the preset scoring conditions is determined as the target grid area, and the target site selection location is determined in the target grid area, which greatly reduces the scope of site selection and improves the accuracy of site selection.

Description of the drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting any creative effort.

Figure 1 is one of the flow diagrams of the site selection method provided by the embodiment of the present application;

Figure 2 is a schematic flowchart 2 of the site selection method provided by the embodiment of the present application;

Figure 3 is a schematic flow chart of scoring by the scoring model in the embodiment shown in Figure 2;

Figure 4 is a schematic structural diagram of the location selection device provided by the embodiment of the present application;

FIG. 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art fall within the scope of protection of this application.

The terms "first", "second", etc. in the description and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It is to be understood that the figures so used are interchangeable under appropriate circumstances so that the embodiments of the present application can be practiced in orders other than those illustrated or described herein, and that "first," "second," etc. are distinguished Objects are usually of one type, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "and/or" in the description and claims indicates at least one of the connected objects, and the character "/" generally indicates that the related objects are in an "or" relationship.

The methods provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings through specific embodiments and application scenarios.

Figure 1 is a schematic flow chart of an address selection method provided by an embodiment of the present application. As shown in Figure 1, the address selection method provided by an embodiment of the present application can be applied to a server and includes the following steps:

Step 101: Divide the target area into multiple grid areas.

Among them, the target area can be one or multiple, it can be an administrative area, or it can be a custom area. Split the target area on the map into multiple closely connected specific graphics. The specific graphics can be triangles or hexagons, and each specific graphics is a grid area.

Step 102: Determine the site selection score of each grid area based on the user residence data and point of interest (POI) data corresponding to each grid area in the plurality of grid areas.

User residence data includes location data, real-name data, residence time data, user portrait data, etc. POI data is a landmark in the geographic information system, which is used to mark the landmark attractions represented by the place (such as protected buildings, square statues), operating institutions from all walks of life (such as gas stations, health clinics, hospitals, hotels, Banks, post offices, etc.), transportation facilities (such as bus stations, subway stations, parking lots, overpasses, etc.). Three types of places can be divided from POI data, namely designated service places (such as hospitals, health clinics, communities), crowd gathering places (such as residential areas, office buildings, schools, supermarkets, etc.), and crowd flow places (such as gas stations). station, highway intersection, bus station, subway station, etc.). Take the location selection of blood donation sites as an example. Different types of places correspond to different blood donation site types. Fixed-point service sites correspond to fixed blood donation sites, crowd gathering places correspond to convenient blood donation sites, and places with mobile crowds correspond to mobile blood donation sites. According to the POI data in the grid area, different types of blood donation sites can be scored, and the first type of score in the grid area can be obtained. According to the user residence data in the grid area, the second type of score in the grid area can be obtained. Comprehensive The first type of score and the second type of score can determine the site selection score of the grid area.

Step 103: Determine a grid area among the plurality of grid areas whose site selection scores meet preset scoring conditions as a target grid area, and determine a corresponding target site location based on the target grid area.

After the site selection score of each grid area is obtained, the grid area whose site selection scores meet the preset scoring conditions among multiple grid areas is determined as the target grid area. Default scoring criteria can be:

The site selection score is greater than the first preset value; or

The highest value in the site selection score corresponding to each grid area; or

The site selection scores of each grid area are sorted in numerical order, and the top n positions are selected, where n is a positive integer, and the value of n can be adjusted according to needs.

After the target grid area is determined, the specific target site location is determined in the target grid area based on factors such as surrounding road conditions, ventilation, and safety.

Through the location selection method of the embodiment of the present application, in the embodiment of the present application, the target area is divided into multiple smaller grid areas, and each network is determined based on the user residence data and POI data corresponding to each grid area. The site selection score corresponding to the grid area is determined, and the grid area whose site selection score meets the preset scoring conditions is determined as the target grid area. The target site selection location is determined in the target grid area, which greatly reduces the scope of site selection and improves the Ensure the accuracy of site selection.

Optionally, determining the site selection score of each grid area based on the user residence data and point of interest POI data corresponding to each grid area in the plurality of grid areas includes:

Obtain user residency data corresponding to each grid area in the plurality of grid areas;

Obtain POI data corresponding to each grid area in the plurality of grid areas;

Based on the user residence data and POI data corresponding to each grid area, the site selection score of each grid area is determined.

Obtain user residence data and POI data for each grid area respectively. User residence data includes location data, real-name data, residence time data, user portrait data, etc. POI data includes coordinates of fixed-point service venues, location of crowd gathering places, etc. Coordinates, coordinates of crowd flow places. Based on the user residence data and POI data of each grid area, the site selection score of each grid area is determined, making the site selection score more scientific and of reference value.

Specifically, the user residence data includes residence time data and user portrait data; and based on the user residence data and POI data corresponding to each grid area, determining the site selection score of each grid area includes: :

Based on the residence time data corresponding to each grid area, determine the corresponding first scoring result;

Based on the user portrait data corresponding to each grid area, determine the corresponding second rating result;

Based on the POI data corresponding to each grid area, determine the corresponding third scoring result;

The first scoring result, the second scoring result and the third scoring result corresponding to each grid area are input into the preset scoring model, and the site selection score corresponding to each grid area is output.

In this embodiment, the user residence data includes residence time data and user portrait data. The residence time data includes the number of days, residence duration and residence period of the user in the corresponding grid area within a certain period of time. and other information. As shown in Table 1, the dwell time data is the first-level data. According to the user’s dwell time data, the second-level attributes such as the resident attributes, natural period attributes, interval period attributes, and dwell time attributes of the grid area can be determined. data. Resident attributes include third-level data such as total number of people, number of residents, number of floating people, number of residents during the day, number of residents at night, etc. Natural period attributes include the number of people in multiple natural periods such as [00, 01), [01, 02) ... [23, 24), which is a variety of third-level data. Interval time period attributes include third-level data such as the number of people in the morning time period, the number of people in the afternoon time period, and the number of people in the evening time period. The residence duration attribute includes the average residence duration and the number of people staying for different durations such as [00, 01), [01, 02) ... [23, 24), etc. It also includes a variety of third-level data.

Table 1

In this embodiment, as shown in Table 2, user portrait data is first-level data, and user portrait data includes second-level data such as gender, age, household registration, and occupation. Gender includes third-level data such as the number of males and the number of females. Age includes third-level data such as the number of young people, the number of middle-aged people, and the number of elderly people. Household registration includes third-level data such as the number of local residents and the number of residents from other places. Occupation includes third-level data such as the number of students, employees of private enterprises, civil servants, and self-employed persons.

Table 2

In this embodiment, as shown in Table 3, POI data is first-level data, and POI data includes second-level data such as fixed-point service places, crowd gathering places, and crowd flow places. Fixed-point service places include third-level data such as the number of hospitals, health clinics, and communities. Crowd gathering places include third-level data such as the number of residential communities, office buildings, and schools. Crowd flow places include the number of bus stations, subway stations, etc. Level 3 data.

table 3

In this embodiment, corresponding coefficients are respectively set for the third-level data, and the product of the value of the third-level data and the corresponding coefficient is the third sub-score corresponding to the third-level data. According to the plurality of third sub-scores, a corresponding second sub-score may be determined. For example, by obtaining the total number of people, the number of residents, the number of floating people, the number of residents during the day, the number of residents at night, and the corresponding coefficients of the total number of people, the number of residents, the number of floating people, the number of residents during the day, and the number of residents at night, the total number can be obtained The third sub-score corresponding to the number of people, fixed number of people, floating number of people, daytime permanent number and nighttime permanent number. The sum of the third sub-scores corresponding to the total number of people, the number of residents, the number of floating people, the number of residents during the day and the number of residents at night is the second sub-score corresponding to the second-level data of the resident attribute. Similarly, multiple second sub-scores such as natural time period attributes, interval time period attributes, residence time attributes, gender, age, household registration, occupation, designated service places, crowd gathering places, and crowd flow places can be determined.

According to the second sub-scores corresponding to the resident attributes, natural period attributes, interval period attributes, and residence duration attributes, the first scoring result corresponding to the grid area can be determined. Optionally, the sum of the second sub-scores corresponding to the resident attribute, the natural period attribute, the interval period attribute, and the residence duration attribute is determined as the first scoring result. Optionally, weighting coefficients are set for second-level data such as resident attributes, natural period attributes, interval period attributes, and residence duration attributes, and the corresponding values of the resident attributes, natural period attributes, interval period attributes, and residence duration attributes are respectively The weighted average of the second sub-score is determined as the first score result.

Similarly, based on the second sub-scores corresponding to gender, age, household registration, and occupation, the second score results corresponding to the grid area can be determined. Based on the second sub-scores corresponding to fixed-point service places, crowd gathering places, and crowd flow places, the third score result corresponding to the grid area can be determined.

It should be noted that in the case of site selection for blood donation sites, when the second self-score corresponding to the POI data in the target grid area is the highest, when the second sub-score corresponding to the designated service location is the highest, in the target grid area Establish a fixed blood donation site within the target grid area; when the second sub-score corresponding to the crowd gathering place in the target grid area is the highest, establish a convenient blood donation site in the target grid area; when the second sub-score corresponding to the crowd flow place in the target grid area When it reaches the highest level, a mobile blood donation station will be established within the target grid area.

In addition, before determining the first scoring result, the second scoring result and the third scoring result, the obtained residence time data, user portrait data and POI data can also be filtered, and the filtered residence time data, user portrait data and POI data can be filtered. Portrait data and POI data are used to determine the first scoring result, the second scoring result and the third scoring result.

Specifically, the screening steps can be:

When there is first data among the plurality of acquired third-level data, the first data is removed and the data amount of the first data is less than the preset data amount; or

Among the multiple third-level data obtained, overlapping data is eliminated. For example, if the sum of the number of men and the number of women is consistent with the total number, the total number of people is removed; or

Dimensionality reduction is carried out through statistics-based dimensionality reduction technology and machine learning-based algorithm technology. Dimensionality reduction technology can use Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA) and factor analysis. Method (Factor Analysis, FA), etc. Machine learning algorithms can use Decision Tree (DT), Random Forest (RF), Gradient Boosting Decision Tree (GBDT), etc.

Determining the site selection score after filtering the dwell time data, user portrait data and POI data can reduce the amount of data that needs to be calculated and improve scoring efficiency.

After determining the first scoring result, the second scoring result and the third scoring result, input the first scoring result, the second scoring result and the third scoring result into the preset scoring model and output the site selection score corresponding to the grid area , among which, the preset scoring model is established using a comprehensive evaluation method. In this embodiment, resident data, user portrait data and POI data are combined to determine the site selection score of the grid area. During the score determination process, the impact of multiple factors on site selection is taken into account, making the site selection score more scientific. More in line with user needs.

Optionally, the obtaining user residency data corresponding to each grid area in the plurality of grid areas includes:

Obtain the base station data corresponding to each grid area;

Based on the corresponding relationship between the base station data and the user, determine the user portrait data corresponding to each grid area, the time information when the user enters the base station, and the time information when the user leaves the base station;

Based on the time information of the user entering the base station and the time information of the user leaving the base station corresponding to each grid area, the residence time data corresponding to each grid area is determined. In this embodiment, the base station is a communication base station, and any base station can be located in only one grid area. Therefore, there is a corresponding relationship between the base station and the grid area as shown in Table 4. The base station data includes: base station number, base station name , base station latitude and longitude and base station address, etc. According to the grid area where the base station is located, the relationship parameters between the base station and the grid area can be obtained, including: grid code, grid boundary coordinates, base station code, base station coordinates, coordinate type, grid ownership, base station ownership, etc. According to the base station number, the mobile phone number associated with the base station can be obtained, as well as the real-name data, location data and portrait data associated with the mobile phone number and the user. Among them, the location data includes date, mobile phone number, province to which the number belongs, province where signaling is reported, district and county where signaling is located, base station code, base station coordinates, time of entering the base station, time of leaving the base station, etc. The real-name data includes ID number, mobile phone number, ID type, name, province of residence, network access time, etc. The portrait data includes ID number, mobile phone number, number of mobile phone numbers, basic attributes (gender, age, education, occupation), value attributes (customer star rating, income level, credit rating), family attributes, interest preferences, etc.

Table 4

From the above data, portrait data such as gender, age, household registration, and occupation, as well as time information on entering the base station and time information on leaving the base station, are obtained.

Based on the time information of the user entering the base station and the time information of leaving the base station every day, the user's residence time data can be determined. For example, those who have stayed for more than 6 hours in a single day for at least 15 days in the past 30 days are identified as permanent people; those who have an average daily stay of less than 1 hour in the past 30 days are identified as mobile people; those who have stayed for more than 6 hours in a single day in the past 30 days are identified as mobile people; Within the day, those who have stayed for more than 6 hours between 8:00 and 18:00 on at least 15 days in a single day are determined as daytime residents; in the past 30 days, those who have stayed between 20:00 and 6:00 the next day for at least 15 days are considered daytime residents. Those who stay for more than 6 hours in a single day are determined to be permanent residents at night.

In this embodiment, the number of residents in the grid area is subdivided to set corresponding scoring coefficients according to the needs of different groups of people, so that the site selection scoring is more reasonable and meets the needs of users.

Optionally, after determining the site selection score of each grid area and before determining the grid area whose site selection score satisfies the preset scoring conditions among the plurality of grid areas as the target grid area, The method also includes:

Generate a grid score map of the target area based on the site selection score of each grid area;

Determining a grid area whose site selection scores meet preset scoring conditions among the plurality of grid areas as a target grid area includes:

The grid area that satisfies the preset scoring conditions is determined based on the grid score map and is determined as the target grid area.

After obtaining the site selection score of each grid area, the site selection score value of each grid area is displayed on the map of the target area to form the grid score map. Optionally, in the grid score map, the gray values of different grid areas are proportional to the numerical size of the site selection score. The target grid area can be intuitively determined based on the gray color of the grid score map, which enhances the user's confidence. Use experience.

FIG. 2 is a schematic flowchart of a site selection method provided by another embodiment of the present application. The site selection method of this embodiment will be described below with reference to FIG. 2 .

When starting to select a site, determine a larger target area and circle all the base station data in the target area. The base station data includes base station number, base station name, base station longitude and latitude, base station address, etc.

In the target area, each mobile phone number corresponds to a base station, and each mobile phone number also corresponds to a user. Therefore, the corresponding relationship between the base station and the user can be established based on the mobile phone number and the base station code. Specifically, the user's dwell time data is collected. The dwell time data includes the time when the user enters the base station and the time when the user leaves the base station. Based on the dwell time data and the correspondence between the base station and the user, it can be calculated that each user is The dwell time and dwell period under a certain base station within a certain period of time is the base station dwell data. In particular, in order to characterize the location and residence information of each user in each natural period, the time of entering or leaving a certain base station in the residence time data needs to be split and processed to ensure that each user is in each natural period. Each time period has a corresponding base station location information. This processing can support dynamic calculation and update of the number of people in the area based on daily or hourly granularity.

Among them, the collection method of the user's residence time data can be to use the mobile phone number as an association and associate the entire network's real-name data including the user's ID card number. For single-card users with one card and one number, the user's location can be determined directly. For multi-card users with one card and multiple numbers, the user's location needs to be aggregated from multiple location data of multiple mobile phone cards. After the user's location is determined, the user's residence time data can be obtained based on the communication relationship between the user and the base station.

Further, the target area is gridded and divided into multiple smaller grid areas. Based on all base station data and multiple grid areas in the target area, the grid where each base station is located can be determined. Create a base station grid mapping table.

According to the base station residency data and the base station grid mapping table, the grid residency data can be determined, that is, within a specific time period, the number of users in the grid and the residence time and residence period of each user in the grid.

According to the relationship between the mobile phone number and the user, not only the user's residence time data can be obtained, but also the user portrait data can be directly obtained. The user portrait data includes gender, age, education, occupation, interest preferences, etc. In order to further enrich the labels related to the grid and site selection, POI data in the target area are collected and summarized according to the distribution of POI types. POI data includes three categories: fixed-point service places, crowd gathering places, and crowd flow places.

Comprehensive grid residence data (user residence time data in the grid), user portrait data, and POI data can accurately reflect the crowd aggregation and mobility rate of the grid, playing a decisive role in the accuracy of site selection. .

For the three first-level data indicators of grid resident data, user portrait data, and POI data, obtain the number of people or the number of landmarks in the sub-data corresponding to each first-level data. The number of people or landmark data corresponding to each sub-data is used as an indicator for scoring the grid. It also supports the combination of various indicators to form derived indicators. The derived indicators are also used to score the grid. In addition, statistical methods or machine learning algorithms that support feature screening can also be used to analyze the correlation between various indicators and complete the extraction of modeling feature indicators. Based on the selected modeling indicators, a comprehensive evaluation method is used to establish a scoring model for grid site selection, output the site selection scores of all grids in the target area, and determine the appropriate location score based on the site selection score and the distance between grids. Site selection location, improve the accuracy of site selection location.

FIG. 3 is a schematic diagram of the process of scoring the grid by the scoring model in the embodiment shown in FIG. 2 . The following describes the scoring process of the grid in detail with reference to FIG. 2 .

The grid scoring indicator data set includes three first-level data indicators such as grid residence data (residency time data), user portrait data, and POI data. Each first-level data indicator includes multiple sub-data indicators. In order to reduce the amount of data and improve site selection efficiency, it is necessary to extract key indicators from the grid scoring indicator data set.

Key indicator extraction includes indicator preprocessing and indicator screening.

Indicator preprocessing generally removes indicators whose missing proportion exceeds a certain threshold, and fills in or corrects other missing or abnormal data.

Indicator screening. When conducting site selection and scoring on the grid, there are a large number of sub-data indicators, and there may be multiple collinear or mutually exclusive relationships between some sub-data indicators. All indicators need to be denoised and dimensionally reduced. . For example, the total number of people corresponding to the gender indicator is equal to the total number of people corresponding to the age indicator. In the gender indicator, the number of people corresponding to the male indicator and the number of female indicators are mutually exclusive. Dimensionality reduction can be carried out through statistics-based dimensionality reduction technology or machine learning-based algorithm technology. Dimensionality reduction technology can use Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA) and Factor Analysis (FA), etc. Machine learning algorithms can use Decision Tree (DT), Random Forest (RF), Gradient Boosting Decision Tree (GBDT), etc.

Through the above key indicator extraction steps, from the three first-level data indicators such as dwell time data, user portrait data, and POI data, the resident attributes, natural period attributes, interval period attributes, dwell time attributes, and gender are determined respectively. , age, household registration, occupation, designated service places, crowd gathering places, crowd flow places and other second-level data indicators. Each second-level data indicator is scored, and the final grid site selection score is obtained based on multiple second-level data indicator scores.

For residence time data and user portrait data, the total number of people, average residence time, number of people of different genders, and number of people of different age groups for each date and time period can be counted according to dimensions such as date, residence period, etc. The number of people in different occupations, etc. Based on a specified period, usually one month or 30 days, indicators such as the number of days, periods, and length of stay of users in the grid during this period can be calculated, and then the resident population in the grid can be identified. , floating population, night-time resident population, daytime resident population and other refined indicators. The statistical caliber of each indicator can be adjusted as needed. Optionally, if you slide the window according to a fixed time, such as 1 day, and continuously take the residence time data and user portrait data of the grid for a period of time, you can establish a time series set of the number of users, and then analyze the data of different categories. The number of users is predicted, and the number of people is dynamically updated according to the sliding window. For example, the autoregressive model (AR) model, moving average (MA) model, autoregressive moving average model (ARMA) model that supports univariate prediction, and the long short-term memory network that supports multivariable prediction ( Long Short Term Memory (LSTM) model.

Calculate the score value corresponding to each indicator according to the specified rules. At the same time, set a reasonable weight based on the impact of each indicator on the site selection. Use the comprehensive evaluation method to perform a weighted sum calculation of the score value and weight of each indicator. , establish a site selection evaluation model.

Among them, the calculation method of the score value of each indicator can use the ranking method, that is, sort the indicator in positive or negative order according to the impact of the indicator on the site selection decision, and use its sorting value as its factor score value, or you can use a certain statistic Values, such as median, mean, mode, etc., are used as the score value of the indicator. The ranking method can eliminate the dimensional differences between different evaluation factors, and the statistical value can eliminate the interference of extreme values. To set the weight of each indicator, the factor analysis method can be used to calculate the contribution rate of each indicator, and its contribution rate value is used as its weight.

Based on the fitting results of the above scoring model, the index scoring algorithm and the weight determination algorithm are adjusted to gradually optimize the model effect, and finally output the final scoring value of each grid. Usually the grids with the highest scores are used as the final alternative locations, providing Make decisions for the demand department.

As needed, the distance between alternative grids can be further calculated, and some site selection points can be appropriately adjusted to suit each location based on the road conditions around the grid and other site selection factors, such as open space, ventilation, flatness, safety and other factors. The opening periods and opening hours of the site selection points can meet the differentiated needs of various groups of people to the greatest extent.

The execution subject of the location selection method provided by the embodiment of the present application may be an location selection device. Taking the location selection device executing the location selection method as an example, the location selection device 200 provided by the embodiment of the present application will be described with reference to Figure 4. The location selection device 200 includes :

The dividing module 201 is used to divide the target area into multiple grid areas;

The determination module 202 is configured to determine the site selection score of each grid area based on the user residence data and point of interest POI data corresponding to each grid area in the plurality of grid areas;

The site selection module 203 is configured to determine a grid area among the plurality of grid areas whose site selection scores meet preset scoring conditions as a target grid area, and determine a corresponding target site based on the target grid area. Location.

Optionally, the determining module 202 includes:

The first acquisition sub-module is used to acquire user residency data corresponding to each grid area in the plurality of grid areas;

The second acquisition sub-module is used to acquire POI data corresponding to each grid area in the plurality of grid areas;

The determination sub-module is used to determine the site selection score of each grid area based on the user residence data and POI data corresponding to each grid area.

Optionally, the user residence data includes residence time data and user portrait data; the determination sub-module includes:

The first scoring unit is used to determine the corresponding first scoring result based on the residence time data corresponding to each grid area;

The second scoring unit is used to determine the corresponding second scoring result based on the user portrait data corresponding to each grid area;

The third scoring unit is used to determine the corresponding third scoring result based on the POI data corresponding to each grid area;

An output unit is used to input the first scoring result, the second scoring result and the third scoring result corresponding to each grid area into the preset scoring model, and output the selection results corresponding to each grid area. Site rating.

Optionally, the first acquisition sub-module includes:

The acquisition unit is used to obtain the base station data corresponding to each grid area;

The first determination unit is configured to determine the user portrait data corresponding to each grid area, the time information of the user entering the base station, and the time information of the user leaving the base station based on the corresponding relationship between the base station data and the user;

The second determination unit is configured to determine the residence time data corresponding to each grid area based on the time information of the user entering the base station and the time information of the user leaving the base station corresponding to each grid area.

Optionally, the device 200 further includes a generation module configured to generate a grid score map of the target area based on the site selection score of each grid area;

The site selection module includes a site selection sub-module, configured to determine a grid area that satisfies preset scoring conditions based on the grid score map as a target grid area.

The location selection device in the embodiment of the present application divides the target area into multiple smaller grid areas, and determines the location selection score corresponding to each grid area based on the user residence data and POI data corresponding to each grid area. The grid area whose site selection score meets the preset scoring conditions is determined as the target grid area, and the target site selection location is determined in the target grid area, which greatly reduces the scope of site selection and comprehensively considers the user's residence situation, Environmental conditions and other conditions make the target location more able to meet the user's needs.

It should be noted that the address selection device 200 provided by the embodiment of the present application can implement all the technical processes of the above resource processing method and achieve the same technical effect. To avoid duplication, the details will not be described again.

An embodiment of the present application also provides a server, including: a processor, a memory, and a program stored on the memory and executable on the processor. When the program is executed by the processor, the above instruction method is implemented. Each process of the embodiment can achieve the same technical effect, so to avoid repetition, it will not be described again here.

Specifically, as shown in Figure 5, this embodiment of the present application also provides a server, including a bus 301, a transceiver 302, an antenna 303, a bus interface 304, a processor 305 and a memory 306.

In this embodiment, the server also includes a computer program stored on memory 306 and executable on processor 305 . Wherein, when the computer program is executed by the processor 305, the following steps can be implemented:

Divide the target area into multiple grid areas;

Based on the user residence data and point of interest POI data corresponding to each grid area in the plurality of grid areas, determine the site selection score of each grid area;

A grid area whose site selection score satisfies a preset scoring condition among the plurality of grid areas is determined as a target grid area, and a corresponding target site selection location is determined based on the target grid area.

Optionally, when the computer program is executed by the processor 305, it can also implement:

Obtain user residency data corresponding to each grid area in the plurality of grid areas;

Obtain POI data corresponding to each grid area in the plurality of grid areas;

Based on the user residence data and POI data corresponding to each grid area, the site selection score of each grid area is determined.

Optionally, when the computer program is executed by the processor 305, it can also implement:

Based on the residence time data corresponding to each grid area, determine the corresponding first scoring result;

Based on the user portrait data corresponding to each grid area, determine the corresponding second rating result;

Based on the POI data corresponding to each grid area, determine the corresponding third scoring result;

The first scoring result, the second scoring result and the third scoring result corresponding to each grid area are input into the preset scoring model, and the site selection score corresponding to each grid area is output.

Optionally, when the computer program is executed by the processor 305, it can also implement:

Obtain the base station data corresponding to each grid area;

Based on the corresponding relationship between the base station data and the user, determine the user portrait data corresponding to each grid area, the time information when the user enters the base station, and the time information when the user leaves the base station;

Based on the time information of the user entering the base station and the time information of the user leaving the base station corresponding to each grid area, the residence time data corresponding to each grid area is determined.

Optionally, when the computer program is executed by the processor 305, it can also implement:

Generate a grid score map of the target area based on the site selection score of each grid area;

Determining a grid area whose site selection scores meet preset scoring conditions among the plurality of grid areas as a target grid area includes:

The grid area that satisfies the preset scoring conditions is determined based on the grid score map and is determined as the target grid area.

In Figure 3, the bus architecture (represented by bus 301), bus 301 may include any number of interconnected buses and bridges, bus 301 will include one or more processors represented by processor 305 and memory represented by memory 306 various circuits linked together. Bus 301 may also link together various other circuits such as peripherals, voltage regulators, power management circuits, etc., which are all well known in the art and therefore will not be described further herein. Bus interface 304 provides an interface between bus 301 and transceiver 302. Transceiver 302 may be one component or may be multiple components, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. The data processed by the processor 305 is transmitted on the wireless medium through the antenna 303. Furthermore, the antenna 303 also receives the data and transmits the data to the processor 305.

Processor 305 is responsible for managing bus 301 and general processing, and may also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory 306 may be used to store data used by processor 305 when performing operations.

Optionally, the processor 305 may be a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or a complex programmable gate array. Logic device (Complex ProgrammableLogic Device, CPLD).

Embodiments of the present application also provide a computer-readable storage medium. A computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, each process of the above-mentioned repeated transmission control method embodiment is implemented, and the same can be achieved. To avoid repetition, the technical effects will not be repeated here. Wherein, the computer-readable storage medium is such as ROM, RAM, magnetic disk or optical disk.

It should be noted that, in this document, the terms "comprising", "comprises" or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or device that includes a series of elements not only includes those elements, It also includes other elements not expressly listed or inherent in the process, method, article or apparatus. Without further limitation, an element defined by the statement "comprises a..." does not exclude the presence of additional identical elements in a process, method, article or apparatus that includes that element.

Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation. Based on this understanding, the technical solution of the present application can be embodied in the form of a software product in essence or that contributes to the existing technology. The computer software product is stored in a storage medium (such as ROM/RAM, disk, CD), including several instructions to cause a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of this application.

The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above-mentioned specific implementations. The above-mentioned specific implementations are only illustrative and not restrictive. Those of ordinary skill in the art will Inspired by this application, many forms can be made without departing from the purpose of this application and the scope protected by the claims, all of which fall within the protection of this application.

Claims (6)

1. A site selection method, characterized in that the method includes: Divide the target area into multiple grid areas; Based on the user residence data and point of interest POI data corresponding to each grid area in the plurality of grid areas, determine the site selection score of each grid area; Determine a grid area among the plurality of grid areas whose site selection scores meet preset scoring conditions as a target grid area, and determine a corresponding target site location based on the target grid area; Determining the site selection score of each grid area based on the user residence data and point of interest POI data corresponding to each grid area in the plurality of grid areas includes: Obtain user residency data corresponding to each grid area in the plurality of grid areas; Obtain POI data corresponding to each grid area in the plurality of grid areas; Based on the user residence data and POI data corresponding to each grid area, determine the site selection score of each grid area; The user residence data includes residence time data and user portrait data; the location score of each grid area is determined based on the user residence data and POI data corresponding to each grid area, including: Based on the residence time data corresponding to each grid area, determine the corresponding first scoring result; Based on the user portrait data corresponding to each grid area, determine the corresponding second rating result; Based on the POI data corresponding to each grid area, determine the corresponding third scoring result; Input the first scoring result, the second scoring result and the third scoring result corresponding to each grid area into the preset scoring model, and output the site selection score corresponding to each grid area; Before determining the first scoring result, the second scoring result and the third scoring result, the method further includes: analyzing the residence time data corresponding to each grid area, each network Filter the user portrait data corresponding to the grid area and the POI data corresponding to each grid area, wherein the residence time data, the user portrait data and the POI data all include a plurality of second-level data, Each Level 2 data includes multiple Level 3 data; Screening steps include: Remove the first data among the plurality of third-level data included in the dwell time data, user portrait data, and POI data. The data amount of the first data is less than the preset data amount. The dwell time data, user portrait data, POI data includes multiple third-level data; or, Remove overlapping data from multiple third-level data included in dwell time data, user portrait data, and POI data; Based on the residence time data, user portrait data and POI data corresponding to each filtered grid area, determine the first scoring result, the second scoring result and the third scoring result respectively; The obtaining user residency data corresponding to each grid area in the plurality of grid areas includes: Obtain the base station data corresponding to each grid area; Based on the corresponding relationship between the base station data and the user, determine the user portrait data corresponding to each grid area, the time information when the user enters the base station, and the time information when the user leaves the base station; Based on the time information of the user entering the base station and the time information of the user leaving the base station corresponding to each grid area, determine the residence time data corresponding to each grid area; The plurality of second-level data of the POI data include: fixed-point service place data, crowd gathering place data, and crowd flow place data. The fixed-point service place data, crowd gathering place data, and crowd flow place data all correspond to a second sub-section. score; The method also includes: Obtain the target data of the target grid area. Among the fixed-point service place data, crowd gathering place data, and crowd flow place data in the target grid area, the corresponding second sub-scoring data with the highest score is the target data; The location type of the target grid area is determined according to the type of the target data. 2. The method according to claim 1, characterized in that after determining the site selection score of each grid area, the site selection scores in the plurality of grid areas meet the preset scoring conditions. Before determining the grid area as the target grid area, the method further includes: Generate a grid score map of the target area based on the site selection score of each grid area; Determining a grid area whose site selection scores meet preset scoring conditions among the plurality of grid areas as a target grid area includes: The grid area that satisfies the preset scoring conditions is determined based on the grid score map and is determined as the target grid area. 3. An address selection device, characterized in that the device includes: The division module is used to divide the target area into multiple grid areas; A determination module configured to determine the site selection score of each grid area based on the user residence data and point of interest POI data corresponding to each grid area in the plurality of grid areas; A site selection module, configured to determine a grid area among the plurality of grid areas whose site selection scores meet preset scoring conditions as a target grid area, and determine a corresponding target site location based on the target grid area. ; The determination module includes: The first acquisition sub-module is used to acquire user residency data corresponding to each grid area in the plurality of grid areas; The second acquisition sub-module is used to acquire POI data corresponding to each grid area in the plurality of grid areas; The determination sub-module is used to determine the site selection score of each grid area based on the user residence data and POI data corresponding to each grid area; The user residence data includes residence time data and user portrait data; the determination sub-module includes: The first scoring unit is used to determine the corresponding first scoring result based on the residence time data corresponding to each grid area; The second scoring unit is used to determine the corresponding second scoring result based on the user portrait data corresponding to each grid area; The third scoring unit is used to determine the corresponding third scoring result based on the POI data corresponding to each grid area; An output unit is used to input the first scoring result, the second scoring result and the third scoring result corresponding to each grid area into the preset scoring model, and output the selection results corresponding to each grid area. address rating; The determination sub-module is also used to filter the residence time data corresponding to each grid area, the user portrait data corresponding to each grid area, and the POI data corresponding to each grid area, Wherein, the residence time data, the user portrait data and the POI data all include a plurality of second-level data, and each second-level data includes a plurality of third-level data; Screening steps include: Excluding the first data among multiple third-level data included in the dwell time data, user portrait data, and POI data, the dwell time data, user portrait data, and POI data all include multiple third-level data, and the third-level data is The data amount of one data is less than the preset data amount; or, Remove overlapping data from multiple third-level data included in dwell time data, user portrait data, and POI data; The first scoring unit is also used to determine the corresponding first scoring result based on the residence time data corresponding to each grid area after screening; The second scoring unit is also used to determine the corresponding second scoring result based on the filtered user portrait data corresponding to each grid area; The third scoring unit is also used to determine the corresponding third scoring result based on the filtered POI data corresponding to each grid area; The first acquisition sub-module includes: The acquisition unit is used to obtain the base station data corresponding to each grid area; The first determination unit is configured to determine the user portrait data corresponding to each grid area, the time information of the user entering the base station, and the time information of the user leaving the base station based on the corresponding relationship between the base station data and the user; The second determination unit is configured to determine the residence time data corresponding to each grid area based on the time information of the user entering the base station and the time information of the user leaving the base station corresponding to each grid area; The plurality of second-level data of the POI data include: fixed-point service place data, crowd gathering place data, and crowd flow place data. The fixed-point service place data, crowd gathering place data, and crowd flow place data all correspond to a second sub-section. score; The location selection module is also used to: Obtain the target data of the target grid area. Among the fixed-point service place data, crowd gathering place data, and crowd flow place data in the target grid area, the corresponding second sub-scoring data with the highest score is the target data; The location type of the target grid area is determined according to the type of the target data. 4. The device according to claim 3, wherein the device further comprises a generating module for generating a grid score map of the target area based on the site selection score of each grid area; The site selection module includes a site selection sub-module, configured to determine a grid area that satisfies preset scoring conditions based on the grid score map as a target grid area. 5. A server, characterized in that it includes: a processor, a memory, and a program stored on the memory and executable on the processor. When the program is executed by the processor, the program as claimed in claim 1 is implemented. Go to the steps of the site selection method described in any one of 2. 6. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method of any one of claims 1 to 2 is implemented. Steps of site selection method.