CN-121998415-A - Building economic risk monitoring method, electronic equipment and storage medium

CN121998415ACN 121998415 ACN121998415 ACN 121998415ACN-121998415-A

Abstract

The invention provides a building economic risk monitoring method, electronic equipment and a storage medium, wherein the method comprises the steps of determining a first target main body, a second target main body and a third target main body according to the characteristic data integrity of main body characteristic data corresponding to each target main body included in a target building; the method comprises the steps of determining a main body negative influence value of a first target main body according to a plurality of main body characteristic data corresponding to the first target main body, determining a main body negative influence value of a second target main body according to the influence type of a characteristic field of the main body characteristic data of the second target main body, determining a main body negative influence value corresponding to a third target main body according to the similarity of main body characteristic data of the same type of main body and main body characteristic data of the third target main body, and weighting the main body negative influence value of the target main body and the building negative influence value to obtain a building economic risk value so as to reflect real risk conditions between a target building and the target main body through the building economic risk value.

Inventors

FANG YAWEN
Yu Congzhe
LIU ZHIMIN
ZHANG CHUN
SUN LINGYAN
GU HAO

Assignees

上海邮电设计咨询研究院有限公司

Dates

Publication Date: 20260508
Application Date: 20260112

Claims (10)

1. A method for monitoring building economic risk, comprising: Step S100, determining the completeness of the characteristic data corresponding to each target main body according to the quantity of the main body characteristic data corresponding to each target main body included in the target building; Step 200, traversing a plurality of target subjects according to the characteristic data integrity corresponding to each target subject, determining a target subject with the characteristic data integrity greater than or equal to a preset first integrity threshold as a first target subject, determining a target subject with the characteristic data integrity smaller than the preset first integrity threshold and greater than a preset second integrity threshold as a second target subject, and determining a target subject with the characteristic data integrity smaller than or equal to the preset second integrity threshold as a third target subject; step S300, carrying out feature analysis on each first target main body according to a plurality of main body feature data corresponding to each first target main body so as to obtain a main body negative influence value corresponding to each first target main body; Step S400, determining an adjustment coefficient corresponding to each second target main body according to the influence type of the characteristic field of each main body characteristic data corresponding to each second target main body; Step S500, processing and analyzing the adjustment coefficient corresponding to each second target main body and the main body characteristic data corresponding to each second target main body to obtain a main body negative influence value corresponding to each second target main body; step S600, determining a main body negative impact value corresponding to each third target main body according to the similarity between the main body feature data of the same type of main body as the main body type of each third target main body and the main body feature data of each third target main body, and the main body negative impact value corresponding to the same type of main body; And step S700, weighting the main negative influence value corresponding to each target main body and the preset building negative influence value corresponding to the target building to obtain the building economic risk value corresponding to the target building.
2. The method according to claim 1, wherein the step S100 comprises: Step S110, obtaining main body characteristic data corresponding to each target main body included in the target building to obtain a plurality of main body characteristic data lists A 1 ,A 2 ,...,A j ,...,A k , wherein j=1, 2, K is the number of target main bodies included in the target building, A j is a main body characteristic data list corresponding to a jth target main body included in the target building; A j =(A j1 ,A j2 ,...,A jm ,...,A jn(j) ), m=1, 2, & gt, n (j) being the number of acquired subject feature data of the jth target subject, a jm being the mth subject feature data of the jth target subject; Step S120, traversing a j , and obtaining a preset field weight of a feature field corresponding to each main body feature data in a j , so as to obtain a preset field weight of a feature field corresponding to a jm from a field weight list B j =(B j1 ,B j2 ,...,B jm ,...,B jn(j) );B jm corresponding to a jth target main body; and step S130, determining the feature data integrity C j =∑ n(j) m=1 B jm corresponding to the jth target main body.
3. The method according to claim 2, wherein the step S300 includes: Step S310, a preset initial feature vector D= (D 1 ,D 2 ,...,D g ,...,D h ) is obtained, wherein g=1, 2,., h is the number of feature data included in the initial feature vector, and D g is the g-th feature data included in the initial feature vector; step S320, determining any one of the first target subjects as a first key subject; Step S330, performing feature encoding on a plurality of main body feature data corresponding to the first key main body to obtain a first feature vector E= (E 1 ,E 2 ,...,E p ,...,E q ), wherein p=1, 2, q is the number of main body feature data of the first key main body, and E p is feature data obtained by performing feature encoding on the p-th main body feature data of the first key main body; Step S340, traversing the initial feature vector D, and if the feature field corresponding to D g is the same as the feature field corresponding to E p , replacing D g in the initial feature vector D with E p ; If the feature field corresponding to the D g is different from the feature field corresponding to any one of the feature data in the first feature vector E, setting D g in the initial feature vector D to be a null value, so as to obtain a target feature vector corresponding to the first key body; Step S350, inputting the target feature vector corresponding to the first key principal into a preset influence value determination model to obtain a principal negative influence value corresponding to the first key principal output by the influence value determination model; The influence value determining model is obtained by training the historical characteristic data corresponding to the target main bodies in the historical time period.
4. A method according to claim 3, wherein the influence value determination model is determined according to the steps of: step S351, performing feature coding on a plurality of historical feature data corresponding to each target main body in a historical period to obtain a plurality of historical coding data corresponding to each target main body, wherein the duration of the historical period is a preset duration, and the ending time of the historical period is before the current time; step S352, integrating the plurality of historical encoding data corresponding to each target subject into the initial feature vector D according to the corresponding feature field, so as to obtain a historical feature vector corresponding to each target subject; step S353, acquiring a preset history negative impact value corresponding to each target subject in a history period; Step S354, taking the historical feature vector corresponding to each target subject as an input sample, taking the historical negative impact value corresponding to each target subject as an output label, and performing supervised learning training on a preset initial large language model to obtain an impact value determination model.
5. The method according to claim 4, wherein the step S400 includes: step S410, determining any one of the second target subjects as a second key subject; Step S420, performing feature encoding on a plurality of main body feature data corresponding to the second key main body to obtain a second feature vector F= (F 1 ,F 2 ,...,F a ,...,F b ), wherein a=1, 2, b is the number of main body feature data of the second key main body, and F a is feature data obtained by performing feature encoding on the a-th main body feature data of the second key main body; Step S430, traversing the initial feature vector D, and if the feature field corresponding to the D g is the same as the feature field corresponding to the F a , replacing the D g in the initial feature vector D with the F a ; If the feature field corresponding to the D g is different from the feature field corresponding to any one of the feature data in the second feature vector F, setting D g in the initial feature vector D to be a null value, so as to obtain a target feature vector corresponding to the second key body; Step S440, traversing the target feature vector corresponding to the second key main body, and determining the feature data with the null value as missing feature data; Step S450, determining an adjustment coefficient corresponding to the second key main body G=((H 1 -I 1 )×J 1 +(H 2 -I 2 )×J 2 +(H 3 -I 3 )×J 3 )/(1-(I 1 +I 2 +I 3 )); Wherein H 1 is the sum of preset field weights of the feature fields corresponding to the feature data of the initial feature vector D, which affects the feature fields of the type positive, H 2 is the sum of preset field weights of the feature fields corresponding to the feature data of the initial feature vector D, which affects the feature fields of the type neutral, H 3 is the sum of preset field weights of the feature fields corresponding to the feature data of the initial feature vector D, which affects the feature fields of the type negative, H 1 +H 2 +H 3 =1; I 1 is the sum of preset field weights of a plurality of missing feature data of the target feature vector corresponding to the second key main body, wherein the influence type is the sum of preset field weights of a positive type of feature fields, I 2 is the sum of preset field weights of a plurality of missing feature data of the target feature vector corresponding to the second key main body, the influence type is the neutral type of feature fields, and I 3 is the sum of preset field weights of a plurality of missing feature data of the target feature vector corresponding to the second key main body; J 1 is the weight coefficient of the preset positive characteristic field, J 2 is the weight coefficient of the preset neutral characteristic field, J 3 is the weight coefficient of the preset negative characteristic field, and J 1 ＜1;J 2 =1;J 3 >1.
6. The method according to claim 5, wherein the step S500 includes: Step S510, inputting the target feature vector corresponding to the second key body into the influence value determining model to obtain an initial negative influence value M corresponding to the second key body output by the influence value determining model; Step S520, determining a main body negative impact value l=m×g corresponding to the second key main body.
7. The method according to claim 6, wherein the step S600 includes: Step S610, determining any one of the third target subjects as a third key subject; step S620, determining the key main body with the same main body type as the third key main body as the main body of the same type in a plurality of preset key main bodies, wherein the key main body is the main body with the characteristic data integrity larger than the first integrity threshold; step 630, traversing the feature fields of the main body feature data of the same type of main body and the feature fields of the main body feature data of the third key main body, and determining the feature fields of the main body of the same type and the main body of the third key main body which correspond together as common fields; Step S640, comparing the similarity of the main body feature data of the common field corresponding to the same type of main body with the main body feature data of the common field corresponding to the third key main body to obtain the feature similarity corresponding to the same type of main body; Step S650, if the feature similarity corresponding to the same type of main body is greater than a preset similarity threshold, determining the same type of main body as a target same type of main body; Step S660, obtaining a preset negative influence value of each target type of main body corresponding to the third key main body to obtain an influence value list N= (N 1 ,N 2 ,...,N c ,...,N d ), wherein c=1, 2, d is the number of target type of main bodies corresponding to the third key main body, and N c is the negative influence value of the c-th target type of main body corresponding to the third key main body; step S670, obtaining the feature similarity between the third key main body and each corresponding target type main body to obtain a feature similarity list T= (T 1 ,T 2 ,...,T c ,...,T d ), wherein T c is the feature similarity between the third key main body and the corresponding c-th target type main body; Step S680, determining a negative impact value r= Σ d c=1 (T c /(∑ d c=1 T c ))×N c of the principal corresponding to the third key principal.
8. The method according to claim 7, wherein the step S700 includes: Step 710, acquiring a preset building negative influence value V corresponding to the target building; Step S720, determining a building economic risk value z=y 1 ×V+Y 2 ×q corresponding to the target building; Wherein Q is the sum of a plurality of main body negative influence values corresponding to the target main body, Y 1 is a preset first risk value coefficient, Y 2 is a preset second risk value coefficient, and Y 1 +Y 2 =1; Y 2 =(S 1 /Q)×W 1 +(S 2 /Q)×W 2 +(S 3 /Q)×W 3 ; S 1 is the sum of the negative impact values of the main bodies corresponding to the first target main bodies, S 2 is the sum of the negative impact values of the main bodies corresponding to the second target main bodies, S 3 is the sum of the negative impact values of the main bodies corresponding to the third target main bodies, S 1 +S 2 +S 3 =q; W 1 is a preset first influence value coefficient, W 2 is a preset second influence value coefficient, W 3 is a preset third influence value coefficient, and 0<W 3 ＜W 2 ＜W 1 <1.
9. A non-transitory computer readable storage medium having stored therein at least one instruction or at least one program loaded and executed by a processor to implement the method of any one of claims 1-8.
10. An electronic device comprising a processor and the non-transitory computer readable storage medium of claim 9.

Description

Building economic risk monitoring method, electronic equipment and storage medium Technical Field The invention relates to the field of building economic monitoring, in particular to a building economic risk monitoring method, electronic equipment and a storage medium. Background In the age of rapid economic development of the current day, building economy is becoming an emerging economic form and is becoming an important driving force for urban economic development. The building economy takes business buildings, functional plates and regional facilities as main carriers, and various enterprises are introduced through developing and renting the building, so that tax sources are introduced, and regional economy development is driven. However, as building economies continue to develop, the risks that they face are becoming increasingly prominent, and businesses and investors want to be able to accurately understand the business risk profile of a resident building in order to make reasonable investment decisions. Meanwhile, an administrator of the building needs an effective risk monitoring method to improve the operation management level of the building so as to guarantee the benefits of tenants. Therefore, an effective risk monitoring and early warning method is urgently needed to ensure the healthy and stable development of building economy. Disclosure of Invention Aiming at the technical problems, the invention adopts the following technical scheme: According to one aspect of the present application, there is provided a building economic risk monitoring method comprising: step S100, determining the completeness of the characteristic data corresponding to each target main body according to the quantity of the main body characteristic data corresponding to each target main body included in the target building; Step 200, traversing a plurality of target subjects according to the feature data integrity corresponding to each target subject, determining the target subject with the feature data integrity greater than or equal to a preset first integrity threshold as a first target subject, determining the target subject with the feature data integrity smaller than the preset first integrity threshold and greater than a preset second integrity threshold as a second target subject, and determining the target subject with the feature data integrity smaller than or equal to the preset second integrity threshold as a third target subject; Step S300, carrying out feature analysis on each first target main body according to a plurality of main body feature data corresponding to each first target main body so as to obtain a main body negative influence value corresponding to each first target main body; Step S400, according to the influence type of the characteristic field of each main body characteristic data corresponding to each second target main body, determining an adjustment coefficient corresponding to each second target main body; step S500, processing and analyzing the adjustment coefficient corresponding to each second target main body and the main body characteristic data corresponding to each second target main body to obtain a main body negative influence value corresponding to each second target main body; Step S600, determining a main body negative influence value corresponding to each third target main body according to the similarity between the main body characteristic data of the same type of main body as that of each third target main body and the main body characteristic data of each third target main body and the main body negative influence value corresponding to the same type of main body; And step S700, weighting the main negative influence value corresponding to each target main body and the preset building negative influence value corresponding to the target building to obtain the building economic risk value corresponding to the target building. According to another aspect of the present application, there is provided a non-transitory computer readable storage medium having stored therein at least one instruction or at least one program loaded and executed by a processor to implement the aforementioned building economic risk monitoring method. According to yet another aspect of the present application, there is provided an electronic device comprising a processor and the aforementioned non-transitory computer-readable storage medium. The invention has at least the following beneficial effects: According to the building economic risk monitoring method, firstly, the feature data integrity corresponding to each target main body is determined according to the quantity of main body feature data corresponding to each target main body included in a target building, the feature data integrity represents the integrity degree of the main body feature data of the obtained target main body, the target main body with the feature data integrity degree smaller than a first integrity degree threshold and lar