CN-122022928-A - Dynamic measurement and calculation method and system for house compensation facing to ground mode

CN122022928ACN 122022928 ACN122022928 ACN 122022928ACN-122022928-A

Abstract

The invention provides a dynamic measurement and calculation method and a system for house compensation oriented to a ground mode, which are characterized in that by carrying out structural modeling on various compensation rules in local policy files, extracting input variables, parameters and weights, accessing a house foundation platform to complete data standardization, constructing a symbolized tracking engine, abstracting a compensation rule calculation process into a directed graph, and realizing dynamic propagation and tracing of influence weights among nodes; in the calculation process, different rule types are distinguished, different weight adjustment is adopted, structural analysis results such as variable influence scores, main driving factor ordering and the like are generated, and finally, compensation amount composition and influence chains are displayed in a visual multi-angle mode.

Inventors

SHEN HAO
CHEN HANG
WU XIAOJIAN
HU HESONG
LIN TAO
LI XIN
LIU YADONG

Assignees

广州市建筑科学研究院集团有限公司
广州市市政集团有限公司房地产置业分公司
广州建设工程质量安全检测中心有限公司

Dates

Publication Date: 20260512
Application Date: 20260109

Claims (10)

1. A dynamic measurement and calculation method for house compensation facing to a land making mode is characterized by comprising the following steps: s1, constructing a compensation rule model based on a local policy file, defining input variables, basic parameters and calculation types of each compensation rule, and initializing basic influence weight values; S2, accessing a house basic data platform and a manual supplementary recording terminal, acquiring and analyzing house physical properties, ownership information and location characteristic data, and generating a standardized input variable set comprising building area, property state and building time; S3, constructing a symbolized tracking engine based on a directed graph structure, abstracting a compensation calculation flow into a node-side relation model, wherein the node comprises an input variable, a rule item and middle output, records the data flow direction and the action direction, and establishes a dynamic influence weight propagation channel; S4, when compensation amount calculation is executed, dynamically adjusting an influence weight propagation strategy according to the calculation type of each rule, and generating dynamic influence weight values corresponding to each intermediate result; s5, calculating the accumulated contribution intensity, the path depth and the action frequency of the input variable based on the propagation path data of the symbolized tracking engine, and generating a dynamic influence weight propagation topological graph; s6, sorting input variables according to the accumulated contribution intensity, and generating a structural analysis information packet by combining the path depth and the action frequency; And S7, carrying out visual synchronous output on the compensation amount value and the structured analysis information packet, presenting an expandable conduction chain flow chart on an interactive interface, and marking the value change quantity of each calculation node and the corresponding influence weight percentage.
2. The method for dynamic measurement and calculation of house compensation for ground-oriented mode according to claim 1, wherein said step S1 specifically comprises: carrying out structural analysis on the local policy file, extracting compensation rule items and applicable conditions thereof, and obtaining a rule item set related to house compensation in the policy text; performing field mapping processing on each rule based on the rule item set, and identifying an input variable on which the rule item set depends; extracting basic parameter values based on semantic analysis results of the policy treaties according to the description contents of the rule items; Based on the logic structure of the rule items, carrying out calculation type identification processing on each rule by utilizing a rule type classification algorithm to generate a rule calculation type identifier; Based on authority level, application range and history execution data of the policy treaty, performing basic influence weight assignment processing on each rule item to generate a basic influence weight configuration matrix; and carrying out structural binding on the rule item set, the input variable, the basic parameter value, the rule calculation type identifier and the basic influence weight configuration matrix to generate a configuration data packet of the compensation rule model.
3. The dynamic measurement and calculation method for house compensation for land making mode according to claim 2, wherein the compensation rule model adopts a text analysis engine to perform OCR recognition and semantic segmentation processing on the policy file of the electronic text or the scanned image, extracts compensation rule items and applicable conditions thereof, and generates a rule item set.
4. The method for dynamic measurement and calculation of house compensation for ground-oriented mode according to claim 1, wherein said step S2 specifically comprises: acquiring mapping data, property registration information and geographic information system location data based on a house base data platform interface protocol, and constructing an original data set of house physical properties and ownership information; Carrying out format standardization processing on building area fields in the original data set, and converting area data of different sources into square meter units for representation by adopting a unified unit conversion algorithm to obtain standardized building area variables; Executing a property status recognition algorithm based on the property registration information, extracting a house property type tag, and generating a standardized property status variable; performing time stamp analysis and classification processing on the house construction year field, and converting specific years into standardized construction year labels by adopting a year interval mapping algorithm; Based on the geographic information system location data, executing a space coordinate analysis and region coding mapping algorithm, converting longitude and latitude information into corresponding administrative division and region number, and generating a standardized location feature variable; And carrying out field combination and structured packaging on the standardized building area variable, the standardized property right state variable, the standardized building time tag and the standardized location characteristic variable to generate a house input variable set in a uniform format.
5. The method for dynamic measurement and calculation of house compensation for ground-oriented mode according to claim 1, wherein said step S3 specifically comprises: defining a node set of the directed graph based on rule items, input variables and intermediate calculation results in the compensation rule model to form a topological structure foundation of a compensation calculation path; constructing an edge set in the directed graph according to the data dependency relationship among the compensation rules, wherein each edge represents the data flow direction and the acting direction from one node to the other node; assigning an action type label to each edge based on the calculation type of the compensation rule, wherein the action type comprises linear addition, multiplication amplification and condition inheritance; constructing a runtime graph structure of a symbolized tracking engine, mapping all nodes and side relations in the compensation amount calculation process to a graph data structure in a memory, wherein the graph structure supports dynamic node expansion and side weight update, and tracks a compensation calculation path in real time; Initializing a dynamic influence weight propagation channel in the graph structure, and distributing an initial influence weight value for each node, wherein the initial influence weight value is assigned based on a basic influence weight value of a rule in a policy system; And dynamically updating node states and edge weight values in the graph structure based on the execution sequence and the data flow direction of the compensation rule, and generating a complete compensation amount influence propagation map.
6. The method of claim 5, wherein the set of nodes includes a house base attribute variable node, a compensation rule node, and an intermediate amount output node.
7. The method for dynamic measurement and calculation of house compensation for ground-oriented mode according to claim 1, wherein said step S4 specifically comprises: Classifying and identifying each rule based on rule calculation types in a predefined compensation rule model, wherein the rule calculation types comprise a linear addition rule, a multiplication rule and a condition judgment rule; performing weight superposition mode processing on the linear addition rule, and calculating the comprehensive influence weight value of the rule output node by adopting a weighted summation algorithm based on the original influence weight value of the input variable and the linear addition coefficient of the rule; performing marginal effect amplification mode processing on the multiplication type rule, and generating an amplification influence weight value of an output node of the rule based on a multiplication synthesis algorithm by utilizing the current influence weight value of an input variable and a multiplication coefficient of the rule; Executing weight inheritance mode processing on the condition judgment rule, and determining an inheritance influence weight value of the rule output node by adopting a branch weight inheritance algorithm based on an original influence weight value of an input variable and a logic branch result of the condition judgment rule; Taking the output influence weight value of each rule node as the input influence weight value of the next calculation node, and executing weight layer-by-layer propagation calculation based on a directed graph structure in a symbolized tracking engine to form a dynamic influence weight propagation sequence of each variable in the compensation amount formation path; And generating a dynamic influence weight value set of each intermediate computing node based on the dynamic influence weight propagation sequence.
8. The method for dynamic measurement and calculation of house compensation for ground-oriented mode according to claim 1, wherein said step S5 specifically comprises: Carrying out topological structure analysis on the node-side relation model output by the symbolized tracking engine, extracting a data flow path set between an input variable and a rule item, and constructing a variable propagation path set; calculating the path depth value of each input variable on each path based on each path in the variable propagation path set; Executing dynamic weight accumulation calculation on the nodes in each path, and calculating the accumulated contribution intensity of the input variable on the whole path by combining the dynamic influence weight value of the rule nodes; Based on the occurrence frequency of the input variable in the variable propagation path set, counting the action frequencies of the input variable in different paths; Based on the three indexes of the path depth value, the accumulated contribution intensity and the action frequency, an input variable influence scoring matrix is constructed, normalization processing is carried out on each variable, and a standardized variable influence scoring result is generated; And generating a dynamic influence weight propagation topological graph based on the variable influence scoring matrix and the propagation path structure.
9. The method for dynamic measurement and calculation of floor-oriented house compensation according to claim 8, wherein the dynamic influence weight propagation topological graph comprises node influence weight labels, path contribution strength labels and variable acting path visualization connecting lines.
10. A house compensation dynamic measurement and calculation system facing to a ground mode is characterized in that the house compensation dynamic measurement and calculation method facing to the ground mode is adopted to conduct the house compensation dynamic measurement and calculation under the ground mode according to any one of claims 1-9.

Description

Dynamic measurement and calculation method and system for house compensation facing to ground mode Technical Field The invention relates to the technical field of house collection compensation measurement and calculation and result interpretability enhancement, in particular to a dynamic measurement method and a system for house compensation oriented to a ground mode. Background The current house collection compensation measuring and calculating system is widely applied to the fields of urban updating, land collection and the like, aims at standardizing complex policy rules, and realizes automatic calculation of house compensation amount by using a data driving method. The mainstream prior art generally adopts pull-through modeling based on a policy rule engine or a parameterized formula, and comprehensive measurement and calculation of multiple indexes such as house physical properties, ownership states, location values, construction ages and the like are automatically completed. The common method comprises a flow-driven rule calling mode, a condition judging mode, a weight adding and subtracting mode and a linear or nonlinear superposition compensation amount calculating mode, and the system is generally in seamless joint with a house basic data management platform, so that great progress is realized in the aspects of improving measuring and calculating efficiency and guaranteeing result compliance. Along with the advancement of informatization, the standardization and localization adaptation degree of the compensation measuring and calculating system are continuously enhanced, and the functions of modularized rule configuration, batch data processing, file archiving, electronic signature and the like are gradually provided, so that the digitalization and the high efficiency of collection management work are promoted; however, the existing house compensation measuring and calculating system still has obvious defects in terms of result interpretation and user interactivity: (1) The currently output compensation result is mainly an amount value, and the automatic generation of explanatory information such as variable contribution degree, policy rule action path and the like which influence the amount is lacked; (2) In a complex multi-link calculation chain, the system is difficult to trace the specific action process of each input variable and the corresponding rule on the final amount, so that a policy making party and a person to be collected cannot intuitively understand calculation logic and weight setting, and the result public confidence is affected; (3) On the premise of lacking sensitivity analysis and variable path visualization capability, a system user is difficult to grasp sensitive driving factors of compensation results, and is also difficult to discover data anomalies or policy adaptation flaws in time, so that negotiation and collection decision efficiency is affected; (4) In order to improve the interpretability, partial external research attempts to introduce a complex proxy model or a statistical sampling method to realize local sensitivity analysis, but the means depend on a large amount of external data, have long training period and high integration cost, are difficult to directly embed an engineering compensation measuring and calculating system and are unfavorable for policy rapid iteration. Disclosure of Invention The invention aims to solve the technical problems and provides a dynamic measurement and calculation method and a system for house compensation oriented to a ground making mode. The technical scheme of the invention is realized in such a way that a dynamic measurement and calculation method of house compensation facing to a ground making mode comprises the following steps: s1, constructing a compensation rule model based on a local policy file, defining input variables, basic parameters and calculation types of each compensation rule, and initializing a basic influence weight value, wherein the basic influence weight value reflects the relative importance degree of the rule in a policy system; S2, accessing a house basic data platform and a manual supplementary recording terminal, acquiring and analyzing house physical properties, ownership information and location feature data, and generating a standardized input variable set comprising building area, property state and building time; S3, constructing a symbolized tracking engine based on a directed graph structure, abstracting a compensation calculation flow into a node-side relation model, wherein the node comprises an input variable, a rule item and middle output, records the data flow direction and the action direction, and establishes a dynamic influence weight propagation channel; S4, when compensation amount calculation is executed, dynamically adjusting an influence weight propagation strategy according to the calculation type of each rule, wherein a weight superposition mode is adopted for the linear addition ru