Management methods and systems for affordable housing applications in smart city based on internet of things

Abstract

This present disclosure provides a management method and system for affordable housing application. The method is executed by a management platform, and includes obtaining a query request for a risk of the affordable housing application from the service platform, wherein the query request is generated by the user platform based on an input of the affordable housing application by an applicant through the user platform; in response to the query request, through a management sub-platform corresponding to the management platform and a service sub-platform corresponding to the service platform, obtaining relevant information of the applicant and his/her related persons thereof, and determining the risk of the affordable housing application of the applicant; and sending the risk of the affordable housing application to the service platform.

What is claimed is: 
     
         1 . A management method for an affordable housing application, which is realized by a risk assessment system for an affordable housing application, the risk assessment system comprising a user platform, a service platform and a management platform, the management method being executed by the management platform, and the management method comprising:
 obtaining a query request for a risk of the affordable housing application from the service platform, wherein the query request is generated by the user platform based on an input of the affordable housing application by an applicant through the user platform;   in response to the query request, through a management sub-platform corresponding to the management platform and a service sub-platform corresponding to the service platform, obtaining relevant information of the applicant and related persons thereof, and determining the risk of the affordable housing application of the applicant;   the management platform adopting different management sub-platforms for data storage, data processing and/or data transmission; wherein the different management sub-platforms correspond to different information sources; and   sending the risk of the affordable housing application to the service platform.   
     
     
         2 . The management method of  claim 1 , wherein the related information includes at least one of housing information, life consumption information, public transportation information, loan information, and basic information. 
     
     
         3 . The management method of  claim 2 , wherein the public transportation information includes: a public transportation riding situation when a road section is congested. 
     
     
         4 . The management method of  claim 2 , wherein the life consumption information includes life track information, and the method further comprises
 generating a life feature vector based on the life track information, wherein the life feature vector includes at least one of a number of times to go to each type of place, average stay time at each type of place, and a proportion of high-consumption places.   
     
     
         5 . The management method of  claim 2 , wherein the life consumption information further comprises consumption situations of high consumption places. 
     
     
         6 . The management method of  claim 1 , wherein the obtaining the relevant information of the applicant and the related persons thereof, and determining the risk of the affordable housing application of the applicant comprises
 inputting the relevant information of the applicant and the related persons thereof into a risk prediction model to determine the risk of the affordable housing application, wherein the risk prediction model is a machine learning model.   
     
     
         7 . The management method of  claim 6 , wherein the risk prediction model comprises a graph neural network model, and the inputting the relevant information of the applicant and the related persons thereof into the risk prediction model to determine the risk of the affordable housing application comprises:
 inputting an associated information graph into the graph neural network model, and determining the risk of the affordable housing application of the applicant based on an output of a node corresponding to the applicant; wherein the associated information graph is obtained based on the relevant information of the applicant and his/her related persons.   
     
     
         8 . The management method of  claim 7 , wherein the node of the associated information graph includes object nodes and feature nodes, the object nodes correspond to the applicant and the related persons thereof, and the feature nodes correspond to the relevant information of the applicant and the related persons thereof;
 an edge of the associated information graph includes a first-type edge and a second-type edge, the first-type edge is used to connect the object nodes, and a feature of the first-type edge is a life similarity between the connected object nodes;   the second-type edge is used to connect the object nodes and the feature nodes corresponding to the object nodes, and a feature of the second-type edge is a feature value of each of the connected feature nodes.   
     
     
         9 . The management method of  claim 8 , wherein the feature nodes comprise: at least one of an age node, an occupation node, a public transportation node, a loan node, and a life consumption node;
 the age node includes nodes of a plurality of preset age groups, and the occupation node includes nodes of a plurality of preset occupations.   
     
     
         10 . The management method of  claim 8 , wherein the input of the risk prediction model further comprises a confidence level, the confidence level being related to a number of nodes and edges included in the associated information graph. 
     
     
         11 . The management method of  claim 10 , wherein the number of nodes and edges included in the associated information graph is obtained by weighted summation based on a hopping relationship between other nodes and edges and an applicant node. 
     
     
         12 . The management method of  claim 11 , wherein the farther a node or an edge of the weighted summation is from the applicant node, the smaller a weight of the node or the edge. 
     
     
         13 . The management method of  claim 10 , further comprising:
 using the confidence level as features of the object nodes of the associated information graph.   
     
     
         14 . A management system for an affordable housing application, comprising a user platform, a service platform, and a management platform;
 the service platform is used to obtain a query request for a risk of the affordable housing application from the service platform, wherein the query request is generated by the user platform based on an input of the affordable housing application by an applicant through the user platform;   the management platform is used to, in response to the query request, through a management sub-platform corresponding to the management platform and a service sub-platform corresponding to the service platform, obtain relevant information of the applicant and his/her related persons thereof, and determine the risk of the affordable housing application of the applicant; the management platform adopting different management sub-platforms for data storage, data processing and/or data transmission; wherein the different management sub-platforms correspond to different information sources; and   send the risk of the affordable housing application to the service platform.   
     
     
         15 . The management system of  claim 14 , the management platform is further configured to
 input the relevant information of the applicant and related persons thereof into a risk prediction model to determine the affordable housing application risk, wherein the risk prediction model is a machine learning model.   
     
     
         16 . The management system of  claim 14 , wherein the risk prediction model comprises a graph neural network model, and the management platform is further configured to:
 input an associated information graph into the graph neural network model, and determining the risk of the affordable housing application of the applicant based on an output of a node corresponding to the applicant; wherein the associated information graph is obtained based on the relevant information of the applicant and the related persons thereof.   
     
     
         17 . The management system of  claim 16 , wherein the node of the associated information graph includes object nodes and feature nodes, the object nodes correspond to the applicant and the related persons thereof, and the feature nodes correspond to the relevant information of the applicant and the related persons thereof;
 an edge of the associated information graph includes a first-type edge and a second-type edge, the first-type edge is used to connect the object nodes, and a feature of the first-type edge is a life similarity between the connected object nodes;   the second-type edge is used to connect the object nodes and the feature nodes corresponding to the object nodes, and a feature of the second-type edge is a feature value of each of the connected feature nodes.   
     
     
         18 . The management system of  claim 17 , wherein the feature node comprises: at least one of an age node, an occupation node, a public transportation node, a loan node, and a life consumption node;
 the age node includes nodes of a plurality of preset age groups, and the occupation node includes nodes of a plurality of preset occupations.   
     
     
         19 . The management system of  claim 17 , wherein the input of the risk prediction model further comprises a confidence level, the confidence level being related to a number of nodes and edges included in the associated information graph. 
     
     
         20 . A non-transitory computer-readable storage medium, comprising a set of instructions, wherein when executed by at least one processor, the management method of  claim 1  is implemented.