CN114529154A

CN114529154A - Method for constructing population scale prediction index system, prediction method, device and system

Info

Publication number: CN114529154A
Application number: CN202210040870.6A
Authority: CN
Inventors: ***; 黄珊珊; 范俞茹; 黄世臻; 郝蓓; 曾涌; 张园林
Original assignee: South China University of Technology SCUT
Current assignee: South China University of Technology SCUT
Priority date: 2022-01-14
Filing date: 2022-01-14
Publication date: 2022-05-24

Abstract

The invention discloses a method, a device and a system for constructing a population scale prediction index system, wherein the prediction method comprises the following steps: acquiring basic data required by predicting the population scale of towns/streets in all years; wherein, the basic data comprises a prediction index in a town/street population scale prediction index system; carrying out time trend extrapolation analysis and population regression analysis on each index in the basic data of the past year to obtain a predicted value of each index; measuring and calculating each index in the historical basic data by adopting an objective weight method to obtain the comprehensive weight of each index; and inputting the predicted values and the comprehensive weights of the indexes into the population scale measuring and calculating model to obtain population scale values serving as population scale prediction results. The method can obtain the prediction results of population scales of different towns/streets, further excavate population scale space distribution characteristics and development trends of various regions, and is favorable for improving the scale prediction efficiency and scientificity of the towns/streets in China.

Description

Method for constructing population scale prediction index system, prediction method, device and system

Technical Field

The invention belongs to the technical field of urban planning and geography, and particularly relates to a method, a device and a system for constructing a population scale prediction index system.

Background

The size of population scale is one of the important characteristics of urban and rural population sites, and the spatial distribution formed by the population scale of each unit in the area clearly reflects the internal characteristics and differences of the area. Since the country is built, due to policy adjustment and regional differences, the population scale of each region of China also undergoes a complex change course and is in dynamic change for a long time.

At present, research mostly focuses on the discussion of macro-scale population scale, but the research and discussion of town/street level population scale with large quantity and wide distribution in China are still not deep enough. Meanwhile, population scale prediction compiled for development in various regions is mostly based on census data, and the data mostly presents the dilemma that urban cities and towns are incomparable, provinces and cities are incomparable for a long time, time is incomparable before and after longitudinally, and statistical calibers are incomparable, so that the problems of unclear population concept, uncertain category, unobjective numerical value and the like exist. The prediction method is limited to single-type index prediction, and the problems of insufficient scientificity and generalization of town/street scale population scale prediction and the like are caused.

In long-term development, the study of town/street population scale is an important issue which is not negligible, population scale prediction is used as a basic link in the planning process of a village and town system, the accuracy of the result directly determines whether the index is matched reasonably, and the study of the evolution characteristics of the county town population scale and the exploration of the future development trend have profound significance. After a transformation period and a transformation development period, more and more people are required to explore the internal development rules of town/street population scale, the method has stronger relevance with the territorial space planning in the aspect of planning transformation, and the rules in the aspect of scale development have better guiding effect on the next planning research in the aspect of transformation development. The objective and scientific prediction method and the intelligent, efficient and systematic system platform have important significance for improving the population scale prediction efficiency of the town/street areas.

Disclosure of Invention

In order to solve the defects of the prior art, the invention provides a method, a device and a system for constructing a population scale prediction index system, which provide a population distribution prediction method of multiple linear regression for town/street population scale prediction and aim to solve the problems that the population prediction index is too single, the statistical aperture of space-time basic data is different and the influence of multiple elements in natural and development conditions on population scale is ignored in the prior art; and a big data technology platform is used for realizing professional collection, analysis and processing of various types of town/street index data and weights in different regions, obtaining prediction results of population scales of different towns/streets, further mining population scale space distribution characteristics and development trends of the regions, contributing to improving scale prediction efficiency and scientificity of the town/street scales in China, and deeply promoting the construction and the improvement of a population scale development system.

The first purpose of the invention is to provide a method for constructing a population scale prediction index system.

A second object of the present invention is to provide a population size prediction method.

A third object of the present invention is to provide a population size prediction device.

A fourth object of the present invention is to provide a population size prediction system.

A fifth object of the present invention is to provide a storage medium.

The first purpose of the invention can be achieved by adopting the following technical scheme:

a method of constructing a population size predictive index system, the method comprising:

based on the relevant research of the population scale influence factors of the urban and rural settlement system, performing criterion layer classification and screening on the factors influencing population scale change to obtain a criterion layer of a prediction index system;

and selecting indexes of the criterion layer of the prediction index system to obtain the index system for predicting the town/street population scale.

Further, the classifying and screening of the criterion layer for the factors affecting the population scale change to obtain the criterion layer of the prediction index system specifically comprises:

classifying factors influencing population scale change into a first-level criterion layer and a second-level criterion layer; wherein:

the first-level criterion layer comprises natural conditions and development conditions;

and screening the natural conditions and the development conditions in the first-level criterion layer, and selecting the land scale of the natural conditions and the population scale and the economic scale in the development conditions as a second-level criterion layer of the prediction index system, namely the criterion layer of the prediction index system.

Further, the index selection is performed on the criterion layer of the prediction index system to obtain the index system for predicting the town/street population scale, which specifically includes:

based on acquireability and scientificity, selecting the construction land area and the cultivated land area as prediction indexes by using a land scale criterion layer;

based on the integrity of population scale in the town/street range, the scientificity of a prediction process and the acquirability of reference data, a population scale criterion layer selects the population scale of the sum of user registration and flow as a prediction index;

based on the acquirability and feasibility, the economic scale criterion layer selects an industrial output value, an agricultural output value, a general public budget and a fixed asset investment amount as prediction indexes.

The second purpose of the invention can be achieved by adopting the following technical scheme:

a method of population size prediction, the method comprising:

acquiring basic data required by predicting the scale of population of towns/streets in all years; wherein, the basic data comprises a prediction index in the index system;

carrying out time trend extrapolation analysis and population regression analysis on each index in the basic data of the past year to obtain a predicted value of each index;

calculating the weight of each index by adopting an objective and subjective comprehensive weight method and utilizing an AHP analytic hierarchy process and a principal component analysis method to obtain the comprehensive weight of each index;

and inputting the predicted value of each index and the comprehensive weight of each index into a population scale measuring and calculating model to obtain a population scale value as a population scale prediction result.

Further, the basic data comprises seven indexes of population scale, agricultural output value of economic scale rule layer, industrial output value, common public budget and fixed asset investment amount, and construction land area and arable land area of land scale rule layer;

the time trend extrapolation analysis and population size regression analysis are carried out on each index in the historical basic data to obtain the predicted value of each index, and the method specifically comprises the following steps:

carrying out time trend extrapolation analysis on each index in the basic data of the past year to obtain a prediction equation;

inputting the predicted years into the prediction equation to obtain the predicted values of all indexes of the predicted years;

performing population scale regression analysis on each index in the historical basic data to obtain a regression equation;

and inputting the agricultural output value, the industrial output value, the general public budget and the fixed asset investment amount index of the economic scale criterion layer and the predicted values of six indexes of the construction land area and the cultivated land area of the land scale criterion layer in the predicted values into the regression equation to obtain the predicted value of the population scale of the predicted year.

Further, the time trend extrapolation analysis is performed on each index in the historical basic data to obtain a prediction equation, and the method specifically includes:

taking seven indexes in the basic data of the calendar year as dependent variables and time as independent variables;

performing curve regression analysis according to the dependent variable and the independent variable, and selecting a prediction model;

supplementing the vacant year data of each dependent variable through the prediction model to obtain complete basic data of each dependent variable in the past year;

taking the complete basic data of each dependent variable of the past year as the dependent variable, and taking the time as the independent variable;

and generating a corresponding prediction equation according to the prediction model.

Further, performing population scale regression analysis on each index in the historical basic data to obtain a regression equation, specifically including:

taking the population scale in the basic data of the past years as a dependent variable, and taking six indexes, namely an agricultural output value, an industrial output value, an index of general public budget and fixed asset investment amount of an economic scale rule layer and an index of construction land area and cultivated land area of a land scale rule layer in the basic data of the past years as independent variables;

and generating a corresponding regression equation according to the prediction model.

Further, performing a curve regression analysis according to the dependent variable and the independent variable, and selecting a prediction model, specifically:

and screening a prediction model which has significant correlation and a certain regression trend of a scatter diagram according to the correlation coefficient and the significance P value of the dependent variable and the independent variable based on the fitting degree and the objectivity as selection bases.

Further, the method for measuring and calculating the weight of each index by using an objective and subjective comprehensive weight method and an AHP analytic hierarchy process and a principal component analysis method to obtain the comprehensive weight of each index specifically includes:

performing factor analysis by using the historical basic data, and extracting component matrix information in the principal component analysis result; the index weight is equal to the weight which takes the variance contribution rate of the principal component as the weight, and the objective weight value of each index is finally obtained by normalizing the weighted average of the coefficients of the index in each principal component linear combination;

aggregating and combining the indexes according to different levels according to the correlation influence and membership among the indexes of the basic data of the past year to form a multi-level analysis structure model, determining the arrangement of relatively important weight or relative quality sequence, and adjusting the result in real time according to the actual situation; thus obtaining the subjective weight value of each index;

and according to different characteristics of different types of towns/streets, integrating the main weight value and the objective weight value of each index to obtain the integrated weight value of each index.

The third purpose of the invention can be achieved by adopting the following technical scheme:

an apparatus for population size prediction, the apparatus comprising:

the historical basic data acquisition module is used for acquiring basic data required by historical town/street population scale prediction; wherein, the basic data comprises the prediction index in the index system;

each index prediction module is used for carrying out time trend extrapolation analysis and population size regression analysis on each index in the basic data of the past year to obtain a prediction value of each index;

each index weight measuring and calculating module is used for measuring and calculating the weight of each index by adopting an objective and subjective comprehensive weight method and utilizing an AHP analytic hierarchy process and a principal component analytic process to obtain the comprehensive weight of each index;

and the population scale prediction module is used for inputting the predicted values of the indexes and the comprehensive weight of the indexes into a population scale measuring and calculating model to obtain population scale values as population scale prediction results.

The fourth purpose of the invention can be achieved by adopting the following technical scheme:

a population size prediction system, the system comprising:

the system display and management function module is used for realizing basic data table management in a database and prediction result data table management;

the scale prediction function module is used for realizing completion of basic data loss and data preprocessing and updating and selecting a fitting equation of a model prediction index;

the prediction result space visualization module is used for realizing rendering of vector data, loading of a map base map, displaying of a basic data chart, setting of model parameters, visualization of a prediction result and an interactive interface between a user and a system;

and the data import and export function module is used for importing data into a database, providing basic data support for model prediction and realizing dynamic display of a prediction result in a map.

The fifth purpose of the invention can be achieved by adopting the following technical scheme:

a storage medium stores a program that, when executed by a processor, implements the population size prediction method described above.

Compared with the prior art, the invention has the following beneficial effects:

1. according to the method, relevant factors influencing the population scale change of the urban and rural settlement system are screened out by layer-by-layer classification, the factors are used as specific analysis indexes for population scale prediction, the internal law of town/street population scale development is excavated, and the method can be used for assisting planning practitioners to carry out town/street population scale measurement and calculation and other planning decisions.

2. According to the method, the subjective and objective weights of the indexes are integrated through different characteristics of different types of towns/streets, so that the weight value of each index which finally influences the population scale of the towns/streets is obtained, and a guide path is provided for improving the accuracy and scientificity of the population scale prediction result; by adopting the multiple linear regression model to predict the town/street population scale, the comprehensive condition of the population scale under the common influence of various factors can be reflected, and a measuring and calculating carrier is provided for the population scale prediction.

3. The system provided by the invention can rapidly analyze the long-term development condition of the town/street through a multi-layer and multi-source regression mathematical model and a multi-dimensional index system of a research area to obtain a relatively objective prediction result, and meanwhile, the system can be used for carrying out multiple visualizations on the population comprehensive scale prediction result, thereby providing effective support for the performance evaluation and optimization guidance of the town/street scale and guiding the development planning of the town/street.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the structures shown in the drawings without creative efforts.

Fig. 1 is a schematic view of a town/street scale prediction index system in example 1 of the present invention.

Fig. 2 is a flowchart of a population size prediction method according to example 2 of the present invention.

Fig. 3 is a schematic diagram of the prediction of the population size of the town/street calendar year in example 2 of the present invention.

Fig. 4 is a block diagram showing a population size prediction system according to embodiment 3 of the present invention.

Fig. 5 is a schematic diagram of a data management module in the population size predicting system according to embodiment 3 of the present invention.

Fig. 6 is a schematic diagram of a scale prediction function module in the population scale prediction system according to embodiment 3 of the present invention.

Fig. 7 is a schematic diagram of a prediction result space visualization module in the population scale prediction system according to embodiment 3 of the present invention.

Fig. 8 is a block diagram showing a population size predicting apparatus according to embodiment 4 of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer and more complete, the technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments, and all other embodiments obtained by a person of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the protection scope of the present invention. It should be understood that the description of the specific embodiments is intended to be illustrative only and is not intended to be limiting.

Example 1:

the embodiment provides a method for constructing a population scale prediction index system, which is based on relevant factors influencing population scale change of an urban and rural settlement system, and comprises natural conditions and development conditions of a first-level criterion layer, wherein the natural conditions comprise two major influence levels of land scale and environmental resources, and the development conditions comprise five major levels of population conditions, location conditions, economic scale, administrative management and service level. Each layer comprises a certain number of specific influence elements, wherein land scale covers a plurality of elements including cultivated land area, construction land area and home base area, environment resources cover natural elements such as landform, natural resources and water area, population conditions cover population elements such as permanent population, household data population and floating population, location conditions cover two elements of traffic location and economic location, economic scale covers elements such as GDP, industrial output value, per capita income and general public budget, administrative management covers elements such as administrative level, government policy and management means, and service level covers elements such as educational medical treatment and cultural and physical facilities. And (3) obtaining a final index system for predicting the town/street population scale by screening the related elements layer by layer, as shown in figure 1.

The method comprises the following steps:

(1) based on the relevant research of the population scale influence factors of the urban and rural settlement system, the factors influencing the population scale change are subjected to criterion layer classification and screening to obtain the criterion layer of the prediction index system.

Factors affecting population size variation are classified into a first level of criteria and a second level of criteria:

(1-1) a first level criterion layer.

The first level of criteria includes natural conditions and developmental conditions.

The urban and rural settlement system is a complex and open regional system, the resident points with different levels are interdependent and closely linked, the urban and rural settlement system has the characteristics of integrity, level and dynamics, the formation, development and evolution of the urban and rural settlement system are influenced by various factors, and the influence modes of different factors are different. Factors influencing the space scale change of the urban and rural settlement system are divided into two categories, namely a first nature and a second nature, namely the first nature and the second nature.

The natural condition part comprises environmental resources and land occupation scale. The ecological environment is a hard constraint of town/street development, the land scale also determines the future development upper limit of the town/street, and the excellent and distinctive ecological environment and the land are core elements influencing the system development of the town/street area.

The development conditions comprise factors such as population scale, regional conditions, economic scale, administrative governance, service level and the like. The current population scale is a basic base number of future population development, directly determines the starting point of population change, and can play a great guiding and correcting role in the standard prediction of villages and towns by extrapolating the situation according to objective historical population data. The location reflects the strength of the accessibility of the colony traffic, and a traffic network which is convenient and developed provides opportunities for population and substance flow, thereby becoming a decisive factor influencing the potential and direction of town/street development. The economic level and the treatment capacity are the core driving force for the development of towns and villages, are the internal driving force for the scale change of the towns and villages, and simultaneously, the perfection degree and the service level of a public service system reflect the development trend of towns/streets population to a certain extent.

And (1-2) a second-level criterion layer.

And screening various secondary criterion layers in the natural conditions and the development conditions of the primary criterion layer through the related research and practice of related scholars on various criterion layers. The environmental resource criterion layer in the natural condition determines the regional moderate population from the resource supply perspective, only represents the ideal population under certain assumed conditions and targets, and has a difference with the actual population scale prediction in the research, so the natural environmental factors in the natural condition level are not included in the index screening range of the town/street scale prediction, and only the land scale of the natural condition is selected as the secondary criterion layer of the prediction index system.

The regional conditions in the development conditions are difficult to significantly change in a short time due to regional traffic or economic locations; there are unstable variations in the administrative levels, government policies and management means in the administrative management criteria layer; most of the infrastructures on the service level are configured in an integrated manner according to the reference caliber of the population of living people borne by the land in an ideal state, so that the configuration of the public service facilities has the characteristics of hysteresis, stage, interval and the like, and the condition of insufficient correlation between the infrastructure and the public service facilities in a research time period can occur, so that the result has large deviation. Based on research, review, analysis and consideration of the three types of criterion layers, the criterion layers are difficult to form rules in population scale prediction, the scale prediction operability of seeking the rules through a specific model is poor, and the change relationship between the criterion layers and the town/street population scale is difficult to comb, so that the regional conditions, the administrative management, the social service street and the like of the development condition layer are not included in the index screening range of the town/future scale prediction, and only the population scale and the economic scale in the development condition are selected as the secondary criterion layers of the prediction index system. The second level of criterion is the criterion layer of the prediction index system.

In conclusion, the factors influencing the population scale change are subjected to criterion layer classification and screening to obtain the criterion layer of the prediction index system, wherein the criterion layer comprises land scale, population scale and economic scale.

(2) And (4) selecting indexes of a criterion layer of the prediction index system to obtain a final index system for predicting the town/street population scale.

And (2-1) selecting the ground scale criterion layer indexes.

The arable land area and the range of arable land reach indirectly influence the change of population scale, and the area of the construction land is also an important index influencing the scale of villages and towns. Therefore, the land scale criterion layer selects the construction land area and the arable land area proportion as reference indexes.

And (2-2) selecting indexes of the population scale criterion layer.

The household population and the regular population are the most obvious indexes for representing the actual population scale of the region, with the development of urbanization, the phenomena of population loss, hollowing and the like occur in a large number of towns/streets, and meanwhile, the increase of the foreign population also enables the floating population to become another key population index. Considering the integrity of population scale within county, the scientificity of the prediction process and the acquirability of reference data, the population scale of the sum of the user membership and the liquidity is uniformly selected as a research object in the town/street scale.

And (2-3) selecting indexes of the economic scale criterion layer.

In the urbanization process, population distribution changes caused by economic growth are the most critical acting forces. The indexes such as GDP, local financial expenditure level, fixed asset investment level, urban and rural resident water storage level and the like can often reflect the most critical information, common public budget is additionally selected as an index according to the statistics condition of each town/street government annual report, four indexes of agricultural output value, industrial output value, fixed asset investment amount and common public budget are selected in total, and the indexes are replaced by service industry output values for towns/streets with part of agricultural output values missing.

Example 2:

as shown in fig. 2, the present embodiment provides a population size prediction method, which includes a collection process of basic data, a regression analysis and a comprehensive weight analysis of each index in the basic data, and a calculation of each index analysis result and a comprehensive weight through a population size calculation model, so as to obtain a final prediction result. The method specifically comprises the following steps:

s201, obtaining basic data required by prediction of the size of the town/street population in the historical years.

The annual data of 2010-plus 2020 in each town/street comes from 'national economy and economy statistics yearbook in China county (city)' and 'Chinese statistics yearbook' in corresponding years, and is appropriately supplemented and improved from 'Chinese county statistics yearbook', 'Guangdong province rural statistics yearbook', 'Guangzhou city statistics yearbook', 'Zengcheng city (district) statistics yearbook', and the like. The development information related to each town/street comes from related economic development data and government work reports in a government information network, and an electronic information atlas, a statistical year, public data of the town government and the like acquired by field reconnaissance of each town/street government department.

TABLE 1 prediction index system for town/street population scale in Zengcheng district

The basic data comprises index data in a population scale criterion layer, an economic scale criterion layer and a land scale criterion layer, wherein the index data in the population scale criterion layer, the economic scale criterion layer and the land scale criterion layer are data such as population scale, agricultural output value, industrial output value, general public budget and fixed asset investment amount index of the economic scale criterion layer, construction land area and arable land area of the land scale criterion layer.

S202, carrying out time trend extrapolation analysis and population size regression analysis on each index in the basic data of the past year to obtain a predicted value of each index.

Further, the step S202 specifically includes:

s2021, carrying out time trend extrapolation analysis on each index in the basic data of the past year to obtain a predicted value of each index.

(1) First, curve fitting was performed based on the collected essential data of each dependent variable in the year 2010-2020, with seven indexes of the population scale, the agricultural output value of the economic scale criterion layer, the industrial output value, the general public budget and the investment amount of the fixed assets, and the construction area and the cultivated area of the land scale criterion layer as dependent variables, and time as an independent variable.

According to the correlation analysis, part of function types with better fitting state at the present stage, such as explosion growth type functions like exponential functions, do not accord with the future scale development trend. Therefore, in the time trend extrapolation model selection of each index, not only the fitting degree at the present stage needs to be considered, but also the objectivity of future development needs to be satisfied. Based on the fitting degree and the objectivity which are taken as selection bases, a function model which has significant correlation and a certain regression trend in a scatter diagram is selected as a prediction model according to correlation coefficients and significance P values of seven different dependent variables and year independent variables. The regression conditions of different dependent variables in different towns/streets and the independent variable of year are different, so that the selected prediction models are also different, but the two principles of good fitting degree and strong objectivity are followed.

(2) And (2) filling the vacant year data of each dependent variable according to the prediction model selected in the step (1) to obtain complete basic data of each dependent variable in the year 2010-2020. Then selecting dependent variable and year in the sps software to perform curve regression analysis, respectively selecting linear regression model, logarithm model, power function model, logistic function model, quadratic function model, etc., according to R²And judging the dependent variable growth trend and selecting a prediction model.

(3) And (3) finally, generating a corresponding prediction equation and a regression curve graph in the sps software through the prediction model selected in the step (2), and substituting the predicted year into the generated prediction equation to obtain the predicted value of the dependent variable of the corresponding year.

And (3) rechecking and comparing the various dependent variable prediction data under the guidance of the prediction model, including population scale prediction values, industrial and agricultural output value prediction values, general public budget and fixed asset investment prediction values, and construction land and arable land area prediction values with respective historical data conditions, and controlling the overall error of the prediction data and the historical data within a reasonable range. And the reasonability and the feasibility of the predicted value of each index layer in the research are verified by comparing the feedback of the error.

And S2022, performing population scale regression analysis on each index in the historical basic data to obtain a predicted value of population scale.

On the basis of time trend extrapolation analysis of an index system, a regression equation of the six indexes and the population scale is constructed according to agricultural output value, industrial output value, general public budget and fixed asset investment amount of an economic scale rule layer, and construction land area and cultivated land area of a land scale rule layer. Curve fitting was performed with population size as the dependent variable and each specific index in the index system described above as the independent variable.

(1) First, selecting2010-2020, performing regression analysis on each specific index data and population scale data, selecting various models such as a linear regression model, a logarithm model, a power function model, a logistic function model, a quadratic function model and the like in the sps software, and performing regression analysis according to R²And selecting a prediction model with optimal fitting degree and objective degree as a prediction model according to the dependent variable growth trend. During function model selection, although the year 2010-2020 fitting degree is good, after calibration is further carried out in consideration of the growth trend of the future population scale and the characteristics of the model, a model which is good in short-term fitting but not in conformity with the actual long-term development situation is omitted, and a relatively ideal prediction result is obtained through other function models. On the other hand, the significance of part of indexes in the specific town/street scale prediction is poor, the weight of the indexes can be reduced by a subjective scoring method, and the influence caused by the fact that the indexes are not significantly related (R2 is more than 0.5) is further reduced. Because different basic data of different factors of different towns have differences, the selected optimal prediction model also has differences.

(2) Next, the numerical value of each index of the predicted year (for example, 2035 years) is obtained by the prediction equation of regression of each index and time in S2021.

(3) And (3) substituting each index value obtained in the step (2) into the selected regression equation to obtain a population scale value of the predicted year.

The actual deviation value of the scale prediction result after inspection is small. Most indexes are well fitted to a multivariate linear regression model, a logarithmic function model, a sigmoid function model and a logistic regression model, and the land indexes are well fitted to the logistic regression model.

And S203, calculating the weight of each index by adopting an objective and subjective comprehensive weight method and using an AHP analytic hierarchy process and a principal component analysis method to obtain the comprehensive weight of each index.

Further, the step S203 specifically includes:

s2031, each index in the basic data is objectively weighted by a principal component analysis method.

In the sps software, factor analysis is carried out on each index volume coefficient value and population scale value in 2010-2020, each index variable and population scale are added into a variable, the extraction method is used as a main component, correlation matrix analysis and a maximum variance method are selected, a score module selects a display factor score coefficient matrix, and after the setting is finished, the determination is clicked to obtain a main component analysis result, wherein the determination result does not depend on subjective determination of people, has strong mathematical theoretical basis and is objective.

In this embodiment, the basic data of the size of the town/street population and each index thereof in 2010-2020 is used for factor analysis, and the component matrix information in the main component analysis result is extracted. And the index weight is equal to the weight which takes the variance contribution rate of the principal component as the weight, and the weight average of the coefficients of the index in each principal component linear combination is normalized to finally obtain the principal component analysis weight value corresponding to each index.

S2032, subjectively empowering each index in the basic data by using an AHP analytic hierarchy process.

The factors are gathered and combined according to different levels according to the mutual correlation influence and the membership relation among the factors to form a multi-level analysis structure model, relatively important weight values or the scheduling of relative quality order are determined, and the result can be adjusted in real time according to the actual situation.

S2033, integrating the two kinds of subjective and objective weights of each index to obtain the integrated weight value of each index.

According to different characteristics of different types of towns/streets, two types of subjective and objective weights of each index are integrated, and the AHP (analytic hierarchy process) weight assignment under the subjective condition of each index and the objective principal component analysis weight assignment are averaged to obtain the integrated weight of each index which finally influences the population scale of the town/street.

And S204, inputting the predicted value of each index and the comprehensive weight of each index into the population scale measuring and calculating model to obtain a population scale value as a population scale prediction result.

In the embodiment, the traditional growth rate method, the related prediction thought of a single element, the environmental resource bearing capacity measurement and the analysis method of multidimensional elements in system mechanics are comprehensively considered, and finally, the multi-linear regression model integrating the characteristics is adopted to predict the town/street population scale, namely, the multi-linear regression model is used as a population scale measuring and calculating model.

The multivariate linear model is a regression analysis method for researching a dependent variable and two or more independent variables, and can reflect the rule that the quantity of one phenomenon or object changes correspondingly according to the change of the quantity of a plurality of phenomena or objects. When the explained variable is influenced by a plurality of influence factors together, in order to better judge the action intensity of different influence factors, a multiple linear regression model is adopted for analysis.

Since the variation of the town/street population scale is the comprehensive result of the multivariate joint influence, the influence degree of different factors on the population scale needs to be adjusted by the differentiated scoring of the weight.

The calculation formula of the multiple linear regression model is as follows:

y＝β₁X_1i+β₂X_2i+…+β_kX_ki i＝(1，2，…，k)

wherein k is the number of independent variables, and Y is a dependent variable, namely the town/street population scale; x is an independent variable, namely the number of predicted population of each index of three standard layers of population, economy and land is represented; beta is a₁、β₂、β₃...β_kThe comprehensive weight of each index obtained by a principal component analysis method and an AHP analytic hierarchy process.

When the regression coefficient beta is a positive value, namely the weight distribution through the correlation method is a positive value, the influence factor has a positive promotion effect on the town/street population aggregation, and the magnitude of the coefficient is used for explaining the effect of the influence factor. The larger the absolute value of the regression coefficient, that is, the larger the weight assignment absolute value by the correlation method, the larger the degree of influence of the explanatory factor on the dependent variable.

According to the multiple linear regression model, a calculation formula of a population scale measurement model is obtained correspondingly as follows:

the method comprises the steps of calculating a population scale of a town/street, wherein the population scale of the town/street is population scale comprehensive weight, population scale self-prediction value, agricultural output value comprehensive weight value, agricultural output value prediction population scale value, industrial output value comprehensive weight value, industrial output value prediction population scale value, general public budget comprehensive weight value, general public budget prediction population scale value, fixed asset investment comprehensive weight value, fixed asset investment prediction population scale value, construction land comprehensive weight value, construction land prediction population scale value and farmland comprehensive weight value.

As shown in fig. 3, the final population size prediction result is obtained by applying each index analysis result and the integrated weight to the population size calculation model.

Example 3:

as shown in fig. 4, the embodiment provides a population size prediction system, which integrates the currently mainstream geographic data spatial technology to realize town/street population size prediction and spatial visualization. The system adopts a C/S architecture, comprises an application layer, a model and algorithm layer and a data layer, and has better openness and structural expansibility. Meanwhile, the system adopts a front-end and back-end low-coupling design, accesses through a data interface, is constructed by adopting an open-source assembly, has higher flexibility, and is specifically explained as follows:

firstly, a system display and management function module.

As shown in fig. 5, the data management module is an interface for uniform reading and writing of files and databases, basic data table management, prediction model data access, and prediction result visualization data access, and mainly implements basic data table management and prediction result data table management in the databases. And a unified data read-write interface is provided, a data access interface is provided for model prediction and data visualization, and the data access interface is used as a basic module of the whole system.

And secondly, a scale prediction function module.

As shown in fig. 6, the module is used to complete the missing of the basic data and preprocess the data by performing sample expansion, function model equation initialization, and model equation optimization on the basic data, and update and select the fitting equation of the model prediction index. And selecting population, land and economic data and various function models in the basic data, grading through time series and subjective and objective assignment, and finally establishing a regression model to calculate the comprehensive population scale of counties (streets).

And thirdly, a prediction result space visualization module.

As shown in fig. 7, the module mainly implements rendering of vector data, map base map loading, basic data chart display, model parameter setting, prediction result visualization, and user-system interaction interface.

And fourthly, importing and exporting the data into and out of the functional module.

The data import function module realizes the given CSV form file, and imports data into a database by reading a header field, wherein the importation comprises the importation of town/street basic data, AHP scoring data and the like, and basic data support is provided for model prediction.

The data export function module mainly realizes dynamic display of the prediction result in the map, including town/street prediction result data display, and supports export forms of GIF and webpage forms.

In addition, in order to provide better data access and data management, collected population data of various types of towns/streets are sorted, screened and verified, a town/street scale structure and a related influence factor database are established and perfected, data related to a county village and town scale structure simulation process are effectively integrated, data support is provided for scale prediction and data display of a system, meanwhile, powerful guarantee is provided for subsequent data management and sharing, and a data service part mainly comprises a database source, a database structure and a table structure.

(1) A database source.

The system adopts a PostgreSQL open-source relation database system with strong functions, spatial data are managed and supported by a PostGIS, and the database system can provide data storage, access and management services for county-area village and town scale structures and relevant influence factor databases.

(2) A database structure.

The county village and town space scale prediction system database mainly comprises basic data, model operation process data and prediction result data, wherein the basic data comprises a town street basic information table (population scale data, economic scale data and land scale data) and a town administrative village development potential index grading table, the model operation process data comprises an index equation fitting table, a town/street weight calculation result table and an administrative village weight calculation result table, and the prediction result data table comprises a town/street prediction result table and an administrative village prediction result table.

(3) And (4) table structure.

The method mainly comprises a town/street weight calculation result table, a town/street prediction result table, wherein:

(3-1) basic information of town street is shown in Table 2:

TABLE 2 basic information table for street town

Name of field	Field information	Type of field
			ZJM	Name of town street	text
CZRK	General population	Int
			YBGGYSSR	General public budget	float8
NYCZ	Agricultural output value	float8
			GSGYCZ	Industrial production value on scale	float8
FWYCZ	Total value of industry	float8
			GSGDZCTZE	Fixed asset investment amount	float8
JZYDMJ	Land for construction	float8
			GDMJ	Cultivation of land	float8
SJNF	Year of year	float8

(3-2) the results of the town/street weight calculation are shown in Table 3:

TABLE 3 weight calculation results table for town/street

Name of field	Field information	Type of field
			ZJM	Name of town street	varchar
AHP1	AHP population weight	float8
			AHP2	AHP agricultural output value population weight	float8
AHP3	AHP industrial production value population weight	float8
			AHP4	AHP public budget population weight	float8
AHP5	AHP fixed asset investment population weight	float8
			AHP6	AHP construction land population weight	float8
AHP7	AHP farmland population weight	float8
			PCA1	PCA Total population weight	float8
PCA2	PCA agricultural output value population weight	float8
			PCA3	PCA Industrial output population weight	float8
PCA4	PCA public budget population weight	float8
			PCA5	PCA fixed asset investment population weight	float8
PCA6	PCA construction land population weight	float8
			PCA7	PCA farmland population weight	float8
Z1	Integrated population weight	float8
			Z2	Integrated agricultural output value population weight	float8
Z3	Integrated industrial value and population weight	float8
			Z4	Integrated public budget population weights	float8
Z5	Synthetic fixed asset investment population weight	float8
			Z6	Population weight of comprehensive construction land	float8
Z7	Population weight of comprehensive cultivated land	float8
			TWO1	Population size criteria layer weights	float8
TWO2	Economic scale criteria layer weights	float8
			TWO3	Land scale criterion layer weight	float8
ONE1	Developing condition criterion layer weights	float8
			ONE2	Natural condition criteria layer weights	float8

(3-3) the results of the town/street prediction are shown in Table 4:

TABLE 4 Town/street prediction results Table (field generated by settings, default 2035 years)

Name of field	Field information	Type of field
			CZM	Name of town street	text
P2021	Population predicted in 2021	Int8
			P2022	Predicted population in 2022	Int8
P2023	Predicted population in 2023 years	Int8
			P2024	Predicted population in 2024 years	Int8
P2025	Population predicted in 2025	Int8
			P2026	Predicted population in 2026 years	Int8
P2027	Predicted population in 2027	Int8
			P2028	Predicted population in 2028	Int8
P2029	Predicted population in 2029 years	Int8
			P2030	Population predicted in 2030	Int8
P2031	Predicted population in 2031 years	Int8
			P2032	Predicted population in 2032 years	Int8
P2033	Predicted population in 2033 years	Int8
			P2034	Predicted population of 2034 years	Int8
P2035	Predicted population in 2035 years	Int8

Because the acquired data structures are different, especially the information of a plurality of original data is numerous and complicated, various data and information need to be cleaned, processed and processed to fully mine the data value.

The system development environment operating system in this embodiment is a Window7 or above version, the software operating environment needs to satisfy 2 cores and above processors, the minimum 2GB memory, and the minimum 20GB hard disk, and the database server hardware needs to satisfy 1 core and above processors, the minimum 1GB memory, and the minimum 50GB hard disk.

Example 4:

as shown in fig. 8, the present embodiment provides a population size prediction apparatus, which includes a calendar year basic data acquisition module 801, an index prediction module 802, an index weight calculation module 803, and a population size prediction module 804, wherein:

a historical basic data acquisition module 801, configured to acquire basic data required by historical town/street population scale prediction; wherein the basic data comprises a prediction index in the index system in the embodiment 1;

each index prediction module 802 is configured to perform time trend extrapolation analysis and population regression analysis on each index in the historical basic data to obtain a prediction value of each index;

each index weight measuring and calculating module 803, which is used for measuring and calculating the weight of each index by using an AHP analytic hierarchy process and a principal component analysis process to obtain the comprehensive weight of each index by adopting an subjective and objective comprehensive weight method for each index in the historical basic data;

and the population size prediction module 804 is used for inputting the predicted values of the indexes and the comprehensive weights of the indexes into a population size measuring and calculating model to obtain population size values as population size prediction results.

For specific implementation of each module in this embodiment, reference may be made to embodiment 1 above, which is not described in detail herein; it should be noted that, the apparatus provided in this embodiment is only illustrated by dividing the functional modules, and in practical applications, the functions may be distributed by different functional modules according to needs, that is, the internal structure is divided into different functional modules to complete all or part of the functions described above.

Example 5:

the present embodiment provides a storage medium, which is a computer-readable storage medium, and stores a computer program, and when the computer program is executed by a processor, the method for predicting the population size of the foregoing embodiment 2 is implemented as follows:

acquiring basic data required by predicting the population scale of towns/streets in all years; wherein the base data comprises predictors of the index system described in example 1;

It should be noted that the computer readable storage medium of the present embodiment may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

In conclusion, the population scale measurement model is screened, the multivariate linear regression model is selected for population scale measurement on the basis of obtaining a town/street population scale criterion layer, an economic scale criterion layer and a land scale criterion layer data, and the result is calibrated by adopting the subjective and objective comprehensive weighting method, so that the method can provide reference for the adjustment of county town/street scale structures under different conditions for planning and functioning departments according to simulated town/street scale structures of county city conditions and conditions of different regions, different types and different development paths; by establishing the town/street population scale prediction system, the multivariate regression mathematical model and the multidimensional index system of the research area can quickly analyze the long-term development condition of the town/street to obtain a relatively objective prediction result, and meanwhile, the comprehensive population scale prediction result is subjected to multiple visualizations by means of the system, so that the prediction dimension and the prediction variable of the prediction model are enriched, and effective support is provided for county and town scale performance evaluation and optimization guidance of counties. The above description is only for the preferred embodiments of the present invention, but the protection scope of the present invention is not limited thereto, and any person skilled in the art can substitute or change the technical solution and the inventive concept of the present invention within the scope of the present invention.

Claims

1. A method for constructing a population size predictive index system, the method comprising:

and selecting indexes of a criterion layer of the prediction index system to obtain the index system for predicting the town/street population scale.

2. The construction method according to claim 1, wherein the classifying and screening of the criterion layer for the factors affecting population scale changes to obtain the criterion layer of the prediction index system specifically comprises:

3. The construction method according to claim 2, wherein the index selection is performed on the criterion layer of the prediction index system to obtain an index system for predicting the town/street population scale, and the method specifically comprises the following steps:

4. A method of population size prediction, the method comprising:

acquiring basic data required by predicting the population scale of towns/streets in all years; wherein the basic data comprises a prediction index in the index system of any one of claims 1 to 3;

and inputting the predicted values of the indexes and the comprehensive weight of the indexes into a population scale measuring and calculating model to obtain population scale values as population scale prediction results.

5. The population size forecasting method according to claim 4, wherein the basic data comprises seven indexes of population size, agricultural output value of economic scale rule layer, industrial output value, general public budget and fixed asset investment amount, and construction land area and arable land area of land scale rule layer;

carrying out population scale regression analysis on each index in the historical basic data to obtain a regression equation;

6. The population size forecasting method according to claim 5, wherein the time trend extrapolation analysis is performed on each index in the historical basic data to obtain a forecasting equation, and the forecasting equation specifically comprises:

taking seven indexes in the basic data of the calendar year as dependent variables, and taking time as independent variables;

7. The population size forecasting method according to claim 5, wherein the population size regression analysis is performed on each index in the historical basic data to obtain a regression equation, and the regression equation specifically comprises:

8. The population size forecasting method according to any one of claims 6 and 7, wherein the curve regression analysis is performed according to the dependent variable and the independent variable to select a forecasting model, specifically:

9. The population size predicting method according to claim 4, wherein the comprehensive weight of each index is obtained by calculating the weight of each index by using an AHP (analytic hierarchy process) and a principal component analysis method by using an objective and subjective comprehensive weight method for each index in the historical basic data, and the method specifically comprises the following steps:

aggregating and combining the indexes according to different levels according to the correlation influence and membership among the indexes of the basic data of the past year to form a multi-level analysis structure model, determining the relatively important weight value or the scheduling of the relative quality sequence, and adjusting the result in real time according to the actual situation so as to obtain the subjective weight value of each index;

10. A population size prediction system, the system comprising: