CN117876103A - Method and system for setting up pedestrian credit investigation user picture - Google Patents
Method and system for setting up pedestrian credit investigation user picture Download PDFInfo
- Publication number
- CN117876103A CN117876103A CN202410278375.8A CN202410278375A CN117876103A CN 117876103 A CN117876103 A CN 117876103A CN 202410278375 A CN202410278375 A CN 202410278375A CN 117876103 A CN117876103 A CN 117876103A
- Authority
- CN
- China
- Prior art keywords
- credit
- features
- user
- credibility
- dimensions
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 238000011835 investigation Methods 0.000 title description 3
- 238000012545 processing Methods 0.000 claims abstract description 23
- 238000007499 fusion processing Methods 0.000 claims abstract description 12
- 238000011156 evaluation Methods 0.000 claims description 27
- 238000012937 correction Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 238000012795 verification Methods 0.000 claims description 4
- 238000012800 visualization Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Landscapes
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention provides a method and a system for setting up a pedestrian credit user image, which belong to the technical field of data processing and specifically comprise the following steps: extracting entity information from different information modules of the user's pedestrian credit data, extracting credit features of different credit dimensions based on the entity information, determining the credibility of the user in the different credit dimensions by checking the data, taking the credit dimensions with the credibility not meeting the requirement as suspected risk dimensions, acquiring the completeness of the credit features in the different credit dimensions, and combining the credibility in the different credit dimensions to determine that the user's pedestrian credit data is reliable, and carrying out fusion processing according to the credit features in the different credit dimensions and semantic networks among the credit features to obtain the user portrait of the user, thereby reducing the risk of credit application processing.
Description
Technical Field
The invention belongs to the technical field of data processing, and particularly relates to a method and a system for setting up a pedestrian credit subscriber image.
Background
The pedestrian credit report is an important data source for user portrait construction by a financial institution, and rich identity, occupation and living information is recorded. However, compared with fragments of report information, such as home address, communication address, residence address and unit address, the user address is marked, and the conclusion integrity problem exists in the traditional method of directly analyzing and obtaining independent labels; and because the pedestrian information is acquired and filled manually, and is not strictly verified, module conflict can exist, and the accuracy is difficult to guarantee by direct application.
In order to solve the technical problems, the prior patent CN202010711327.5 (SDK system and processing method thereof based on credit message variable processing and customer portrait) provides a method for analyzing pedestrian credit message information and outputting credit derived variable and customer portrait code according to analysis result, but conflict conditions and alignment processing among modules of different pedestrian information are not considered yet, so that the accuracy of the generated customer portrait is difficult to meet the requirements.
Aiming at the technical problems, the invention provides a method and a system for setting up a pedestrian credit subscriber image.
Disclosure of Invention
In order to achieve the purpose of the invention, the invention adopts the following technical scheme:
according to one aspect of the invention, a method for setting up a human credit investigation user image is provided.
The method for constructing the pedestrian credit user image is characterized by comprising the following steps of:
s1, extracting entity information from different information modules of pedestrian credit information data of a user, and extracting credit characteristics of different credit dimensions based on the entity information;
s2, constructing a semantic network among different credit features, performing alignment checking processing on the different credit features through the semantic network to obtain checking data of the credible features of different credit dimensions, determining the credibility of the user under the different credit dimensions through the checking data, and entering the next step when the credit dimensions with the credibility not meeting the requirements exist;
s3, taking the credit dimension with the reliability not meeting the requirement as a suspected risk dimension, determining the comprehensive reliability of pedestrian credit data according to the number of the suspected risk dimensions and the reliability of different credit dimensions, and entering the next step when the comprehensive reliability meets the requirement;
s4, acquiring the integrity of credit features in different credit dimensions, and when the reliability of the credit feature data of the user is determined by combining the reliability in different credit dimensions, carrying out fusion processing according to the credit features in different credit dimensions and semantic networks among the credit features to obtain the user portrait of the user.
The invention has the beneficial effects that:
1. the credibility of the user under different credit dimensions is determined by checking the data, the difference of the credibility of different credit features due to the difference of semantic conflict situations and consistency situations with other credit features is fully considered, the accurate evaluation of the credibility of different credit dimensions is realized, and a foundation is laid for carrying out the generation of user portraits of the user in a differentiated manner and the screening of non-credible users.
2. The comprehensive credibility of the pedestrian credit data is determined according to the number of the suspected risk dimensions and the credibility of different credit dimensions, so that the screening of the unreliable users from the angles of a plurality of credit dimensions is realized, and the technical problem of lower processing efficiency caused by processing the user portrait of the user with the unreliable pedestrian credit data is avoided.
3. And carrying out fusion processing according to the credit characteristics under different credit dimensions and semantic networks among the credit characteristics to obtain user portraits of the users, namely taking the difference of the credit characteristics under the different credit dimensions into consideration, and simultaneously, enabling the generation processing result of the user portraits to more accurately reflect the real credit risk of the users through the identification of the semantic networks, thereby laying a foundation for further effectively controlling the credit risk.
The credit module comprises user identity, occupation, living information and historical credit data.
The credit dimension comprises user identity information, credit overdue information, credit application information, credit support information and credit inquiry information.
When the comprehensive credibility does not meet the requirement, determining that the credibility of the pedestrian credit information data of the user does not meet the requirement, and directly outputting credit risk without constructing a user portrait of the user.
The credit feature integrity is determined according to the ratio of the number of credit features of the pedestrian credit data of the user in the credit dimension to the number of preset credit features in the credit dimension.
In a second aspect, the present invention provides a computer system comprising: a communicatively coupled memory and processor, and a computer program stored on the memory and capable of running on the processor, characterized by: and executing the method for constructing the pedestrian credit subscriber image by the processor when the processor runs the computer program.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the invention as set forth hereinafter.
In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
The above and other features and advantages of the present invention will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings;
FIG. 1 is a flow chart of a method of constructing a representation of a pedestrian credit subscriber;
FIG. 2 is a flow chart of a method of determining a user's trustworthiness in the credit dimension;
FIG. 3 is a flow chart of a method of determining the overall confidence of pedestrian credit data;
FIG. 4 is a flow chart of a method of determining that user's pedestrian credit data is reliable;
FIG. 5 is a block diagram of a computer system.
Detailed Description
In order to make the technical solutions in the present specification better understood by those skilled in the art, the technical solutions in the embodiments of the present specification will be clearly and completely described below with reference to the drawings in the embodiments of the present specification, and it is obvious that the described embodiments are only some embodiments of the present specification, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present disclosure.
The pedestrian credit data is used as a key reference means for credit application processing of a financial institution, analysis of the pedestrian credit data is realized, construction of a user portrait is carried out, and the control of credit risk of the financial institution is important, but because credit features among different modules of the pedestrian credit data possibly have places with semantic conflict, such as residence addresses, working addresses, communication addresses and the like, the requirement of the portrait construction processing efficiency and the portrait accuracy cannot be met if whether the construction of the user portrait based on the pedestrian credit data is needed cannot be determined according to the recognition result of the credit features with the semantic conflict of the pedestrian credit data.
In order to solve the technical problems, the invention adopts the duty ratio of suspected problem features with semantic conflict in credit features of different credit dimensions of pedestrian credit data to determine the credibility of the different credit dimensions, and determines the comprehensive credibility of the pedestrian credit data according to the number duty ratio of the credit dimensions with different credibility not meeting the requirements, when the comprehensive credibility meets the requirements, the invention carries out fusion processing according to the credit features under the different credit dimensions and semantic networks among the credit features to obtain the user portrait of the user, wherein the credit dimensions comprise user identity information, overdue credit information, credit application information, credit branch information and credit inquiry information.
Specifically, the invention adopts the following technical scheme:
firstly, extracting entity information from different information modules of pedestrian credit information data of a user, and extracting credit characteristics of different credit dimensions according to the entity information;
then constructing semantic networks among different credit features, performing alignment checking processing on the different credit features through the semantic networks to obtain checking data of the credit features with different credit dimensions, determining the credibility of the user under the different credit dimensions through the checking data, specifically determining the credibility of the different credit dimensions through the occupation ratio of suspected problem features with semantic conflicts in the credit features with different credit dimensions of pedestrian credit data, entering the next step when the credit dimension with lower credibility exists, and performing fusion processing according to the credit features with different credit dimensions and the semantic networks among the credit features when the credit dimension with lower credibility does not exist to obtain a user portrait of the user;
taking the credit dimension with the reliability not meeting the requirement as a suspected risk dimension, determining the comprehensive reliability of pedestrian credit data according to the number of the suspected risk dimensions and the reliability of different credit dimensions, specifically determining the comprehensive reliability of the pedestrian credit data by multiplying the number of the suspected risk dimensions by the average value of the reliability of different credit dimensions, entering the next step when the comprehensive reliability is larger, and determining that the reliability of the pedestrian credit data of the user is not meeting the requirement when the comprehensive reliability is smaller, without constructing a user portrait of the user, and directly outputting the credit giving risk;
and finally, determining whether the pedestrian credit data of the user is reliable or not according to the credit feature integrity and the credibility under different credit dimensions, specifically determining whether the pedestrian credit data of the user is reliable or not according to the credit feature integrity and the credibility under different credit dimensions and the average value of the weight sum of the credibility, when the pedestrian credit data is reliable, performing visualization processing according to the credit features under different credit dimensions and the semantic network among the credit features to obtain the user portrait of the user under different credit dimensions, and performing fusion processing on the user portrait of the user under different credit dimensions to obtain the user portrait of the user, and when the pedestrian credit data of the user is unreliable, determining that the credibility of the pedestrian credit data of the user does not meet the requirement, and directly outputting the credit risk without constructing the user portrait of the user.
Further explanation will be made below from two perspectives of the method class embodiment and the system class embodiment.
In order to solve the above-mentioned problems, according to one aspect of the present invention, as shown in fig. 1, there is provided a method for constructing a pedestrian credit user image, which is characterized by comprising:
s1, extracting entity information from different information modules of pedestrian credit information data of a user, and extracting credit characteristics of different credit dimensions based on the entity information;
specifically, the credit module comprises user identity, occupation, residence information and historical credit data.
It should be noted that the credit dimension includes user identity information, credit overdue information, credit application information, credit support information, and credit inquiry information.
S2, constructing a semantic network among different credit features, performing alignment checking processing on the different credit features through the semantic network to obtain checking data of the credible features of different credit dimensions, determining the credibility of the user under the different credit dimensions through the checking data, and entering the next step when the credit dimensions with the credibility not meeting the requirements exist;
further, the semantic network among the credit features is determined according to semantic logic association relations among the credit features, wherein the semantic logic association relations are determined according to matching results of the credit features.
It will be appreciated that the verification data includes the number of other credit features used for the credit feature verification and the number of other credit features that conflict with the semantics of the credit features.
In a possible embodiment, as shown in fig. 2, the method for determining the credibility of the user in the credit dimension in step S1 is as follows:
determining that credit features of a semantic network do not exist with other credit features in the credit dimension according to the checking data, taking the credit features as unchecked features, and determining the basic credibility of the user in the credit dimension according to the number of unchecked features and the number ratio of the credit features in the credit dimension;
taking the credit features which have no semantic conflict with other credit features in the credit dimension as the credit features, and determining the credit feature correction amount of the user in the credit dimension according to the number of the other credit features, the number of the credit features and the number ratio of the credit features in the credit dimension, wherein the other credit features are associated with different credit features;
taking the credit features which have semantic conflict with other credit features in the credit dimension as the unreliable credit features, and determining the feature credibility of the different unreliable credit features according to the number of the other credit features which are associated with the different unreliable credit features, the number of the other credit features which have semantic conflict with the unreliable credit features and the number of the other credit features which have no semantic conflict;
and determining deviation feature correction quantity of the user in the credit dimension according to the quantity of different unreliable credit features, the feature credibility and the quantity ratio of the credit features in the credit dimension, and determining the credibility of the user in the credit dimension by combining the credibility feature correction quantity and the basic credibility.
In another embodiment, the method for determining the credibility of the user in the credit dimension in step S1 is as follows:
determining that the credit features of the semantic network do not exist with other credit features in the credit dimension according to the checking data, taking the credit features as unchecked features, and taking the credit features which have semantic conflict with other credit features in the credit dimension as unreliable credit features;
when the unverified features and the unreliable credit features do not exist in the credit dimension, determining the credibility of the user in the credit dimension through the number of the credit features of the user in the credit dimension;
when there are unchecked features or untrusted credit features in the credit dimension:
determining a base trustworthiness of the user in the credit dimension by a number of unverified features and a number of credit features in the credit dimension;
when the unreliable credit features do not exist in the credit dimension, judging whether the quantity proportion of the unverified features is larger than a preset quantity proportion, if so, determining the credibility of the user in the credit dimension through the basic credibility of the user in the credit dimension, and if not, determining the credibility of the user in the credit dimension through the quantity of the credit features of the user in the credit dimension;
when there is an untrusted credit feature in the credit dimension:
taking the credit features which have semantic conflict with other credit features in the credit dimension as the unreliable credit features, and determining the feature credibility of the different unreliable credit features according to the number of the other credit features which are associated with the different unreliable credit features, the number of the other credit features which have semantic conflict with the unreliable credit features and the number of the other credit features which have no semantic conflict;
and determining deviation feature correction quantity of the user in the credit dimension according to the quantity of different unreliable credit features, feature credibility and the quantity ratio of the credit features in the credit dimension, and determining the credibility of the user in the credit dimension by combining the quantity of the credit features and the basic credibility.
In another embodiment, the method for determining the credibility of the user in the credit dimension in step S2 is as follows:
s21, taking the credit characteristics which have semantic conflict with other credit characteristics in the credit dimension as the unreliable credit characteristics, judging whether the unreliable credit characteristics exist in the credit dimension, if so, entering the next step, and if not, entering the step S24;
s22, judging whether the number of the unreliable credit features in the credit dimension is larger than the preset feature number, if so, determining the credibility of the user in the credit dimension through the number of the unreliable credit features;
s23, determining feature credibility of different unreliable credit features according to the number of other credit features associated with the different unreliable credit features, the number of other credit features having semantic conflict with the unreliable credit features and the number of other credit features having no semantic conflict, determining deviation feature correction quantity of the user in the credit dimension according to the number of different unreliable credit features, the feature credibility and the number proportion of the credit features in the credit dimension, judging whether the deviation feature correction quantity meets the requirement, if yes, entering a next step, and if no, determining the credibility of the user in the credit dimension through the deviation feature correction quantity;
s24, determining that credit features of a semantic network do not exist with other credit features in the credit dimension according to the checking data, taking the credit features as unchecked features, and determining the basic credibility of the user in the credit dimension according to the number of unchecked features and the number ratio of the credit features in the credit dimension;
s25, taking the credit features which have no semantic conflict with other credit features in the credit dimension as the credit features, determining the credit feature correction quantity of the user in the credit dimension according to the quantity of the other credit features, the quantity of the credit features and the quantity ratio of the credit features in the credit dimension, and determining the credibility of the user in the credit dimension by combining the deviation feature correction quantity and the basic credibility.
Further, the confidence level of the user in different credit dimensions ranges from 0 to 1, and when the confidence level of the user in the credit dimensions is smaller than a preset confidence level, the confidence level of the credit dimensions is determined to not meet the requirement.
S3, taking the credit dimension with the reliability not meeting the requirement as a suspected risk dimension, determining the comprehensive reliability of pedestrian credit data according to the number of the suspected risk dimensions and the reliability of different credit dimensions, and entering the next step when the comprehensive reliability meets the requirement;
when the user does not have the credibility dimension which does not meet the requirement, the user portrait of the user is obtained directly according to the credit characteristics under different credit dimensions and the semantic network among the credit characteristics.
In a possible embodiment, as shown in fig. 3, the method for determining the comprehensive credibility of the pedestrian credit data in step S3 is as follows:
determining the credibility evaluation amount of the suspected risk dimension based on the number of the suspected risk dimensions and the credibility of different suspected risk dimensions;
taking the credit dimension with the suspected risk dimension removed as a credibility credit dimension, and determining the credibility evaluation amount of the credibility dimension based on the number of the credibility dimensions and the credibility of different credibility dimensions;
and obtaining the average value of the credibility of different credit dimensions, and determining the comprehensive credibility of the pedestrian credit data by combining the credibility evaluation value of the credibility risk dimension and the credibility evaluation value of the suspected risk dimension.
It can be understood that when the integrated credibility does not meet the requirement, the credibility of the pedestrian credit information data of the user is determined to be not met, the user portrait of the user is not required to be constructed, and the credit risk is directly output.
In another embodiment, the method for determining the comprehensive credibility of the pedestrian credit data in step S3 is as follows:
when the number of suspected risk dimensions of the pedestrian credit data is larger than the number of preset dimensions, determining that the comprehensive credibility of the pedestrian credit data does not meet the requirement;
acquiring the credibility of the suspected risk dimensions when the number of the suspected risk dimensions of the pedestrian credit data is larger than the number of preset dimensions, and determining that the comprehensive credibility of the pedestrian credit data does not meet the requirement when the number of the suspected risk dimensions with the credibility smaller than the credibility threshold is larger than the limit number of the preset dimensions;
when the number of the suspected risk dimensions with the credibility smaller than the credibility threshold is not larger than the limit number of the preset dimensions, determining the credibility evaluation amount of the suspected risk dimensions based on the number of the suspected risk dimensions and the credibility of different suspected risk dimensions, and when the credibility evaluation amount of the suspected risk dimensions does not meet the requirements, determining that the comprehensive credibility of the pedestrian credit data does not meet the requirements;
when the credibility evaluation amount of the suspected risk dimension meets the requirement, taking the credit dimension with the suspected risk dimension removed as a credibility dimension, and determining the credibility evaluation amount of the credibility dimension based on the number of the credibility dimensions and the credibility of different credibility dimensions;
and obtaining the average value of the credibility of different credit dimensions, and determining the comprehensive credibility of the pedestrian credit data by combining the credibility evaluation value of the credibility risk dimension and the credibility evaluation value of the suspected risk dimension.
In another embodiment, the method for determining the comprehensive credibility of the pedestrian credit data in step S3 is as follows:
s31, judging whether the average value of the credibility of different credit dimensions of the pedestrian credit data meets the requirement, if so, entering the next step, and if not, determining that the comprehensive credibility of the pedestrian credit data does not meet the requirement;
s32, judging whether the number of suspected risk dimensions of the pedestrian sign data is larger than the number of preset dimensions, if so, entering the next step, and if not, entering the step S34;
s33, judging whether the sum of the credibility of the suspected risk dimensions of the pedestrian credit data meets the requirement, if so, entering the next step, and if not, determining that the comprehensive credibility of the pedestrian credit data does not meet the requirement;
s34, taking a suspected risk dimension with the credibility smaller than a credibility threshold value as a problem risk dimension, determining the credibility evaluation amount of the suspected risk dimension according to the number of the problem risk dimension, the number of the suspected risk dimension and the credibility of different suspected risk dimensions, judging whether the credibility evaluation amount of the suspected risk dimension does not meet the requirement, if so, determining that the comprehensive credibility of the pedestrian credit data does not meet the requirement, and if not, entering the next step;
and S35, taking the credit dimension with the suspected risk dimension removed as a credibility dimension, determining the credibility evaluation amount of the credibility dimension based on the number of the credibility dimensions and the credibility of different credibility dimensions, acquiring the average value of the credibility of different credit dimensions, and determining the comprehensive credibility of pedestrian credit data by combining the credibility evaluation amount of the credibility dimension and the credibility evaluation amount of the suspected risk dimension.
S4, acquiring the integrity of credit features in different credit dimensions, and when the reliability of the credit feature data of the user is determined by combining the reliability in different credit dimensions, carrying out fusion processing according to the credit features in different credit dimensions and semantic networks among the credit features to obtain the user portrait of the user.
It should be noted that, the credit feature integrity is determined according to a ratio of the number of credit features of the user's pedestrian credit data in the credit dimension to the number of preset credit features in the credit dimension.
In a possible embodiment, as shown in fig. 4, the determining that the pedestrian credit data of the user in step S4 is reliable specifically includes:
determining reliability of dimension data in different credit dimensions according to the credibility in different credit dimensions and the integrity of credit features;
acquiring reliability and types of dimension data under different credit dimensions, determining the reliability of the pedestrian credit data of the user by combining the number and types of the credit dimensions of the pedestrian credit data of the user, and determining whether the pedestrian credit data of the user is reliable or not through the reliability.
In another embodiment, the determining in step S4 that the user' S pedestrian credit data is reliable specifically includes:
s41, determining the number of incomplete credit dimensions through credit feature integrality under different credit dimensions, judging whether the number of the incomplete credit dimensions meets the requirement, if so, entering the next step, and if not, determining that the pedestrian credit data of the user is unreliable;
s42, judging whether the sum of the number of the incomplete credit dimensions and the number of the suspected risk dimensions meets the requirement, if so, entering the next step, and if not, determining that the pedestrian credit data of the user is unreliable;
s43, determining the reliability of dimension data in different credit dimensions according to the reliability in the different credit dimensions and the integrity of the credit features, judging whether the credit dimensions with the reliability of the dimension data not meeting the requirements exist, if so, entering the next step, and if not, entering the step S46;
s44, taking the credit dimension of which the reliability of the dimension data does not meet the requirement as an unreliable credit dimension, judging whether the number of the unreliable credit dimensions meets the requirement, if so, entering the next step, and if not, determining that the pedestrian credit data of the user is unreliable;
s45, determining a reliable influence value of the unreliable credit dimension through the number of the unreliable credit dimension, the type of the unreliable credit dimension and the dimension data reliability of different unreliable credit dimensions, judging whether the reliable influence value of the unreliable credit dimension meets the requirement, if so, entering the next step, and if not, determining that the pedestrian credit data of the user is unreliable;
s46, acquiring reliability and types of dimension data under different credit dimensions, determining the reliability of the pedestrian credit data of the user by combining the quantity and types of the credit dimensions of the pedestrian credit data of the user and the reliability influence value of the unreliable credit dimensions, and determining whether the pedestrian credit data of the user is reliable or not through the reliability.
The user portrait construction method of the user includes:
and carrying out visualization processing according to the credit features in different credit dimensions and semantic networks among the credit features to obtain user portraits of the user in the different credit dimensions, and carrying out fusion processing on the user portraits of the user in the different credit dimensions to obtain the user portraits of the user.
In another aspect, as shown in FIG. 5, the present invention provides a computer system comprising: a communicatively coupled memory and processor, and a computer program stored on the memory and capable of running on the processor, characterized by: and executing the method for constructing the pedestrian credit subscriber image by the processor when the processor runs the computer program.
The method for constructing the image of the pedestrian credit subscriber specifically comprises the following steps:
extracting entity information from different information modules of pedestrian credit information data of a user, and extracting credit characteristics of different credit dimensions based on the entity information;
constructing semantic networks among different credit features, performing alignment checking processing on the different credit features through the semantic networks to obtain checking data of the credible features of different credit dimensions, determining the credibility of the user under the different credit dimensions through the checking data, and entering the next step when the credibility does not meet the required credit dimensions;
when the number of suspected risk dimensions of the pedestrian credit data is larger than the number of preset dimensions, determining that the comprehensive credibility of the pedestrian credit data does not meet the requirement;
acquiring the credibility of the suspected risk dimensions when the number of the suspected risk dimensions of the pedestrian credit data is larger than the number of preset dimensions, and determining that the comprehensive credibility of the pedestrian credit data does not meet the requirement when the number of the suspected risk dimensions with the credibility smaller than the credibility threshold is larger than the limit number of the preset dimensions;
when the number of the suspected risk dimensions with the credibility smaller than the credibility threshold is not larger than the limit number of the preset dimensions, determining the credibility evaluation amount of the suspected risk dimensions based on the number of the suspected risk dimensions and the credibility of different suspected risk dimensions, and when the credibility evaluation amount of the suspected risk dimensions does not meet the requirements, determining that the comprehensive credibility of the pedestrian credit data does not meet the requirements;
when the credibility evaluation amount of the suspected risk dimension meets the requirement, taking the credit dimension with the suspected risk dimension removed as a credibility dimension, and determining the credibility evaluation amount of the credibility dimension based on the number of the credibility dimensions and the credibility of different credibility dimensions;
acquiring the average value of the credibility of different credit dimensions, and determining the comprehensive credibility of the pedestrian credit data by combining the credibility evaluation value of the credibility risk dimension and the credibility evaluation value of the suspected risk dimension;
and when the credit feature integrity under different credit dimensions is obtained and the reliability under different credit dimensions is combined to determine that the pedestrian credit data of the user is reliable, carrying out visualization processing according to the credit features under different credit dimensions and semantic networks among the credit features to obtain user figures of the user under different credit dimensions, and carrying out fusion processing on the user figures of the user under different credit dimensions to obtain the user figures of the user.
Through the above embodiments, the present invention has the following beneficial effects:
1. the credibility of the user under different credit dimensions is determined by checking the data, the difference of the credibility of different credit features due to the difference of semantic conflict situations and consistency situations with other credit features is fully considered, the accurate evaluation of the credibility of different credit dimensions is realized, and a foundation is laid for carrying out the generation of user portraits of the user in a differentiated manner and the screening of non-credible users.
2. The comprehensive credibility of the pedestrian credit data is determined according to the number of the suspected risk dimensions and the credibility of different credit dimensions, so that the screening of the unreliable users from the angles of a plurality of credit dimensions is realized, and the technical problem of lower processing efficiency caused by processing the user portrait of the user with the unreliable pedestrian credit data is avoided.
3. And carrying out fusion processing according to the credit characteristics under different credit dimensions and semantic networks among the credit characteristics to obtain user portraits of the users, namely taking the difference of the credit characteristics under the different credit dimensions into consideration, and simultaneously, enabling the generation processing result of the user portraits to more accurately reflect the real credit risk of the users through the identification of the semantic networks, thereby laying a foundation for further effectively controlling the credit risk.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for apparatus, devices, non-volatile computer storage medium embodiments, the description is relatively simple, as it is substantially similar to method embodiments, with reference to the section of the method embodiments being relevant.
The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing is merely one or more embodiments of the present description and is not intended to limit the present description. Various modifications and alterations to one or more embodiments of this description will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, or the like, which is within the spirit and principles of one or more embodiments of the present description, is intended to be included within the scope of the claims of the present description.
Claims (11)
1. The method for constructing the pedestrian credit user image is characterized by comprising the following steps of:
extracting entity information from different information modules of pedestrian credit information data of a user, and extracting credit characteristics of different credit dimensions based on the entity information;
constructing semantic networks among different credit features, performing alignment checking processing on the different credit features through the semantic networks to obtain checking data of the credible features of different credit dimensions, determining the credibility of the user under the different credit dimensions through the checking data, and entering the next step when the credibility does not meet the required credit dimensions;
taking the credit dimension with the credibility not meeting the requirement as a suspected risk dimension, determining the comprehensive credibility of pedestrian credit data according to the number of the suspected risk dimensions and the credibility of different credit dimensions, and entering the next step when the comprehensive credibility meets the requirement;
and acquiring the integrity of credit features under different credit dimensions, and when the reliability of the credit sign data of the user is determined by combining the reliability under different credit dimensions, carrying out fusion processing according to the credit features under different credit dimensions and semantic networks among the credit features to obtain the user portrait of the user.
2. The method for constructing a pedestrian credit user image according to claim 1, wherein the credit module comprises user identity, occupation, living information and historical credit data.
3. The method of claim 1, wherein the credit dimension includes user identity information, credit overdue information, credit application information, credit support information, and credit inquiry information.
4. The pedestrian credit user image construction method of claim 1, wherein the verification data includes the number of other credit features used for the credit feature verification and the number of other credit features that conflict with the semantics of the credit features.
5. The method for constructing a human credit user image according to claim 1, wherein the method for determining the credibility of the user in the credit dimension is as follows:
determining that credit features of a semantic network do not exist with other credit features in the credit dimension according to the checking data, taking the credit features as unchecked features, and determining the basic credibility of the user in the credit dimension according to the number of unchecked features and the number ratio of the credit features in the credit dimension;
taking the credit features which have no semantic conflict with other credit features in the credit dimension as the credit features, and determining the credit feature correction amount of the user in the credit dimension according to the number of the other credit features, the number of the credit features and the number ratio of the credit features in the credit dimension, wherein the other credit features are associated with different credit features;
taking the credit features which have semantic conflict with other credit features in the credit dimension as the unreliable credit features, and determining the feature credibility of the different unreliable credit features according to the number of the other credit features which are associated with the different unreliable credit features, the number of the other credit features which have semantic conflict with the unreliable credit features and the number of the other credit features which have no semantic conflict;
and determining deviation feature correction quantity of the user in the credit dimension according to the quantity of different unreliable credit features, the feature credibility and the quantity ratio of the credit features in the credit dimension, and determining the credibility of the user in the credit dimension by combining the credibility feature correction quantity and the basic credibility.
6. The method for setting up a human credit user image according to claim 1, wherein the confidence level of the user in different credit dimensions ranges from 0 to 1, and when the confidence level of the user in the credit dimensions is smaller than a preset confidence level, it is determined that the confidence level of the credit dimensions does not meet the requirement.
7. The method for constructing the human credit user image according to claim 1, wherein when the user does not have a credible dimension with credibility not meeting the requirement, the user image of the user is obtained directly according to credit features in different credit dimensions and semantic networks among the credit features.
8. The method for constructing the image of the pedestrian credit subscriber as set forth in claim 1, wherein the method for determining the comprehensive credibility of the pedestrian credit data is as follows:
determining the credibility evaluation amount of the suspected risk dimension based on the number of the suspected risk dimensions and the credibility of different suspected risk dimensions;
taking the credit dimension with the suspected risk dimension removed as a credibility credit dimension, and determining the credibility evaluation amount of the credibility dimension based on the number of the credibility dimensions and the credibility of different credibility dimensions;
and obtaining the average value of the credibility of different credit dimensions, and determining the comprehensive credibility of the pedestrian credit data by combining the credibility evaluation value of the credibility risk dimension and the credibility evaluation value of the suspected risk dimension.
9. The method for constructing a user image for pedestrian credit as claimed in claim 1, wherein when the integrated credibility does not meet the requirement, the credibility of the user's pedestrian credit data is determined to be not met, and the credit risk is directly output without constructing a user portrait of the user.
10. The method for constructing the user portrait of the pedestrian credit as claimed in claim 1, wherein the method for constructing the user portrait of the user comprises the following steps:
and carrying out visualization processing according to the credit features in different credit dimensions and semantic networks among the credit features to obtain user portraits of the user in the different credit dimensions, and carrying out fusion processing on the user portraits of the user in the different credit dimensions to obtain the user portraits of the user.
11. A computer system, comprising: a communicatively coupled memory and processor, and a computer program stored on the memory and capable of running on the processor, characterized by: the processor, when executing the computer program, performs a method of constructing a pedestrian credit user image as claimed in any one of claims 1 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410278375.8A CN117876103B (en) | 2024-03-12 | 2024-03-12 | Method and system for setting up pedestrian credit investigation user picture |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410278375.8A CN117876103B (en) | 2024-03-12 | 2024-03-12 | Method and system for setting up pedestrian credit investigation user picture |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117876103A true CN117876103A (en) | 2024-04-12 |
| CN117876103B CN117876103B (en) | 2024-06-18 |
Family
ID=90597174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410278375.8A Active CN117876103B (en) | 2024-03-12 | 2024-03-12 | Method and system for setting up pedestrian credit investigation user picture |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117876103B (en) |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080059364A1 (en) * | 2006-09-01 | 2008-03-06 | Tidwell Lisa C | Systems and methods for performing a financial trustworthiness assessment |
| US8504470B1 (en) * | 2011-08-31 | 2013-08-06 | BT Patent LLC | Methods and systems for financial transactions |
| CN109191281A (en) * | 2018-08-21 | 2019-01-11 | 重庆富民银行股份有限公司 | A kind of group's fraud identifying system of knowledge based map |
| CN110110093A (en) * | 2019-04-08 | 2019-08-09 | 深圳众赢维融科技有限公司 | A kind of recognition methods, device, electronic equipment and the storage medium of knowledge based map |
| CN110163201A (en) * | 2019-03-01 | 2019-08-23 | 腾讯科技（深圳）有限公司 | Image measurement method and apparatus, storage medium and electronic device |
| CN110717816A (en) * | 2019-07-15 | 2020-01-21 | 上海氪信信息技术有限公司 | Artificial intelligence technology-based global financial risk knowledge graph construction method |
| CN114282003A (en) * | 2021-11-30 | 2022-04-05 | 中国建设银行股份有限公司 | Financial risk early warning method and device based on knowledge graph |
| DE202022104425U1 (en) * | 2022-08-03 | 2022-08-09 | Sayed Sayeed Ahmad | Intelligent system for secure integration of credit checks and banking systems through machine learning |
| CN115907786A (en) * | 2022-11-29 | 2023-04-04 | 国网江西省电力有限公司信息通信分公司 | Supplier portrait construction method based on knowledge graph |
| CN116629456A (en) * | 2023-07-20 | 2023-08-22 | 杭银消费金融股份有限公司 | Method, system and storage medium for predicting overdue risk of service |
| CN117035985A (en) * | 2023-07-31 | 2023-11-10 | 浙江孚临科技有限公司 | Trusted overdue risk control method and system |
| CN117078026A (en) * | 2023-10-17 | 2023-11-17 | 杭银消费金融股份有限公司 | Wind control index management method and system based on data blood margin |
| CN117495548A (en) * | 2023-12-07 | 2024-02-02 | 中国农业银行股份有限公司 | Risk early warning method, device, equipment and medium |
| CN117556225A (en) * | 2024-01-12 | 2024-02-13 | 杭银消费金融股份有限公司 | Pedestrian credit data risk management system |
-
2024
- 2024-03-12 CN CN202410278375.8A patent/CN117876103B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080059364A1 (en) * | 2006-09-01 | 2008-03-06 | Tidwell Lisa C | Systems and methods for performing a financial trustworthiness assessment |
| US8504470B1 (en) * | 2011-08-31 | 2013-08-06 | BT Patent LLC | Methods and systems for financial transactions |
| CN109191281A (en) * | 2018-08-21 | 2019-01-11 | 重庆富民银行股份有限公司 | A kind of group's fraud identifying system of knowledge based map |
| CN110163201A (en) * | 2019-03-01 | 2019-08-23 | 腾讯科技（深圳）有限公司 | Image measurement method and apparatus, storage medium and electronic device |
| CN110110093A (en) * | 2019-04-08 | 2019-08-09 | 深圳众赢维融科技有限公司 | A kind of recognition methods, device, electronic equipment and the storage medium of knowledge based map |
| CN110717816A (en) * | 2019-07-15 | 2020-01-21 | 上海氪信信息技术有限公司 | Artificial intelligence technology-based global financial risk knowledge graph construction method |
| CN114282003A (en) * | 2021-11-30 | 2022-04-05 | 中国建设银行股份有限公司 | Financial risk early warning method and device based on knowledge graph |
| DE202022104425U1 (en) * | 2022-08-03 | 2022-08-09 | Sayed Sayeed Ahmad | Intelligent system for secure integration of credit checks and banking systems through machine learning |
| CN115907786A (en) * | 2022-11-29 | 2023-04-04 | 国网江西省电力有限公司信息通信分公司 | Supplier portrait construction method based on knowledge graph |
| CN116629456A (en) * | 2023-07-20 | 2023-08-22 | 杭银消费金融股份有限公司 | Method, system and storage medium for predicting overdue risk of service |
| CN117035985A (en) * | 2023-07-31 | 2023-11-10 | 浙江孚临科技有限公司 | Trusted overdue risk control method and system |
| CN117078026A (en) * | 2023-10-17 | 2023-11-17 | 杭银消费金融股份有限公司 | Wind control index management method and system based on data blood margin |
| CN117495548A (en) * | 2023-12-07 | 2024-02-02 | 中国农业银行股份有限公司 | Risk early warning method, device, equipment and medium |
| CN117556225A (en) * | 2024-01-12 | 2024-02-13 | 杭银消费金融股份有限公司 | Pedestrian credit data risk management system |
Non-Patent Citations (1)
| Title |
|---|
| 陈芊汝;: "区块链与大数据技术的结合对互联网征信发展的启示", 甘肃金融, no. 11, 15 November 2016 (2016-11-15) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117876103B (en) | 2024-06-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109815803B (en) | Face examination risk control method and device, computer equipment and storage medium | |
| CN112084383A (en) | Information recommendation method, device and equipment based on knowledge graph and storage medium | |
| CN109697665A (en) | Loan checking method, device, equipment and medium based on artificial intelligence | |
| CN111652096B (en) | Face recognition method, device and system based on block chain | |
| CN116629456B (en) | Method, system and storage medium for predicting overdue risk of service | |
| CN112181835B (en) | Automatic test method, device, computer equipment and storage medium | |
| CN111709718A (en) | Intelligent warranty asset service platform, method and storage medium based on artificial intelligence | |
| CN110309099A (en) | Interface managerial method, device, equipment and computer readable storage medium | |
| CN112700250A (en) | Identity authentication method, device and system in financial scene | |
| CN112988896A (en) | Synchronous consensus method, system, device and medium based on block chain | |
| CN115034886A (en) | Default risk prediction method and device | |
| CN117876103B (en) | Method and system for setting up pedestrian credit investigation user picture | |
| CN111461191B (en) | Method and device for determining image sample set for model training and electronic equipment | |
| CN112966304A (en) | Method and device for preventing process document from being tampered, computer equipment and medium | |
| CN112801767A (en) | Accounting data processing method and device for newly added channels | |
| CN111275540A (en) | Consumption credit approval method and device and terminal equipment | |
| CN109636572A (en) | Risk checking method, device, equipment and the readable storage medium storing program for executing of bank card | |
| CN115150150A (en) | Information recording method based on block chain technology | |
| CN115050042A (en) | Claims data entry method and device, computer equipment and storage medium | |
| CN109636573A (en) | Credit analysis method, device, equipment and readable storage medium storing program for executing | |
| CN111770080A (en) | Method and device for recovering device fingerprint | |
| CN111091354A (en) | Real estate information processing system based on big data | |
| CN117764710B (en) | Monitoring method for housing financial risk behaviors | |
| CN113450124B (en) | Outbound method and device based on user behavior, electronic equipment and medium | |
| CN110310020B (en) | Project scheme management method based on data analysis, related device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |