CN-121767109-B - Intelligent asset management system risk data processing method and system

CN121767109BCN 121767109 BCN121767109 BCN 121767109BCN-121767109-B

Abstract

The invention provides a risk data processing method and system of an intelligent asset management system, and relates to the technical field of intelligent asset management. The method comprises the steps of obtaining target property asset data, carrying out standardized processing on the target property asset data to obtain standard data, executing multidimensional association matching based on the standard data, calculating association confidence between the target property asset data and pre-stored core property archive data, and establishing logic association between the target property asset data and the core property archive data when the association confidence meets a preset threshold condition. The method aims to solve the problem that the existing real estate asset risk management system is limited in processing nonstandard data, so that risks of novel real estate assets cannot be effectively evaluated, real estate asset data from nonstandard input channels can be effectively processed, and accurate risk evaluation of the novel real estate assets is facilitated.

Inventors

HUANG YANBO
GUO HONGYU
WANG JUNJIE

Assignees

佛山建发智慧城市科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260303

Claims (4)

1. A method for processing risk data of an intelligent asset management system, comprising the steps of: s1, acquiring target property asset data which originate from a non-standard input channel and contain geographic position information and property identification information in a non-standard format; S2, carrying out standardized processing on the target property asset data to obtain standard data, wherein the specific steps in the step S2 comprise: converting the geographic position information in the non-standard format into geographic coordinates in a standard coordinate system by using a preset coordinate conversion rule, and converting the title identification information into standard title codes by using a preset code mapping relation to obtain standard data; S3, based on the standard data, executing multi-dimensional association matching and calculating association confidence coefficient between the target property asset data and pre-stored core property archive data, wherein the multi-dimensional association matching at least comprises determining spatial distance approximation degree of geographic coordinates, semantic similarity of property types and matching degree of association subjects, the association confidence coefficient is a calculation result of weighted summation of all dimensions, and the specific steps of determining the semantic similarity of the property types comprise: B1. extracting functional feature words and hierarchy qualifiers in the property type description by analyzing the property type description in the target property asset data; B2. step-by-step comparison is carried out on the functional feature words and the hierarchy qualifiers and a preset property type classification system in the core property file data, and the feature word matching quantity and the hierarchy depth are determined, wherein the specific steps in the step B2 comprise: B21. Identifying each matching path of the functional feature words and the hierarchy qualifiers in the property type classification system; B22. extracting auxiliary information related to each matching path from the target property asset data for each matching path identified; B23. And (3) evaluating the semantic consistency of each matching path and the auxiliary information to obtain a semantic consistency evaluation result of each matching path, wherein the specific steps in the step (B23) comprise: B231. constructing semantic feature vectors of all the matching paths and semantic feature vectors of the auxiliary information; B232. Calculating the similarity between the semantic feature vector of each matching path and the semantic feature vector of the auxiliary information, wherein the specific steps in the step B232 comprise: According to the semantic feature vector of each matching path and the semantic feature vector of the auxiliary information, calculating the similarity between the semantic feature vector of each matching path and the semantic feature vector of the auxiliary information by adopting a calculation method based on a vector included angle or an inner product; B233. according to the similarity, semantic consistency assessment results of all the matching paths are obtained; B24. Selecting a matching path with optimal consistency according to the semantic consistency evaluation result, and determining the matching quantity of the feature words and the hierarchical depth; B3. Determining a property type matching score according to the hierarchical depth and the feature word matching quantity; S4, when the association confidence degree meets a preset threshold value condition, establishing logic association between the target property asset data and the core property archive data.
2. The smart asset management system risk data processing method of claim 1, wherein the specific step of determining the spatial distance proximity of the geographic coordinates comprises: A1. Calculating the spatial distance between the target property asset data and the geographic location coordinates in the core property profile data according to the following formula: ; Wherein, the In order to be a spatial distance from each other, For the radius of the earth, For an arc corresponding to the latitude of the geographic coordinates in the target property asset data, For an arc corresponding to the latitude of the geographic coordinates of the geographic location feature in the core property profile data, For an arc corresponding to the longitude of the geographic coordinates in the target property asset data, An arc corresponding to the longitude of the geographic coordinates of the geographic location feature in the core property profile data; A2. Calculating the spatial distance approximation from the spatial distance: ; Wherein, the For the spatial distance approximation, Is a preset geographic error threshold.
3. The smart asset management system risk data processing method of claim 1, wherein the specific step of determining the degree of matching of the associated subject comprises: C1. identifying a primary legal entity name associated with subject information in the target property asset data; C2. Inquiring an association legal entity with an association relation with the main legal entity name in the core property archive data, and determining a relation weight according to the type of the association relation; C3. and determining the matching score of the association subject and taking the matching score as the matching degree of the association subject according to all queried association legal entities and corresponding relation weights thereof.
4. A smart asset management system risk data processing system employing the smart asset management system risk data processing method as claimed in any one of claims 1 to 3, comprising: The acquisition module is used for acquiring the target property asset data; the standardized processing module is used for carrying out standardized processing on the target property asset data to obtain standard data; the computing module is used for executing multidimensional association matching based on the standard data and computing the association confidence between the target property asset data and the pre-stored core property archive data; And the association generating module is used for establishing logic association between the target property asset data and the core property archive data when the association confidence degree meets a preset threshold condition.

Description

Intelligent asset management system risk data processing method and system Technical Field The invention relates to the technical field of intelligent asset management, in particular to a risk data processing method and system of an intelligent asset management system. Background With the deep advancement of smart city construction, the importance of the house property asset risk management system in the fields of city planning, financial decision making and the like is increasingly highlighted. The traditional system mainly aims at standardized property data, and the data processing flow and the evaluation mechanism are designed based on a relatively stable market environment. However, with the adjustment of urban development strategies, new property assets such as innovative industrial parks, talent apartments and the like are continuously emerging, which have risk characteristics that are distinct from those of traditional properties, and the value evaluation and risk monitoring requirements are more complex and dynamic. A core problem faced by existing systems is the limitation of their data processing capabilities. Firstly, the system adopts a fixed processing sequence, and the processing flow cannot be dynamically adjusted according to the risk priority of the novel asset, so that the analysis of key risk information is lagged. Secondly, to cope with the urgent need, an analyst has to enter data through a general data complement module whose check rule is not strict, which results in a large amount of geographical location information and title identification information in a non-standard format entering the system. Because the format of the data is not standard, the data cannot be effectively associated with the core property file, and a large amount of information islands are formed. More seriously, this lack of data correlation directly affects the accuracy of risk assessment. When the system performs multidimensional risk assessment, the risk assessment result of the novel asset deviates seriously from the actual situation because complete and accurate input data cannot be obtained, and even effective assessment cannot be generated. This not only affects the decision quality of the city manager, but also weakens their confidence in the system. The root of the problem is that the existing system lacks effective processing capability for nonstandard data, and particularly has technical defects in data standardization and multidimensional correlation matching. The system cannot perform unified standardized processing on the property data with different sources and different formats, and cannot establish an accurate multidimensional association relationship, so that the risk assessment capability of the system on novel property assets is directly restricted. In view of the above problems, no effective technical solution is currently available. Disclosure of Invention The invention aims to provide a risk data processing method and system of an intelligent asset management system, which aim to solve the problems that the existing real estate asset risk management system has limitations in processing nonstandard data, particularly has technical defects in data standardization and multidimensional association matching, so that risks of novel real estate assets cannot be effectively evaluated, real estate asset data from nonstandard input channels can be effectively processed, and accurate risk evaluation of the novel real estate assets is facilitated. In a first aspect, the present invention provides a method for processing risk data of an intelligent asset management system, including the steps of: s1, acquiring asset data of a target real estate; s2, carrying out standardized processing on the target property asset data to obtain standard data; S3, based on the standard data, performing multidimensional association matching, and calculating association confidence between the target property asset data and pre-stored core property archive data; S4, when the association confidence degree meets a preset threshold value condition, establishing logic association between the target property asset data and the core property archive data. The risk data processing method of the intelligent asset management system can effectively process real estate asset data from a non-standard input channel, solves the problem that a traditional system cannot identify and associate non-standard data through standardized processing and multidimensional association matching, and accordingly can accurately evaluate the risk of novel real estate assets and overcomes the defects of limitation of data processing capability and inaccurate risk evaluation in the prior art. In a second aspect, the present invention provides a smart asset management system risk data processing system comprising: The acquisition module is used for acquiring the target property asset data; the standardized processing module is used for carrying out standa