CN-121981342-A - Land scale prediction method based on weight calculation

Abstract

The invention provides a land scale prediction method based on weight calculation, which relates to the technical field of data processing, and comprises the steps of carrying out space resource retrieval on a target space data set to obtain a space data list; the method comprises the steps of performing space topology mapping on a space data list to obtain a space adjacent matrix, performing space weighting update on data nodes to obtain a mutual plastic feature vector, performing similarity evaluation on the mutual plastic feature vector to obtain a similarity weight coefficient, performing weighting correction on the space adjacent matrix to obtain a mutual plastic adjacent weight matrix, performing topology analysis on the mutual plastic adjacent weight matrix to obtain space structural features, performing feature-structure cooperative weight distribution on the data nodes to obtain an initial weight distribution field, performing compatibility weighting correction on the initial weight distribution field to obtain an optimized weight distribution field, and performing space overall planning on the optimized weight distribution field to obtain a scale scheme.

Inventors

• ZHENG MENG
• Ji Xiangran

Assignees

• 山东省地质测绘院

Dates

Publication Date

20260505

Application Date

20260408

Claims (10)

1. 1. A land scale prediction method based on weight calculation, the method comprising: s01, carrying out space resource retrieval on a target space data set to obtain a space data list of the target space data set; s02, performing space topology mapping on the data nodes of the space data list to obtain a space adjacency matrix of the data nodes; S03, based on the space adjacent matrix, carrying out space weighted update on basic attribute characteristics of the data nodes to obtain mutual plastic characteristic vectors of the data nodes, carrying out similarity quantitative evaluation on the mutual plastic characteristic vectors to obtain similarity weight coefficients of the mutual plastic characteristic vectors, and carrying out element-by-element weighted correction on the space adjacent matrix based on the similarity weight coefficients to obtain the mutual plastic adjacent weight matrix of the data nodes; s04, carrying out topological structure analysis on the mutual plastic adjacent weight matrix to obtain the spatial structure characteristics of the data nodes; S05, based on the mutual plastic feature vector and the space structure feature, performing feature-structure cooperative weight distribution on the data nodes to obtain an initial weight distribution field of the target space data set; S06, based on preset space constraint, carrying out compatibility weighted correction on the initial weight distribution field to obtain an optimized weight distribution field of the target space data set; S07, performing space overall planning on the optimized weight distribution field to obtain a scale scheme of the target space data set.
2. 2. The land scale prediction method based on weight calculation of claim 1, wherein the performing spatial resource retrieval on the target spatial data set to obtain a spatial data list of the target spatial data set comprises: performing space information retrieval on a target space data set to obtain a space distribution list of the target space data set; carrying out structural analysis on the entity objects of the spatial distribution list to obtain spatial attribute description of the entity objects; performing spatial feature screening on the spatial attribute description to obtain spatial key features of the entity object; And based on the space key characteristics, performing entity object space merging on the space distribution list to obtain a space data list of the target space data set.
3. 3. The land scale prediction method based on weight calculation of claim 1, wherein said performing spatial topology mapping on data nodes of said spatial data list to obtain a spatial adjacency matrix of said data nodes comprises: acquiring the space position coordinates of the data nodes in the space data list; based on the space position coordinates, performing space proximity detection on the data nodes to obtain a node pair set of the data nodes; Carrying out shared boundary discrimination on the node pair set to obtain a space adjacent node pair set of the data node; Carrying out shared boundary extension measurement on the space adjacent node pair set to obtain the boundary contact length of the space adjacent node pair set; based on the boundary contact length, carrying out connection intensity quantification on the space adjacent node pair set to obtain adjacent weight of the space adjacent node pair; and carrying out topology association reconstruction on the data nodes based on the adjacent weights to obtain a space adjacent matrix of the data nodes.
4. 4. The land scale prediction method based on weight calculation of claim 1, wherein said spatial adjacency matrix based on said spatial adjacency matrix performs spatial weighting update on basic attribute features of said data nodes to obtain inter-plastic feature vectors of said data nodes, performs similarity quantization evaluation on said inter-plastic feature vectors to obtain similarity weight coefficients of said inter-plastic feature vectors, performs element-by-element weighting correction on said spatial adjacency matrix based on said similarity weight coefficients to obtain inter-plastic adjacency weight matrices of said data nodes, comprising: Performing matrix row normalization on the space adjacent matrix to obtain a space propagation weight matrix of the space adjacent matrix; based on the space propagation weight matrix, carrying out neighborhood weighting collection on the basic attribute characteristics of the data nodes to obtain neighborhood influence characteristics of the data nodes; performing feature blending mapping on the neighborhood influence features and the basic attribute features to obtain a mutual plastic feature vector of the data node; Performing vector projection coincidence degree evaluation on the mutual plastic characteristic vector to obtain similarity measure of the mutual plastic characteristic vector; Carrying out numerical normalization on the similarity measure to obtain a similarity weight coefficient of the mutual plastic characteristic vector; Performing element-by-element weighting correction on the space adjacent matrix based on the similarity weight coefficient to obtain a mutual plastic adjacent weight matrix of the space adjacent matrix; And carrying out iterative updating on the inter-plastic feature vector and the inter-plastic adjacent weight matrix until the difference degree of the inter-plastic feature vector and the inter-plastic adjacent weight matrix produced by iteration are lower than a preset tolerance threshold, taking the inter-plastic feature vector produced by the last iteration as the inter-plastic feature vector of the data node, and taking the inter-plastic adjacent weight matrix produced by the last iteration as the inter-plastic adjacent weight matrix of the data node.
5. 5. The land scale prediction method based on weight calculation as set forth in claim 4, wherein said performing a vector projection overlap ratio evaluation on said inter-plastic feature vector to obtain a similarity measure of said inter-plastic feature vector comprises: Performing direction projection deconstructment on the mutual plastic characteristic vector to obtain a projection component of the mutual plastic characteristic vector; Product aggregation is carried out on the projection components, so that the projection coincidence degree of the projection components is obtained; Performing direction deviation punishment weighting on the projection overlap ratio to obtain a direction correction overlap ratio of the projection overlap ratio; and calculating the similarity measure of the mutual plastic characteristic vector based on the direction correction coincidence degree, wherein the calculation formula of the similarity measure is as follows: ; In the formula, For the measure of similarity to be described, Is the first The inter-plastic feature vectors of the individual data nodes, Is the first Data node and the first The cosine value of the normal azimuth of the shared boundary of the individual data nodes, Is the first Data node and the first The sine value of the normal azimuth of the shared boundary of the individual data nodes, Is the first The inter-plastic feature vectors of the individual data nodes, In order to perform the modular arithmetic operation, As a function of the natural index of refraction, Is the first Data node and the first The normal azimuth of the shared boundary of the individual data nodes, Is the first Data node and the first The direction angle of the data node center line, As an absolute value operator, Penalty scale parameters for a preset directional deviation, Is a natural constant which is used for the production of the high-temperature-resistant ceramic material, Is the first Data node and the first The individual data nodes share the contact length of the boundary, And modulating the scale parameter for the preset contact length.
6. 6. The land scale prediction method based on weight calculation of claim 1, wherein the performing topology analysis on the inter-plastic abutment weight matrix to obtain the spatial structure feature of the data node comprises: traversing and extracting non-zero elements of the mutual plastic adjacent weight matrix to obtain position indexes and weight values of the non-zero elements; based on the position index and the weight value, carrying out weighted connection topology recombination on the data node to obtain a directional weighted topological graph of the data node; Performing bidirectional contribution sum-up on the in-edge strength and the out-edge strength of the topological nodes in the directional weighted topological graph to obtain the node comprehensive strength of the topological nodes; based on the comprehensive strength of the nodes, carrying out node importance sorting screening on the directional weighted topological graph to obtain core nodes of the directional weighted topological graph; and carrying out node attribute normalization on the core nodes to obtain the spatial structure characteristics of the data nodes.
7. 7. The land scale prediction method based on weight calculation of claim 1, wherein said performing feature-structure cooperative weight distribution on said data nodes based on said inter-plastic feature vector and said spatial structure feature to obtain an initial weight distribution field of said target spatial data set comprises: Based on the spatial structure characteristics, performing spatial topology flow direction tracing on the data nodes to obtain directed dependent paths of the data nodes; Carrying out path bearing capacity quantification on the directional dependent paths to obtain the dependent strength of the directional dependent paths; carrying out path orientation analysis on the directional dependent paths to obtain dependent direction feature vectors of the directional dependent paths; marking a data node with zero degree in the directed dependent path as a penetration starting point, and marking a data node with zero degree in the directed dependent path as a penetration end point; injecting the source bearing capacity into the permeation starting point based on the mutual plastic characteristic vector to obtain the initial bearing capacity of the permeation starting point; Carrying out radial permeation conveying on the initial bearing capacity based on the directional dependent path and the permeation end point to obtain the accumulated bearing capacity of the permeation start point; carrying out load characteristic coupling modulation on the data nodes based on the accumulated load capacity to obtain cooperative allocation weights of the data nodes; and carrying out space global normalization allocation on the target space data set based on the collaborative allocation weight to obtain an initial weight allocation field of the target space data set.
8. 8. The land scale prediction method based on weight calculation as set forth in claim 7, wherein the calculation formula of the cumulative load is: ; In the formula, Is the first The cumulative load-bearing capacity of the individual data nodes, Is the first The data node is configured to store data, Is the first in the directed dependent path All upstream nodes of the data nodes, Is the first The initial load-bearing capacity of the individual data nodes, Is the first Data node and the first The strength of the dependence of the individual data nodes, Is the first The data node is configured to store data, Is the first in the directed dependent path All downstream nodes of the individual data nodes, Is the first Data node and the first The strength of the dependence of the individual data nodes, Is a natural constant which is used for the production of the high-temperature-resistant ceramic material, Is a preset coupling coefficient of the bearing attribute, Is the first Transpose of the inter-plastic feature vectors of the individual data nodes, Is the first Data node and the first The dependency direction feature vector of each data node.
9. 9. The land scale prediction method based on weight calculation of claim 1, wherein said performing compatibility weighted correction on said initial weight distribution field based on a preset spatial constraint to obtain an optimized weight distribution field of a target spatial data set comprises: Performing constraint type screening on preset space constraints to obtain an upper limit constraint subset, a forbidden association constraint subset and an efficiency weight subset of the space constraints; Performing upper limit truncation treatment on the weight values of the initial weight distribution field based on the upper limit constraint subset to obtain an upper limit truncation weight distribution field of the target space data set; Based on the forbidden associated constraint subset, carrying out forbidden associated pair suppression on the upper limit truncated weight distribution field to obtain a suppression weight distribution field of the target space data set; performing efficiency weighted correction on the suppression weight distribution field based on the efficiency weight subset to obtain a corrected weight distribution field of the target space data set; and taking the corrected weight distribution field as an optimized weight distribution field of the target space data set.
10. 10. The land scale prediction method based on weight calculation of claim 1, wherein the spatially unifying the optimized weight distribution field to obtain the scale scheme of the target spatial data set comprises: Performing weight value traversal on the optimized weight distribution field to obtain a weight value sequence of the optimized weight distribution field; Based on the weight value sequence, carrying out weight grade demarcation on the data node to obtain a resource priority sequence of the data node; performing resource hierarchical allocation on the total amount of resources of the target space dataset based on the resource priority sequence to obtain the resource allocation amount of the data node; and carrying out space aggregation and integration on the resource allocation amount to obtain a scale scheme of the target space data set.

Description

Land scale prediction method based on weight calculation Technical Field The invention relates to the technical field of data processing, in particular to a land scale prediction method based on weight calculation. Background The existing method has the defects of insufficient mining depth of spatial topological association of a target spatial data set, does not carry out weighted update and mutual modeling analysis of spatial dimension on basic attribute features of data nodes, and carries out weight calculation only by means of single basic attribute features, so that the weight coefficients cannot accurately reflect spatial association features among the data nodes, the accuracy of feature characterization is lost, errors of basic data layers are brought to subsequent land scale prediction, and the technical effectiveness of prediction results is directly affected. In the prior art, in the weight distribution link of land scale prediction, the cooperative consideration of the data node inter-modeling feature vector and the space structure feature is not realized, the construction of an initial weight distribution field lacks scientific topology flow direction and bearing capacity penetration logic support, meanwhile, the correction means of the initial weight is single, the hierarchical compatibility weighting treatment is not carried out on different types of space constraints, the optimization suitability of the weight distribution field is insufficient, in addition, the space overall stage lacks a scientific resource distribution mechanism based on weight levels, the operation flow efficiency of the whole land scale prediction is finally lower, the matching degree of the predicted scale scheme and the actual space resource allocation requirement is poorer, and therefore, how to improve the land scale prediction efficiency based on weight calculation becomes the problem to be solved. Disclosure of Invention The invention provides a land scale prediction method based on weight calculation, which aims to solve the problems in the background technology. In order to achieve the above object, the present invention provides a land scale prediction method based on weight calculation, including: s01, carrying out space resource retrieval on a target space data set to obtain a space data list of the target space data set; s02, performing space topology mapping on the data nodes of the space data list to obtain a space adjacency matrix of the data nodes; S03, based on the space adjacent matrix, carrying out space weighted update on basic attribute characteristics of the data nodes to obtain mutual plastic characteristic vectors of the data nodes, carrying out similarity quantitative evaluation on the mutual plastic characteristic vectors to obtain similarity weight coefficients of the mutual plastic characteristic vectors, and carrying out element-by-element weighted correction on the space adjacent matrix based on the similarity weight coefficients to obtain the mutual plastic adjacent weight matrix of the data nodes; s04, carrying out topological structure analysis on the mutual plastic adjacent weight matrix to obtain the spatial structure characteristics of the data nodes; S05, based on the mutual plastic feature vector and the space structure feature, performing feature-structure cooperative weight distribution on the data nodes to obtain an initial weight distribution field of the target space data set; S06, based on preset space constraint, carrying out compatibility weighted correction on the initial weight distribution field to obtain an optimized weight distribution field of the target space data set; S07, performing space overall planning on the optimized weight distribution field to obtain a scale scheme of the target space data set. In a preferred embodiment, the performing spatial resource retrieval on the target spatial data set to obtain a spatial data list of the target spatial data set includes: performing space information retrieval on a target space data set to obtain a space distribution list of the target space data set; carrying out structural analysis on the entity objects of the spatial distribution list to obtain spatial attribute description of the entity objects; performing spatial feature screening on the spatial attribute description to obtain spatial key features of the entity object; And based on the space key characteristics, performing entity object space merging on the space distribution list to obtain a space data list of the target space data set. In a preferred embodiment, the performing spatial topology mapping on the data nodes of the spatial data list to obtain a spatial adjacency matrix of the data nodes includes: acquiring the space position coordinates of the data nodes in the space data list; based on the space position coordinates, performing space proximity detection on the data nodes to obtain a node pair set of the data nodes; Carrying out shared boundary discrimi