A multi-source data sharing method for urban and rural planning

Abstract

The present invention relates to the field of data processing technology, and in particular to a multi-source data sharing method for urban and rural planning. The method provided by the present invention comprises obtaining multiple data sources for urban and rural planning; analyzing the degree of association between the multiple data sources and the quality reliability of each data source, and calculating the information support of each data source in urban and rural planning based on the degree of association and the quality reliability; based on the information support, using the principal component analysis algorithm to perform weighted dimensionality reduction on the multiple data sources to obtain and share the reduced-dimensional data. The method provided by the present invention can not only accurately identify key data in multi-source data, but also avoid the influence of erroneous data or low-reliability data on urban and rural planning decisions. At the same time, it can also effectively integrate and analyze multi-source data to achieve data sharing, making comprehensive analysis and decision-making of urban and rural planning easier.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a multi-source data sharing method for urban and rural planning. Background Art

[0002] In the current rapidly developing information age, urban and rural planning faces unprecedented challenges and opportunities. With the widespread application of technologies such as big data, the Internet of Things, and cloud computing, a large amount of multi-source data has been accumulated in the field of urban and rural planning. These data come from a wide range of sources and are of various types, covering everything from remote sensing images, socio-economic statistics to real-time traffic information, social media content, etc. Effective sharing of multi-source data can improve the scientificity, accuracy, and efficiency of urban and rural planning.

[0003] However, in the process of storing and sharing urban and rural planning data sets from different sources, a large amount of computing resources are required for data sharing due to the characteristics of large sample size, diverse types and rich information of multi-source data for urban and rural planning. At present, the principal component analysis (PCA) method is used to reduce the dimension of multi-source data for urban and rural planning to obtain the principal component scores, and then evaluate the planning area. However, in the process of dimensionality reduction, PCA only retains the first few principal components with the largest variance, and discards information in other directions, which may lead to the loss of some key data, especially when the discarded components contain factors that have an important impact on urban and rural planning. Therefore, in the process of PCA dimensionality reduction, how to retain the key data in multi-source data to prevent the components with important influence from being discarded is an urgent problem to be solved. Summary of the invention

[0004] In order to solve the above-mentioned technical problem of how to retain key data in multi-source data, the purpose of the present invention is to provide a multi-source data sharing method for urban and rural planning. The technical solution adopted is as follows:

[0005] The present invention provides a method for sharing multi-source data for urban and rural planning, the method comprising:

[0006] Access multiple data sources for urban and rural planning;

[0007] Analyze the correlation between the multiple data sources and the quality reliability of each data source, and calculate the information support of each data source in urban and rural planning based on the correlation and quality reliability;

[0008] Based on the information support, the principal component analysis algorithm is used to perform weighted dimensionality reduction on the multiple data sources to obtain and share the reduced-dimensional data.

[0009] Furthermore, the analyzing the correlation between the multiple data sources and the quality reliability of each data source includes:

[0010] Analyze the average change synchronization between two data sources over time and the change synchronization within the neighborhood respectively, and obtain the first correlation degree and the second correlation degree accordingly;

[0011] Determining the credibility of the data source according to the difference between the first correlation degree and the second correlation degree, and determining the quality reliability according to the credibility and the first correlation degree;

[0012] Accordingly, the information support degree of each data source in urban and rural planning is obtained by calculating the degree of association and the quality reliability, including:

[0013] determining a reference priority of each data source based on the first association degree and the total number of the data sources;

[0014] The information support is obtained by calculation according to the quality reliability and the reference priority.

[0015] Furthermore, the step of analyzing the average change synchronization between two data sources over time to obtain a first correlation degree includes:

[0016] Obtaining corresponding change trend information in two data sources, wherein the corresponding change trend information is a future change trend represented by corresponding data values ​​in two data sources at the same time;

[0017] The average change synchronization between any two data sources is calculated according to the number of the data values ​​and the change trend information to obtain a first correlation degree.

[0018] Further, the determining the reference priority of each data source based on the first association degree and the total number of the data sources includes:

[0019] Taking a first correlation degree greater than or equal to a preset first correlation degree threshold value among the multiple first correlation degrees as a target correlation degree, and taking the pairwise data sources corresponding to the target correlation degree as strongly correlated data sources;

[0020] Count the number of data sources that are strongly associated with the current data source, and calculate the proportion of strongly associated data sources based on the total number of data sources;

[0021] The first correlation degree average of the current data source is calculated using the multiple first correlation degrees, and the reference priority of the current data source is calculated based on the first correlation degree average and the proportion of the strongly correlated data sources.

[0022] Further, the future change trend is determined according to the slope of a line segment between a current data value and a next adjacent data value, and the change synchronization within the neighborhood range is the change synchronization within the neighborhood range of the current data value.

[0023] Further, determining the quality reliability according to the credibility and the first correlation degree includes:

[0024] The credibility is used as the weight of the first correlation degree, and the weighted sum of the M-1 first correlation degrees of the current data source is calculated, and the mean of the weighted sum results is used as the average correlation reliability of the current data source;

[0025] The ratio between the average correlation reliability and the average of M-1 first correlation degrees is used as the quality reliability of the current data source; wherein M represents the total number of data sources.

[0026] Furthermore, based on the information support, the plurality of data sources are weightedly reduced in dimension using a principal component analysis algorithm to obtain dimension-reduced data, including:

[0027] Multiplying the information support of each data source by each data value in the multiple data sources to obtain weighted data;

[0028] Performing covariance calculation on the weighted data to obtain weighted covariance;

[0029] Performing eigendecomposition on the weighted covariance to obtain eigenvalues ​​and corresponding eigenvectors, and determining principal components based on the eigenvalues ​​and eigenvectors;

[0030] The multiple data sources are projected onto the determined principal components to obtain dimensionally reduced data.

[0031] Furthermore, the multiple data sources include point of interest data, socio-economic statistics, real-time traffic data, and remote sensing image data.

[0032] Furthermore, after obtaining a plurality of data sources for urban and rural planning, the method further comprises:

[0033] The original formats of the multiple data sources are uniformly converted into a target format, abnormal data and erroneous data are eliminated from the converted data sources, missing values ​​are filled in the eliminated data, and finally the dimensions of the filled data are unified to obtain data sources with consistent dimensions;

[0034] Accordingly, the analyzing the correlation between the multiple data sources and the quality reliability of each data source includes:

[0035] Analyze the degree of correlation between multiple data sources with consistent dimensions and the quality reliability of each data source.

[0036] Furthermore, the shared dimension-reduced data includes:

[0037] The reduced-dimensionality data is uploaded to a cloud platform so that authorized users can obtain the reduced-dimensionality data from the cloud platform.

[0038] The present invention has the following beneficial effects: by analyzing the correlation degree and quality reliability of the data sources, the correlation degree and quality reliability of each data source are obtained, and the information support degree is obtained according to the correlation degree and quality reliability. The larger the information support degree is, the current data source has a strong dependency relationship with other data sources and has higher accuracy and reliability, that is, the current data source has a higher importance in urban and rural planning. Further, by using the information support degree to perform weighted calculation on each data value, it can mine potential key data, avoid the key data from being discarded in the subsequent data dimensionality reduction, and avoid the influence of erroneous data or data with low reliability on urban and rural planning. Finally, the principal component analysis algorithm is used to perform data dimensionality reduction, thereby simplifying the data structure, reducing the data dimension, and avoiding the waste of data storage resources. The multi-source data sharing method for urban and rural planning provided by the present invention can not only accurately identify the key data in the multi-source data, but also avoid the influence of erroneous data or data with low reliability on urban and rural planning decisions. At the same time, it can also effectively integrate and analyze multi-source data, realize data sharing, and make the comprehensive analysis and decision-making of urban and rural planning easier. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A schematic flow chart of a multi-source data sharing method for urban and rural planning provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0041] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the multi-source data sharing method for urban and rural planning proposed by the present invention, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0042] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0043] The specific scheme of a multi-source data sharing method for urban and rural planning provided by the present invention is described in detail below with reference to the accompanying drawings.

[0044] See also Figure 1 , which shows a method flow chart of a multi-source data sharing method for urban and rural planning provided by an embodiment of the present invention, the multi-source data sharing method for urban and rural planning comprises the following steps:

[0045] S101, obtaining multiple data sources for urban and rural planning.

[0046] Among them, multiple data sources include POI (point of interest) data, socioeconomic statistics, real-time traffic data, remote sensing image data, social media data, etc., which can be obtained from different government development data platforms, urban planning and design system platforms, geographic information systems and other channels.

[0047] S102, analyzing the correlation degree between the multiple data sources and the quality reliability of each data source, and calculating the information support degree of each data source in urban and rural planning based on the correlation degree and the quality reliability.

[0048] The degree of correlation between data sources can reflect the degree of mutual dependence between data sources. If the degree of correlation between two data sources is greater, the degree of dependence between the two data sources is higher. If the degree of correlation between a data source and the remaining data sources is greater, it means that the data source has a high degree of dependence on all the remaining data sources, and the data in the data source is more critical. Therefore, by analyzing the degree of correlation between data sources, representative and more critical data sources can be obtained.

[0049] The quality reliability of a data source can reflect the accuracy and reliability of the data source, which is obtained through further analysis based on the degree of association. In actual scenarios, if the degree of association of a data source is high, but the quality reliability of this data source is low, then this data source cannot provide very important information for urban and rural planning, and cannot be used as an important reference factor, such as data sources with backward equipment or more manual intervention. Therefore, it is necessary to further analyze the quality reliability of the data source to further avoid the impact of erroneous data or low-reliability data on urban and rural planning decisions.

[0050] In summary, it is necessary to comprehensively consider the correlation between the current data source and other data sources and the quality reliability of the current data source to calculate the information support of the current data source. Only when the correlation and quality reliability are both large, the information support will be large. The larger the information support, the more important the current data source is in urban and rural planning.

[0051] S103: Based on the information support, a principal component analysis algorithm is used to perform weighted dimensionality reduction on the multiple data sources to obtain and share reduced-dimensionality data.

[0052] In this step, the information support is used as a weighting coefficient to perform weighted calculations with the data values ​​in multiple data sources, and then the weighted data is subjected to dimensionality reduction processing through the principal component analysis algorithm to obtain the reduced-dimensional data and share it.

[0053] In urban and rural planning, the original different data sources contain multiple indicators related to social economy, environmental quality, traffic flow, etc. The principal component scores are obtained by performing weighted PCA dimensionality reduction on multiple data sources. The principal component is formed by combining these indicators with different weighting coefficients of multiple data sources. Planners can use these principal component scores to evaluate the comprehensive development level, environmental conditions or traffic pressure of different regions. In the process of weighted PCA dimensionality reduction, the accuracy of the weighting coefficients of each data source is the key to mining potential key data. The present invention uses the information support obtained by the above calculation to perform weighted dimensionality reduction, so as to better reflect the relative importance of data sources in weighted principal component analysis.

[0054] The multi-source data sharing method for urban and rural planning provided by the embodiment of the present invention obtains the degree of association and quality reliability of each data source by analyzing the degree of association and quality reliability of the data source, and obtains the information support according to the degree of association and quality reliability. The larger the information support, the stronger the dependence relationship between the current data source and other data sources, and the higher the accuracy and reliability, that is, the higher the importance of the current data source in urban and rural planning. Further, by using the information support to perform weighted calculation on each data value, it can mine potential key data, avoid the key data from being discarded in the subsequent data dimensionality reduction, and avoid the influence of erroneous data or data with low reliability on urban and rural planning. Finally, the principal component analysis algorithm is used to reduce the data dimensionality, thereby simplifying the data structure, reducing the data dimension, and avoiding the waste of data storage resources. The multi-source data sharing method for urban and rural planning provided by the present invention can not only accurately identify the key data in the multi-source data, but also avoid the influence of erroneous data or data with low reliability on urban and rural planning decisions. At the same time, it can also effectively integrate and analyze multi-source data, realize data sharing, and make the comprehensive analysis and decision-making of urban and rural planning easier.

[0055] In some embodiments of the present invention, analyzing the degree of association between the multiple data sources and the quality reliability of each data source includes:

[0056] The average change synchronization between the two data sources over time and the change synchronization within the neighborhood are analyzed respectively, and the first correlation degree and the second correlation degree are obtained accordingly.

[0057] Among them, the first degree of association is the analysis result of the average change synchronization. The larger the value of the first degree of association, the more consistent the change trends in the two data sources are, that is, the closer the correlation between the two data sources is. Conversely, the smaller the value of the first degree of association, the more different the change trends in the two data sources are, and the correlation between the two data sources is low. The second degree of association is the analysis result of the synchronization of changes within the neighborhood. The first degree of association is calculated based on the future change trend of a single data value, and the second degree of association is calculated based on the change trend within the neighborhood where the single data value is located. If the first degree of association is closer to the second degree of association, it means that the two types of data sources are stable and have consistent change trends.

[0058] The credibility of the data source is determined according to the difference between the first correlation degree and the second correlation degree, and the quality reliability is determined according to the credibility and the first correlation degree.

[0059] The credibility is calculated by the following formula:

[0060]

[0061] In the formula, Indicates Class data source and The credibility of similar data sources in urban and rural planning; is the total number of data values; Indicates Class data source and The corresponding data source of the class The synchronization of changes within the neighborhood of a data value is the second degree of association; Indicates Class data source and The average change synchronization between class data sources is the first degree of association. is a natural exponential function.

[0062] Indicates Class data source and N mean differences between class data, where That is, The difference between the first degree of association and the second degree of association at the data value. The smaller the value, the Class data source and If the data sources are stable and have a consistent trend of change, and the correlation relationship remains consistent, then the corresponding credibility of the two types of data sources in urban and rural planning is greater, and the greater the credibility, the higher the reliability of the data source. The larger the value, the smaller the credibility.

[0063] Since the above credibility is the credibility of each data source, in order to calculate the quality reliability of each data source, the first correlation degree can be introduced to calculate it. Specifically, the credibility is used as the weight of the first correlation degree to perform weighted calculation on the M-1 first correlation degrees of the current data source, and the quality reliability of the current data source is obtained based on the weighted calculation result; where M represents the total number of data sources.

[0064] Accordingly, the information support degree of each data source in urban and rural planning is obtained by calculating the degree of association and the quality reliability, including:

[0065] The reference priority of each data source is determined based on the first degree of association and the total number of the data sources. The larger the reference priority, the stronger the change synchronization between the current data source and the other multiple data sources. The current data source can provide more comprehensive and important information for urban and rural planning, and is the preferred data reference object. Conversely, the smaller the reference priority, the less important or less comprehensive the corresponding data source is.

[0066] The information support is obtained by calculation according to the quality reliability and the reference priority.

[0067] Among them, if the quality reliability is high and the reference priority is also high, the calculated information support is also greater, which means that the corresponding data source is both comprehensive and accurate, and its importance is greater. On the contrary, it means that the importance of the corresponding data source is small. The information support is specifically calculated by the following formula:

[0068]

[0069] In the formula, Indicates The information support of similar data sources in urban and rural planning; Indicates Reference priority of class data sources in urban and rural planning; Indicates The quality reliability of the data source.

[0070] Indicates when When the reference priority and quality reliability of this type of data source in urban and rural planning are both large, it plays a key role in urban and rural planning, and the greater the information support of this type of data source.

[0071] In some embodiments of the present invention, analyzing the average change synchronization between two data sources over time to obtain the first correlation degree includes:

[0072] The corresponding change trend information in two data sources is obtained, wherein the corresponding change trend information is the future change trend represented by the corresponding data values ​​in the two data sources at the same time.

[0073] The average change synchronization between any two data sources is calculated according to the number of the data values ​​and the change trend information to obtain a first correlation degree.

[0074] In this embodiment, a line graph of data value changes is drawn for each data source with time as the horizontal axis and data value as the vertical axis, and average change synchronization analysis is performed based on the line graph. The following is an example of calculating the first correlation degree between the i-th data source and the j-th data source:

[0075] The nth data value is obtained from the change line graphs corresponding to the i-th data source and the j-th data source, and the time information of the two selected data values ​​is the same. Then the slope of the line segment between the two selected data values ​​and the next adjacent data value in their respective change line graphs is determined, so as to obtain the nth data value of the i-th data source. Future trend of data values , the jth data source Future trend of data values , and That is, the corresponding change trend information. Then, by calculating the future change trend And future trends The difference between them is used to determine the similarity of the change trends of the nth data value of the i-th data source and the j-th data source. Finally, the average similarity of the change trends of the N data values ​​in the i-th data source and the j-th data source is calculated, and the first correlation degree between the i-th data source and the j-th data source is obtained based on the average similarity of the change trends. , see the following formula for details:

[0076]

[0077] In the formula, Indicates Class data source The slope of the line segment between a data value and the next adjacent data value in the line graph represents the Class Data Source The future trend of changes at the data value; Indicates Class data source The slope of the line segment between the jth data value and the next adjacent data value in the line graph represents the The future trend of changes at the data value, is a natural exponential function.

[0078] It should be noted that each data source includes N data values. Indicates Class data source and The smaller the similarity of the changing trends between the corresponding data values ​​of the same type of data source, the greater the consistency of the changing trends of each data between the i-th data source and the j-th data source, and the stronger the average changing synchronization between the two types of data sources, the more consistent the corresponding data changing trends are, that is, the closer the correlation between the two types of data sources is.

[0079] All data sources are calculated through the above-mentioned first correlation degree calculation process, and each data source corresponds to M-1 first correlation degrees; wherein M represents the total number of data sources.

[0080] The multi-source data sharing method for urban and rural planning provided by the embodiment of the present invention determines the similarity of the change trends between two data sources by analyzing the future change trends of each data value, thereby determining the first degree of correlation between the two data sources. It can fully mine potential key data to avoid being discarded in principal component analysis, and provide important data for urban and rural planning.

[0081] In some embodiments of the present invention, the stronger the average change synchronization between data sources represented by the above-mentioned first correlation degree, the more consistent the change trend of the data sources, and the closer the correlation between the data sources. Therefore, if a data source has a strong correlation with multiple data sources among the remaining M-1 data sources, it means that the data source is very important in providing comprehensive information on urban and rural planning, and thus it can be regarded as an important reference data source for urban and rural planning. When the first correlation degree between a data source and multiple other data sources is greater and there are multiple such first correlation degrees, the reference priority of the data source in urban and rural planning is higher. Based on this, the first correlation degree and the preset first correlation degree threshold can be used to calculate the specific reference priority value, specifically:

[0082] A first correlation degree among the multiple first correlation degrees that is greater than or equal to a preset first correlation degree threshold is used as a target correlation degree, and the pairwise data sources corresponding to the target correlation degree are used as strongly correlated data sources.

[0083] Count the number of data sources that are strongly associated with the current data source, and calculate the proportion of strongly associated data sources based on the total number of data sources.

[0084] The first correlation degree average of the current data source is calculated using the multiple first correlation degrees, and the reference priority of the current data source is calculated based on the first correlation degree average and the proportion of the strongly correlated data sources. That is, the reference priority is calculated based on the following formula:

[0085] The quality reliability is calculated by the following formula:

[0086]

[0087] In the formula, Indicates Quality reliability of class data sources; Indicates Class data source and The synchronization of changes between class data sources is the first degree of association; Indicates Class data source and The corresponding credibility of the class data source in urban and rural planning; is the average correlation reliability, is the mean of the first degree of association.

[0088] Reliability is the first The quality reliability is calculated by the weight of the first correlation degree between the data source and other data sources. Class data source and The greater the first correlation degree of the class data source and the greater the corresponding credibility, the calculated The larger the quality reliability value of a class data source is, the greater the reliability of the data source is, and M represents the total number of data sources.

[0089] In some embodiments of the present invention, based on the information support, using a principal component analysis algorithm to perform weighted dimensionality reduction on the multiple data sources to obtain dimensionality-reduced data includes:

[0090] The information support of each data source is multiplied by each data value in the multiple data sources to obtain weighted data.

[0091] Covariance calculation is performed on the weighted data to obtain weighted covariance.

[0092] The weighted covariance is subjected to eigendecomposition to obtain eigenvalues ​​and corresponding eigenvectors, and principal components are determined according to the eigenvalues ​​and the eigenvectors.

[0093] The multiple data sources are projected onto the determined principal components to obtain dimensionally reduced data.

[0094] In this embodiment, since urban and rural planning usually involves data from different sources with high dimensions and different scales and importance, high-dimensional data processing is usually computationally complex and resource-intensive. The information support of each data source for urban and rural planning obtained above provides necessary information for urban and rural planning. Information support weighted PCA can be used to reduce the dimension of the data, thereby improving computing efficiency and storage efficiency, making data processing and analysis more efficient, facilitating real-time updates and decision support, and making complex data in urban and rural planning more intuitive and easy to understand, thereby supporting better communication and decision-making.

[0095] Specifically, the above multiple data sources are organized into a matrix, each row represents a data source, and each column represents a data value. Each data value in the data matrix is ​​multiplied by the information support of the corresponding data source to form a weighted data matrix. The weighted data matrix is ​​subjected to covariance calculation to obtain a weighted covariance matrix. The weighted covariance matrix reflects the relationship between the data sources and their weighted influence. Then, the weighted covariance matrix is ​​subjected to eigenvalue decomposition to obtain the eigenvalues ​​and corresponding eigenvectors, and the principal components are further selected, and the original multiple data sources are projected onto the selected principal components to form the reduced-dimensional data, which is of key reference significance for urban and rural planning.

[0096] In some embodiments of the present invention, after obtaining a plurality of data sources for urban and rural planning, the method further comprises:

[0097] The original formats of the multiple data sources are uniformly converted into a target format, abnormal data and erroneous data are eliminated from the converted data sources, and missing values ​​are filled in the eliminated data. Finally, the dimensions of the filled data are unified to obtain a data source with consistent dimensions, so as to facilitate subsequent correlation degree and quality reliability analysis.

[0098] Among them, unified format conversion can be achieved through data processing software such as ArcGIS and Python. The elimination of abnormal data and erroneous data can be completed by filtering data values ​​by setting filtering thresholds. Missing value filling can be achieved through interpolation algorithms. The algorithm for unified dimension can be normalization, standardization, etc.

[0099] The multi-source data sharing method for urban and rural planning provided by the embodiment of the present invention realizes quality inspection of multiple data sources by eliminating abnormal data and erroneous data, ensures data integrity by filling missing values, and eliminates the influence of data of different dimensions by unifying dimensions.

[0100] In some embodiments of the present invention, the sharing of dimension-reduced data includes:

[0101] The reduced-dimensionality data is uploaded to a cloud platform so that authorized users can obtain the reduced-dimensionality data from the cloud platform.

[0102] Specifically, the urban and rural planning data after weighted PCA dimensionality reduction can be uploaded to the cloud platform using a network sharing service. Before uploading, the dimensionality-reduced data can be organized into a compatible format (e.g., JSON) to ensure a smooth upload process. Then, select a suitable cloud platform (e.g., AWS S3) and upload the dimensionality-reduced data to a specified storage bucket or directory in the cloud platform.

[0103] In addition, in order to achieve data sharing, a unified access interface can be established, such as through the API provided by the cloud platform or a custom developed interface to allow authorized users to query, download and access data, while supporting different data formats and query parameters. Set access permissions based on user identity to ensure that only authorized users can access the reduced data. Regularly monitor access logs to ensure compliance with data usage policies and make necessary permission adjustments to maintain the security and effectiveness of the multi-source data sharing system for urban and rural planning.

[0104] In summary, the multi-source data sharing method for urban and rural planning provided by the embodiment of the present invention can more accurately understand and apply the interdependence between these data, evaluate the accuracy and reliability of the data, and find potential data deviations or anomalies by comparing the information of different data sources, thereby ensuring the reliability of the decision-making basis. According to the information support of the data source, multiple data sources are weighted and reduced in dimension, thereby simplifying the data structure and reducing the dimension, while retaining the most important information, effectively integrating and analyzing multiple data sources, and then realizing data sharing, so as to make it easier to conduct comprehensive analysis and decision-making on urban and rural planning.

[0105] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0106] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A multi-source data sharing method for urban and rural planning, characterized in that: The multi-source data sharing method for urban and rural planning includes: Access multiple data sources for urban and rural planning; Analyze the correlation between the multiple data sources and the quality reliability of each data source, and calculate the information support of each data source in urban and rural planning based on the correlation and quality reliability; Based on the information support, the plurality of data sources are weightedly reduced in dimension using a principal component analysis algorithm to obtain and share reduced-dimensional data; The analyzing the correlation between the multiple data sources and the quality reliability of each data source includes: Analyze the average change synchronization between two data sources over time and the change synchronization within the neighborhood respectively, and obtain the first correlation degree and the second correlation degree accordingly; Determining the credibility of the data source according to the difference between the first correlation degree and the second correlation degree, and determining the quality reliability according to the credibility and the first correlation degree; Accordingly, the information support degree of each data source in urban and rural planning is obtained by calculating the degree of association and the quality reliability, including: determining a reference priority of each data source based on the first association degree and the total number of the data sources; Calculate and obtain information support according to the quality reliability and the reference priority; The step of performing weighted dimensionality reduction on the plurality of data sources based on the information support by using a principal component analysis algorithm to obtain dimensionality-reduced data includes: Multiplying the information support of each data source by each data value in the multiple data sources to obtain weighted data; Performing covariance calculation on the weighted data to obtain weighted covariance; Performing eigendecomposition on the weighted covariance to obtain eigenvalues ​​and corresponding eigenvectors, and determining principal components based on the eigenvalues ​​and eigenvectors; Projecting the multiple data sources onto the determined principal components to obtain dimensionally reduced data; Obtaining corresponding change trend information in two data sources, wherein the corresponding change trend information is the future change trend represented by the corresponding data values ​​in the two data sources at the same time; the future change trend is determined according to the slope of the line segment between the current data value and the next adjacent data value, and the change synchronization within the neighborhood is the change synchronization within the neighborhood of the current data value; The first correlation degree is calculated based on the future change trend of a single data value, and the first correlation degree is the analysis result of the average change synchronization; the second correlation degree is calculated based on the change trend within the neighborhood where the single data value is located, and the second correlation degree is the analysis result of the change synchronization within the neighborhood; The average change synchronization between the two data sources is calculated according to the number of the data values ​​and the change trend information to obtain a first correlation degree. The calculation formula of the first correlation degree is: In the formula, Indicates Class data source and The average change synchronization between the class data sources is the first degree of association, N is the total number of data values, Indicates Class data source The slope of the line segment between a data value and the next adjacent data value in the line graph represents the Class Data Source The future trend of changes at the data value; Indicates Class data source The slope of the line segment between the jth data value and the next adjacent data value in the line graph represents the The future trend of changes at the data value, is a natural exponential function; The determining the quality reliability according to the credibility and the first correlation degree includes: The credibility is used as the weight of the first correlation degree, and the weighted sum of the M-1 first correlation degrees of the current data source is calculated, and the mean of the weighted sum results is used as the average correlation reliability of the current data source; The ratio between the average correlation reliability and the average of M-1 first correlation degrees is used as the quality reliability of the current data source; wherein M represents the total number of data sources.

2. A multi-source data sharing method for urban and rural planning according to claim 1, characterized in that: The determining the reference priority of each data source based on the first association degree and the total number of the data sources includes: Taking a first correlation degree greater than or equal to a preset first correlation degree threshold value among the multiple first correlation degrees as a target correlation degree, and taking the pairwise data sources corresponding to the target correlation degree as strongly correlated data sources; Count the number of data sources that are strongly associated with the current data source, and calculate the proportion of strongly associated data sources based on the total number of data sources; The first correlation degree average of the current data source is calculated using the multiple first correlation degrees, and the reference priority of the current data source is calculated based on the first correlation degree average and the proportion of the strongly correlated data sources.

3. A multi-source data sharing method for urban and rural planning according to claim 1, characterized in that: The multiple data sources include point of interest data, socio-economic statistics data, real-time traffic data, and remote sensing image data.

4. A multi-source data sharing method for urban and rural planning according to claim 1, characterized in that: After obtaining the plurality of data sources for urban and rural planning, the method further comprises: The original formats of the multiple data sources are uniformly converted into a target format, abnormal data and erroneous data are eliminated from the converted data sources, missing values ​​are filled in the eliminated data, and finally the dimensions of the filled data are unified to obtain data sources with consistent dimensions; Accordingly, the analyzing the correlation between the multiple data sources and the quality reliability of each data source includes: Analyze the degree of correlation between multiple data sources with consistent dimensions and the quality reliability of each data source.

5. A multi-source data sharing method for urban and rural planning according to claim 1, characterized in that: Share the reduced-dimensional data, including: The reduced-dimensionality data is uploaded to a cloud platform so that authorized users can obtain the reduced-dimensionality data from the cloud platform.

Images