CN111383102A - Financial credit risk identification method, model construction method and device - Google Patents
Financial credit risk identification method, model construction method and device Download PDFInfo
- Publication number
- CN111383102A CN111383102A CN202010229807.8A CN202010229807A CN111383102A CN 111383102 A CN111383102 A CN 111383102A CN 202010229807 A CN202010229807 A CN 202010229807A CN 111383102 A CN111383102 A CN 111383102A
- Authority
- CN
- China
- Prior art keywords
- financial
- map
- information
- credit risk
- risk identification
- 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 41
- 238000010276 construction Methods 0.000 title claims abstract description 13
- 238000010801 machine learning Methods 0.000 claims abstract description 24
- 238000004364 calculation method Methods 0.000 claims description 16
- 238000000605 extraction Methods 0.000 claims description 14
- 238000012546 transfer Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 8
- 230000001364 causal effect Effects 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 238000003062 neural network model Methods 0.000 description 5
- 238000013473 artificial intelligence Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 241000220225 Malus Species 0.000 description 2
- 235000021016 apples Nutrition 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000006855 networking Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000011144 upstream manufacturing Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000009193 crawling Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000007723 transport mechanism Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Technology Law (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
A financial credit risk identification model construction method, a financial credit risk identification model construction device, a financial credit risk identification method, a financial credit risk identification device and a computer readable storage medium, wherein the financial credit risk identification method comprises the following steps: generating a financial credit knowledge map according to the information of the loan subject; generating a financial affair map according to the financial event data and the financial field ontology; generating a financial credit risk identification model through machine learning based on the information of the loan subject, the financial credit knowledge map and the financial affairs map; and updating the financial affair map according to the real-time financial event, and outputting financial credit risk information of the real-time financial event to the loan subject to be predicted through the financial credit risk identification model according to the updated financial affair map and the information of the loan subject to be predicted. According to the method and the device, the financial credit risk can be predicted according to real-time financial events, the dimensionality of risk identification is enriched, and the prequalification of credit risk of a credit main body is improved.
Description
Technical Field
The present disclosure relates to the field of risk identification, and more particularly, to a method and an apparatus for constructing a financial credit risk identification model, a method and an apparatus for identifying a financial credit risk, and a computer-readable storage medium.
Background
There are two main risks in a financial credit business scenario: fraud risk and credit risk can be effectively identified by a risk identification means in a business application stage, the credit risk identification can only be researched and judged according to past historical data of an application user, however, in a credit business environment, particularly in the credit environment of a small enterprise owner, the business condition is influenced by various aspects such as market environment, regulations, surrounding environment, upstream and downstream industry development and the like, the identification in a traditional mode is mainly concentrated in a pre-loan link, corresponding analysis and prediction are lacked in loan, and corresponding measures cannot be taken aiming at later-stage risks.
In the financial loan link, the main method for identifying overdue risks of the user is to establish the expectation of predicting future repayment of the user through the past repayment record of the user by using a user behavior scoring card. However, when a loan entity operates a loan, the operation result of the loan entity determines the later repayment capacity of the loan, and the loan entity belongs to the main influence factors of the credit risk, while the operation condition of the operation entity is often influenced by other objective conditions, such as the industry environment, the policy environment, the upstream and downstream environment, and other factors, and the current financial risk identification uses the historical repayment condition of the loan entity, and has hysteresis. In addition, the identification mode mainly depends on the historical repayment condition of the credit owner and basic personal information, and has one-sidedness.
Disclosure of Invention
The application provides a financial credit risk identification model construction method and device, a financial credit risk identification method and device and a computer readable storage medium, so as to improve the financial credit risk identification capability.
The embodiment of the application provides a financial credit risk identification method, which comprises the following steps:
generating a financial credit knowledge map according to the information of the loan subject;
generating a financial affair map according to the financial event data and the financial field ontology;
generating a financial credit risk identification model through machine learning based on the information of the loan subject, the financial credit knowledge map and the financial affairs map;
and updating the financial affair map according to the real-time financial event, and outputting financial credit risk information of the real-time financial event to the loan subject to be predicted through the financial credit risk identification model according to the updated financial affair map and the information of the loan subject to be predicted.
The embodiment of the application further provides a financial credit risk identification model construction method, which comprises the following steps:
generating a financial credit knowledge map according to the information of the loan subject;
generating a financial affair map according to the financial event data and the financial field ontology;
and generating a financial credit risk identification model through machine learning based on the information of the loan subject, the financial credit knowledge map and the financial affairs map.
The embodiment of the present application further provides a financial credit risk identification apparatus, including:
the financial credit knowledge map module is used for generating a financial credit knowledge map according to the information of the loan subject;
the financial affair map module is used for generating a financial affair map according to the financial event data and the financial field ontology;
the credit risk transfer identification module is used for generating a financial credit risk identification model through machine learning based on the information of the loan subject, the financial credit knowledge map and the financial affair map;
and the risk pre-judging module is used for updating the financial affair map according to the real-time financial event, and outputting the financial credit risk information of the real-time financial event to the loan subject to be predicted through the financial credit risk identification model according to the updated financial affair map and the information of the loan subject to be predicted.
The embodiment of the present application further provides a financial credit risk identification model building apparatus, including:
the financial credit knowledge map module is used for generating a financial credit knowledge map according to the information of the loan subject;
the financial affair map module is used for generating a financial affair map according to the financial event data and the financial field ontology;
and the credit risk transfer identification module is used for generating a financial credit risk identification model through machine learning based on the information of the loan body, the financial credit knowledge map and the financial affair map.
The embodiment of the present application further provides a financial credit risk identification apparatus, including: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the financial credit risk identification method when executing the program.
The embodiment of the present application further provides a financial credit risk identification model building apparatus, including: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the financial credit risk identification model construction method when executing the program.
Embodiments of the present application further provide a computer-readable storage medium storing computer-executable instructions for executing the financial credit risk identification method.
The embodiment of the application also provides a computer-readable storage medium, which stores computer-executable instructions for executing the financial credit risk identification model construction method.
Compared with the related art, the method comprises the following steps: generating a financial credit knowledge map according to the information of the loan subject; generating a financial affair map according to the financial event data and the financial field ontology; generating a financial credit risk identification model through machine learning based on the information of the loan subject, the financial credit knowledge map and the financial affairs map;
and updating the financial affair map according to the real-time financial event, and outputting financial credit risk information of the real-time financial event to the loan subject to be predicted through the financial credit risk identification model according to the updated financial affair map and the information of the loan subject to be predicted. According to the method and the device, the financial credit risk can be predicted according to real-time financial events, the dimensionality of risk identification is enriched, and the prequalification of credit risk of a credit main body is improved.
Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the application. Other advantages of the present application may be realized and attained by the instrumentalities and combinations particularly pointed out in the specification and the drawings.
Drawings
The accompanying drawings are included to provide an understanding of the present disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the examples serve to explain the principles of the disclosure and not to limit the disclosure.
FIG. 1 is a flow chart of a financial credit risk identification model construction method according to an embodiment of the present application;
FIG. 2 is a flowchart of steps 101 and 501 of an embodiment of the present application;
FIG. 3 is a flowchart of steps 102 and 502 of an embodiment of the present application;
FIG. 4 is a flowchart of steps 103 and 503 of an embodiment of the present application;
FIG. 5 is a flowchart of a financial credit risk identification method according to an embodiment of the present application;
FIG. 6 is a flowchart of step 504 according to an embodiment of the present application;
fig. 7 is a schematic composition diagram of a financial credit risk identification model construction device according to an embodiment of the present application;
FIG. 8 is a schematic diagram showing the components of the financial credit risk identifying apparatus according to the embodiment of the present application;
fig. 9 is a schematic diagram of an application example of the present application.
Detailed Description
The present application describes embodiments, but the description is illustrative rather than limiting and it will be apparent to those of ordinary skill in the art that many more embodiments and implementations are possible within the scope of the embodiments described herein. Although many possible combinations of features are shown in the drawings and discussed in the detailed description, many other combinations of the disclosed features are possible. Any feature or element of any embodiment may be used in combination with or instead of any other feature or element in any other embodiment, unless expressly limited otherwise.
The present application includes and contemplates combinations of features and elements known to those of ordinary skill in the art. The embodiments, features and elements disclosed in this application may also be combined with any conventional features or elements to form a unique inventive concept as defined by the claims. Any feature or element of any embodiment may also be combined with features or elements from other inventive aspects to form yet another unique inventive aspect, as defined by the claims. Thus, it should be understood that any of the features shown and/or discussed in this application may be implemented alone or in any suitable combination. Accordingly, the embodiments are not limited except as by the appended claims and their equivalents. Furthermore, various modifications and changes may be made within the scope of the appended claims.
Further, in describing representative embodiments, the specification may have presented the method and/or process as a particular sequence of steps. However, to the extent that the method or process does not rely on the particular order of steps set forth herein, the method or process should not be limited to the particular sequence of steps described. Other orders of steps are possible as will be understood by those of ordinary skill in the art. Therefore, the particular order of the steps set forth in the specification should not be construed as limitations on the claims. Further, the claims directed to the method and/or process should not be limited to the performance of their steps in the order written, and one skilled in the art can readily appreciate that the sequences may be varied and still remain within the spirit and scope of the embodiments of the present application.
As shown in fig. 1, an embodiment of the present application provides a method for constructing a financial credit risk identification model, including:
Wherein, the loan subject may be a company, an individual, or the like. The information of the loan body can comprise basic information, social information, geographic information, financial attribute information, business and other dimensional data of the loan body.
From the AI (Artificial Intelligence) perspective, the knowledge graph is a knowledge base for understanding human language, and from the database perspective, the knowledge graph is a novel knowledge storage structure; from the perspective of knowledge representation, a knowledge graph is a method for calculating mechanism solution knowledge; from a web perspective, a knowledge graph is a semantic interconnection between knowledge data.
The knowledge map is a knowledge base based on binary relations, and is used for describing entities (or concepts, the concepts are abstractions of the entities, for example, "fruits" are concepts of "apples") in the real world and the mutual relations thereof, the basic composition unit of the knowledge map is a triple (entity) -relation-entity triple, and the entities are mutually connected through relations to form a mesh structure. Through the knowledge graph, a user can be supported to search according to subjects instead of character strings, and therefore information search on a semantic level is truly realized. The search engine based on the knowledge graph can directly feed back the structured knowledge to the user, and the user can find the knowledge which the user wants to obtain without browsing a large number of webpages.
As shown in fig. 2, in one embodiment, step 101 includes:
The information of the loan body can be obtained according to modes of actively reporting and/or networking inquiry and the like of the loan body.
The financial domain ontology may be a company, an individual, etc. The main bodies of the finance credit knowledge-graph correspond to the finance field ontology, and the relationship of the finance credit knowledge-graph refers to the association relationship between the main bodies, for example, the main bodies of the applicant: applicant's body and the partner's body, both of which are in a partner relationship. Company principal A and company principal B, B being a sub-company relationship of A, and so on.
And step 203, extracting financial credit knowledge map data from the information of the loan body according to the body and the relation of the financial credit knowledge map.
Wherein corresponding data may be extracted from the loan principal's information against the principal and relationship of the financial credit knowledge-graph.
And step 204, constructing the financial credit knowledge-graph according to the financial credit knowledge-graph data.
Wherein the financial credit knowledge-graph may be constructed from the financial credit knowledge-graph data based on a form of a graphical database, which may include a Neo4j graphical database.
And 102, generating a financial affairs map according to the financial event data and the financial field ontology.
The financial domain ontology base may be constructed by an expert. The financial domain ontology library includes a plurality of financial domain ontologies.
The event map is a logical knowledge base of events and describes evolution rules and modes between events.
As shown in FIG. 3, in one embodiment, step 102 includes:
The financial event data of the internet can be crawled by crawlers to obtain the financial event data.
Wherein the events can be extracted in various ways, such as extraction using rules, extraction using neural network models, logic-based extraction, and the like.
And step 303, establishing a financial affairs map of the causal relationship of the event for the financial event based on the financial field ontology.
The event cause and effect relationship refers to the relationship between the former and the later, wherein the former event can cause the latter to occur.
And 103, generating a financial credit risk identification model through machine learning based on the information of the loan body, the financial credit knowledge map and the financial affairs map.
In one embodiment, as shown in FIG. 4, step 103 comprises:
Feature engineering calculation and label calculation can be performed according to the multidimensional data in the information of the loan body to obtain feature data and labels of the loan body, wherein the feature data may include feature names, feature values and the like. The label is the output (Y) in the supervised machine learning algorithm, for example, when the financial field identifies the fraud risk, the characteristics which can be extracted through the characteristic engineering according to the bank flow information of the user are the transfer amount of each month, the consumption amount of the last half year and the like, if the person is judged to be a fraudulent user, the fraud is the label of the person.
And 402, performing vectorization calculation on the financial credit knowledge graph based on the characteristic data to obtain vector data.
The vectorization calculation may have various forms, for example, the vectorization calculation may be performed in a node2vec (node-to-vector) manner.
And 403, setting a conduction weight parameter based on the financial affairs map.
Wherein, the corresponding conduction weight parameter can be set according to the possibility of event conversion in the financial affairs map. For example, the conduction weight parameter may be set based on PageRank (web page ranking), or may be set accordingly according to actual situations.
And 404, training a preset machine learning model according to the characteristic data, the vector data and the conduction weight parameters to obtain a financial credit risk identification model, wherein the output of the financial credit risk identification model is financial credit risk information.
The machine learning model may take many forms, such as a decision tree, a neural network model, and so on.
The financial credit risk information may include fraud risk and credit risk, and may be divided into different levels according to the level of risk, and a predicted risk level is output.
The financial credit risk identification model is constructed on the basis of the financial credit knowledge graph and the financial affairs graph, so that the real-time financial credit risk can be predicted by using the model.
As shown in fig. 5, an embodiment of the present application further provides a financial credit risk identification method, including:
Wherein, the loan subject may be a company, an individual, or the like. The information of the loan body can comprise basic information, social information, geographic information, financial attribute information, business and other dimensional data of the loan body.
From the AI (Artificial Intelligence) perspective, the knowledge graph is a knowledge base for understanding human language, and from the database perspective, the knowledge graph is a novel knowledge storage structure; from the perspective of knowledge representation, a knowledge graph is a method for calculating mechanism solution knowledge; from a web perspective, a knowledge graph is a semantic interconnection between knowledge data.
The knowledge map is a knowledge base based on binary relations, and is used for describing entities (or concepts, the concepts are abstractions of the entities, for example, "fruits" are concepts of "apples") in the real world and the mutual relations thereof, the basic composition unit of the knowledge map is a triple (entity) -relation-entity triple, and the entities are mutually connected through relations to form a mesh structure. Through the knowledge graph, a user can be supported to search according to subjects instead of character strings, and therefore information search on a semantic level is truly realized. The search engine based on the knowledge graph can directly feed back the structured knowledge to the user, and the user can find the knowledge which the user wants to obtain without browsing a large number of webpages.
As shown in fig. 2, in one embodiment, step 501 includes:
The information of the loan body can be obtained according to modes of actively reporting and/or networking inquiry and the like of the loan body.
The financial domain ontology may be a company, an individual, etc. The main bodies of the finance credit knowledge-graph correspond to the finance field ontology, and the relationship of the finance credit knowledge-graph refers to the association relationship between the main bodies, for example, the main bodies of the applicant: applicant's body and the partner's body, both of which are in a partner relationship. Company principal A and company principal B, B being a sub-company relationship of A, and so on.
And step 203, extracting financial credit knowledge map data from the information of the loan body according to the body and the relation of the financial credit knowledge map.
Wherein corresponding data may be extracted from the loan principal's information against the principal and relationship of the financial credit knowledge-graph.
And step 204, constructing the financial credit knowledge-graph according to the financial credit knowledge-graph data.
Wherein the financial credit knowledge-graph may be constructed from the financial credit knowledge-graph data based on a form of a graphical database, which may include a Neo4j graphical database.
The financial domain ontology base may be constructed by an expert. The financial domain ontology library includes a plurality of financial domain ontologies.
The event map is a logical knowledge base of events and describes evolution rules and modes between events.
As shown in FIG. 3, in one embodiment, step 502 includes:
The financial event data of the internet can be crawled by crawlers to obtain the financial event data.
Wherein the events can be extracted in various ways, such as extraction using rules, extraction using neural network models, logic-based extraction, and the like.
And step 303, establishing a financial affairs map of the causal relationship of the event for the financial event based on the financial field ontology.
The event cause and effect relationship refers to the relationship between the former and the later, wherein the former event can cause the latter to occur.
And 503, generating a financial credit risk identification model through machine learning based on the information of the loan body, the financial credit knowledge map and the financial affairs map.
In one embodiment, as shown in FIG. 4, step 503 comprises:
Feature engineering calculation and label calculation can be performed according to the multidimensional data in the information of the loan body to obtain feature data of the loan body, wherein the feature data can include feature names, feature values, label names and the like.
And 402, performing vectorization calculation on the financial credit knowledge graph based on the characteristic data to obtain vector data.
The vectorization calculation may have various forms, for example, the vectorization calculation may be performed in a node2vec (node-to-vector) manner.
And 403, setting a conduction weight parameter based on the financial affairs map.
Wherein, the corresponding conduction weight parameter can be set according to the possibility of event conversion in the financial affairs map. For example, the conduction weight parameter may be set based on PageRank (web page ranking), or may be set accordingly according to actual situations.
And 404, training a preset machine learning model according to the characteristic data, the vector data and the conduction weight parameters to obtain a financial credit risk identification model, wherein the output of the financial credit risk identification model is financial credit risk information.
The machine learning model may take many forms, such as a decision tree, a neural network model, and so on.
The financial credit risk information may include fraud risk and credit risk, and may be divided into different levels according to the level of risk, and a predicted risk level is output.
And 504, updating the financial affair map according to the real-time financial event, and outputting financial credit risk information of the real-time financial event to the loan subject to be predicted through the financial credit risk identification model according to the updated financial affair map and the information of the loan subject to be predicted.
In one embodiment, as shown in FIG. 6, step 504 includes:
Referring to steps 301-302, internet financial event data can be obtained in real time, real-time financial events are obtained, and then a financial affair map is updated.
And step 602, performing characteristic engineering processing according to the information of the loan body to be predicted to obtain characteristic data to be predicted.
In which, referring to step 401, the feature data to be predicted may be determined.
Wherein, with reference to step 402, the vector data to be predicted can be determined.
And step 604, updating the conduction weight parameters based on the updated financial affairs map.
Therein, the conduction weight parameter may be updated with reference to step 403.
And 605, inputting the characteristic data to be predicted, the vector data to be predicted and the updated conduction weight parameter into the financial credit risk identification model to obtain financial credit risk information of the real-time financial event to be predicted on the loan subject to be predicted, which is output by the financial credit risk identification model.
According to the method and the device, the financial credit risk can be predicted according to real-time financial events, the dimensionality of risk identification is enriched, and the prequalification of credit risk of a credit main body is improved.
As shown in fig. 7, an embodiment of the present application further provides a financial credit risk identification model building apparatus, including:
the financial credit knowledge map module 71 is used for generating a financial credit knowledge map according to the information of the loan body;
a financial affairs map module 72 for generating a financial affairs map based on the financial event data and the financial domain ontology;
and the credit risk transfer identification module 73 is used for generating a financial credit risk identification model through machine learning based on the information of the loan body, the financial credit knowledge map and the financial affairs map.
In one embodiment, the financial credit knowledge-map module 71 is configured to:
obtaining information of the loan body;
constructing the main body and the relation of the financial credit knowledge-graph according to the financial field ontology;
extracting financial credit knowledge map data from the information of the loan subject according to the subject and the relationship of the financial credit knowledge map;
and constructing the financial credit knowledge-graph according to the financial credit knowledge-graph data.
In one embodiment, the financial affairs map module 72 is configured to:
acquiring financial event data;
performing event extraction on the financial event data to obtain a financial event;
and establishing a financial affair map of event causal relationship for the financial event based on the financial field ontology.
In one embodiment, the credit risk transfer identification module 73 is configured to:
performing characteristic engineering processing on the information of the loan body to obtain characteristic data of the loan body;
vectorizing calculation is carried out on the financial credit knowledge graph to obtain vector data;
setting a conducted weight parameter based on the financial affairs map;
and training a preset machine learning model according to the characteristic data, the vector data and the conduction weight parameters to obtain a financial credit risk identification model, wherein the output of the financial credit risk identification model is financial credit risk information.
As shown in fig. 8, an embodiment of the present application further provides a financial credit risk identification apparatus, including:
the financial credit knowledge map module 71 is used for generating a financial credit knowledge map according to the information of the loan body;
a financial affairs map module 72 for generating a financial affairs map based on the financial event data and the financial domain ontology;
a credit risk transfer identification module 73 for generating a financial credit risk identification model through machine learning based on the loan principal's information, the financial credit knowledge-graph and financial affairs-graph;
and the risk pre-judging module 81 is used for updating the financial affair map according to the real-time financial event, and outputting financial credit risk information of the real-time financial event to the loan subject to be predicted through the financial credit risk identification model according to the updated financial affair map and the information of the loan subject to be predicted.
In one embodiment, the financial credit knowledge-map module 71 is configured to:
obtaining information of the loan body;
constructing the main body and the relation of the financial credit knowledge-graph according to the financial field ontology;
extracting financial credit knowledge map data from the information of the loan subject according to the subject and the relationship of the financial credit knowledge map;
and constructing the financial credit knowledge-graph according to the financial credit knowledge-graph data.
In one embodiment, the financial affairs map module 72 is configured to:
acquiring financial event data;
performing event extraction on the financial event data to obtain a financial event;
and establishing a financial affair map of event causal relationship for the financial event based on the financial field ontology.
In one embodiment, the credit risk transfer identification module 73 is configured to:
performing characteristic engineering processing on the information of the loan body to obtain characteristic data of the loan body;
vectorizing calculation is carried out on the financial credit knowledge graph to obtain vector data;
setting a conducted weight parameter based on the financial affairs map;
and training a preset machine learning model according to the characteristic data, the vector data and the conduction weight parameters to obtain a financial credit risk identification model, wherein the output of the financial credit risk identification model is financial credit risk information.
In an embodiment, the risk pre-judging module 81 is configured to:
performing characteristic engineering processing according to the information of the loan subject to be predicted to obtain characteristic data to be predicted;
obtaining vector data to be predicted based on the feature data to be predicted and the financial credit knowledge base;
updating a conducted weight parameter based on the updated financial affairs map;
and inputting the characteristic data to be predicted, the vector data and the updated conduction weight parameters into the financial credit risk identification model to obtain financial credit risk information of the real-time financial event output by the financial credit risk identification model to the loan subject to be predicted.
According to the method and the device, the financial credit risk can be predicted according to real-time financial events, the dimensionality of risk identification is enriched, and the prequalification of credit risk of a credit main body is improved.
For example, if a bank loans a subsidiary company B under the company A, due to the fact that the reputation of the related signing company A is damaged due to tax stealing and tax losing events of the artist signed by the company A, the fund chain of the company is affected, and the business fluctuates to the company B for bank loan in the later period, the business has bad account risk, if the business passes through a case map, the risk can be immediately recognized after the event comes out, the bank is prompted to take measures such as fund conservation and the like for the business in advance, and the bad account situation is prevented from occurring.
The following is a description of an application example. Referring to fig. 9, the financial credit risk identification method may include the steps of:
a first module: finance credit knowledge-map module 71
Step 1, obtaining basic information, social information, geographic information, financial attribute information, business and other dimensional data of a loan body;
step 2, constructing a financial field body according to the financial business field knowledge;
step 3, constructing a main body and a relation of the knowledge graph according to the domain ontology in the step 2; for example, Applicant's subject: applicant's body and spouse's body, both are spouse relationships;
step 4, extracting atlas data from the data obtained in the step 1 according to the main body and the relation constructed by the knowledge atlas in the step 3;
step 5, constructing a financial credit knowledge graph based on the neo4j graph database by using the graph data acquired in the step 4;
wherein, a knowledge graph is formed according to the extracted main bodies and the relationship;
and a second module: financial domain affairs map module 72
Step 6, constructing a financial field ontology library;
step 7, crawling the financial events of the internet by using a crawler;
step 8, performing event extraction on the data obtained in the step 7;
wherein, the extraction can be performed by using a rule, a neural network model, logic-based extraction and the like;
step 9, based on the ontology library in the step 6, establishing a cause and effect relationship graph of the event extracted in the step 8;
and a third module: credit risk transfer identification module 73
Step 10, performing characteristic engineering calculation and label calculation on the multi-dimensional data acquired in the step 1;
step 11, vectorizing calculation is carried out on the knowledge graph;
and a module IV: financial risk anticipation module 81
Step 14, acquiring an internet event in real time; capturing real-time reported events, for example, from a web portal;
step 15, fusing the event in the step 14 into the affair map established in the step 9;
step 16, recalculating step 10 to step 12;
and step 17, inputting the characteristic data of the step 16 into the model of the step 13, and outputting the risk prediction of the current event to the borrowing subject.
The embodiment of the present application further provides a financial credit risk identification model building apparatus, including: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the financial credit risk identification model construction method when executing the program.
The embodiment of the present application further provides a financial credit risk identification apparatus, including: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the financial credit risk identification method when executing the program.
The embodiment of the application also provides a computer-readable storage medium, which stores computer-executable instructions for executing the financial credit risk identification model construction method.
Embodiments of the present application further provide a computer-readable storage medium storing computer-executable instructions for executing the financial credit risk identification method.
In this embodiment, the storage medium may include, but is not limited to: a U-disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic or optical disk, and other various media capable of storing program codes.
It will be understood by those of ordinary skill in the art that all or some of the steps of the methods, systems, functional modules/units in the devices disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. In a hardware implementation, the division between functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be performed by several physical components in cooperation. Some or all of the components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as is well known to those of ordinary skill in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, Digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can accessed by a computer. In addition, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media as known to those skilled in the art.
Claims (10)
1. A financial credit risk identification method, comprising:
generating a financial credit knowledge map according to the information of the loan subject;
generating a financial affair map according to the financial event data and the financial field ontology;
generating a financial credit risk identification model through machine learning based on the information of the loan subject, the financial credit knowledge map and the financial affairs map;
and updating the financial affair map according to the real-time financial event, and outputting financial credit risk information of the real-time financial event to the loan subject to be predicted through the financial credit risk identification model according to the updated financial affair map and the information of the loan subject to be predicted.
2. The method of claim 1, wherein generating a financial credit knowledge-map from the lender's information comprises:
obtaining information of the loan body;
constructing the main body and the relation of the financial credit knowledge-graph according to the financial field ontology;
extracting financial credit knowledge map data from the information of the loan subject according to the subject and the relationship of the financial credit knowledge map;
and constructing the financial credit knowledge-graph according to the financial credit knowledge-graph data.
3. The method of claim 1, wherein generating the financial affairs map from the financial event data and the financial domain ontology comprises:
acquiring financial event data;
performing event extraction on the financial event data to obtain a financial event;
and establishing a financial affair map of event causal relationship for the financial event based on the financial field ontology.
4. The method of claim 1, wherein generating a finance credit risk identification model through machine learning based on the information of the lending subject, the finance credit knowledge-graph and a finance affairs graph comprises:
performing characteristic engineering processing on the information of the loan body to obtain characteristic data of the loan body;
vectorizing calculation is carried out on the financial credit knowledge graph to obtain vector data;
setting a conducted weight parameter based on the financial affairs map;
and training a preset machine learning model according to the characteristic data, the vector data and the conduction weight parameters to obtain a financial credit risk identification model, wherein the output of the financial credit risk identification model is financial credit risk information.
5. The method according to claim 1, wherein the outputting the financial credit risk information of the real-time financial event to the loan subject to be predicted through the financial credit risk identification model according to the updated financial affairs map and the information of the loan subject to be predicted comprises:
performing characteristic engineering processing according to the information of the loan subject to be predicted to obtain characteristic data to be predicted;
obtaining vector data to be predicted based on the feature data to be predicted and the financial credit knowledge base;
updating a conducted weight parameter based on the updated financial affairs map;
and inputting the characteristic data to be predicted, the vector data and the updated conduction weight parameters into the financial credit risk identification model to obtain financial credit risk information of the real-time financial event output by the financial credit risk identification model to the loan subject to be predicted.
6. A financial credit risk identification model construction method comprises the following steps:
generating a financial credit knowledge map according to the information of the loan subject;
generating a financial affair map according to the financial event data and the financial field ontology;
and generating a financial credit risk identification model through machine learning based on the information of the loan subject, the financial credit knowledge map and the financial affairs map.
7. A financial credit risk identification apparatus, comprising:
the financial credit knowledge map module is used for generating a financial credit knowledge map according to the information of the loan subject;
the financial affair map module is used for generating a financial affair map according to the financial event data and the financial field ontology;
the credit risk transfer identification module is used for generating a financial credit risk identification model through machine learning based on the information of the loan subject, the financial credit knowledge map and the financial affair map;
and the risk pre-judging module is used for updating the financial affair map according to the real-time financial event, and outputting the financial credit risk information of the real-time financial event to the loan subject to be predicted through the financial credit risk identification model according to the updated financial affair map and the information of the loan subject to be predicted.
8. A financial credit risk identification model building apparatus, comprising:
the financial credit knowledge map module is used for generating a financial credit knowledge map according to the information of the loan subject;
the financial affair map module is used for generating a financial affair map according to the financial event data and the financial field ontology;
and the credit risk transfer identification module is used for generating a financial credit risk identification model through machine learning based on the information of the loan body, the financial credit knowledge map and the financial affair map.
9. A financial credit risk identification apparatus comprising: memory, processor and computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1 to 5 when executing the program.
10. A computer-readable storage medium storing computer-executable instructions for performing the method of any one of claims 1-6.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010229807.8A CN111383102B (en) | 2020-03-27 | 2020-03-27 | Financial credit risk identification method, model construction method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010229807.8A CN111383102B (en) | 2020-03-27 | 2020-03-27 | Financial credit risk identification method, model construction method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111383102A true CN111383102A (en) | 2020-07-07 |
CN111383102B CN111383102B (en) | 2023-10-24 |
Family
ID=71215696
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010229807.8A Active CN111383102B (en) | 2020-03-27 | 2020-03-27 | Financial credit risk identification method, model construction method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111383102B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112131392A (en) * | 2020-08-01 | 2020-12-25 | 赛飞特工程技术集团有限公司 | Public health epidemic situation early warning method and system based on knowledge graph |
CN112529681A (en) * | 2021-02-08 | 2021-03-19 | 南京审计大学 | Credit risk transfer method based on credit subject correlation strength |
CN112598507A (en) * | 2020-12-25 | 2021-04-02 | 北京知因智慧科技有限公司 | Excessive credit granting risk prediction system and method based on knowledge graph |
CN112785414A (en) * | 2021-01-04 | 2021-05-11 | 上海海事大学 | Credit risk prediction method based on knowledge graph and ontology inference engine |
CN112785418A (en) * | 2021-01-22 | 2021-05-11 | 深圳前海微众银行股份有限公司 | Credit risk modeling method, credit risk modeling device, credit risk modeling equipment and computer readable storage medium |
CN113610626A (en) * | 2021-07-26 | 2021-11-05 | 建信金融科技有限责任公司 | Bank credit risk identification knowledge graph construction method and device, computer equipment and computer readable storage medium |
CN115269879A (en) * | 2022-09-05 | 2022-11-01 | 北京百度网讯科技有限公司 | Knowledge structure data generation method, data search method and risk warning method |
CN115689720A (en) * | 2022-09-02 | 2023-02-03 | 睿智合创(北京)科技有限公司 | Newly-increased risk monitoring system based on risk attribution |
CN116452323A (en) * | 2023-04-28 | 2023-07-18 | 广州市良策网络技术有限公司 | Risk assessment method, system, equipment and storage medium |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107945024A (en) * | 2017-12-12 | 2018-04-20 | 厦门市美亚柏科信息股份有限公司 | Identify that internet finance borrowing enterprise manages abnormal method, terminal device and storage medium |
CN108596439A (en) * | 2018-03-29 | 2018-09-28 | 北京中兴通网络科技股份有限公司 | A kind of the business risk prediction technique and system of knowledge based collection of illustrative plates |
CN108763556A (en) * | 2018-06-01 | 2018-11-06 | 北京奇虎科技有限公司 | Usage mining method and device based on demand word |
CN109726819A (en) * | 2018-12-29 | 2019-05-07 | 东软集团股份有限公司 | A kind of method and device for realizing reasoning |
CN109977237A (en) * | 2019-05-27 | 2019-07-05 | 南京擎盾信息科技有限公司 | A kind of dynamic law occurrence diagram spectrum construction method towards legal field |
CN110134797A (en) * | 2019-04-29 | 2019-08-16 | 贳巽(北京)国际商业数据技术股份公司 | A method of change of financial market is studied and judged based on reason map and multi-sector model |
US20190259033A1 (en) * | 2015-06-20 | 2019-08-22 | Quantiply Corporation | System and method for using a data genome to identify suspicious financial transactions |
CN110363449A (en) * | 2019-07-25 | 2019-10-22 | 中国工商银行股份有限公司 | A kind of Risk Identification Method, apparatus and system |
-
2020
- 2020-03-27 CN CN202010229807.8A patent/CN111383102B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190259033A1 (en) * | 2015-06-20 | 2019-08-22 | Quantiply Corporation | System and method for using a data genome to identify suspicious financial transactions |
CN107945024A (en) * | 2017-12-12 | 2018-04-20 | 厦门市美亚柏科信息股份有限公司 | Identify that internet finance borrowing enterprise manages abnormal method, terminal device and storage medium |
CN108596439A (en) * | 2018-03-29 | 2018-09-28 | 北京中兴通网络科技股份有限公司 | A kind of the business risk prediction technique and system of knowledge based collection of illustrative plates |
CN108763556A (en) * | 2018-06-01 | 2018-11-06 | 北京奇虎科技有限公司 | Usage mining method and device based on demand word |
CN109726819A (en) * | 2018-12-29 | 2019-05-07 | 东软集团股份有限公司 | A kind of method and device for realizing reasoning |
CN110134797A (en) * | 2019-04-29 | 2019-08-16 | 贳巽(北京)国际商业数据技术股份公司 | A method of change of financial market is studied and judged based on reason map and multi-sector model |
CN109977237A (en) * | 2019-05-27 | 2019-07-05 | 南京擎盾信息科技有限公司 | A kind of dynamic law occurrence diagram spectrum construction method towards legal field |
CN110363449A (en) * | 2019-07-25 | 2019-10-22 | 中国工商银行股份有限公司 | A kind of Risk Identification Method, apparatus and system |
Non-Patent Citations (1)
Title |
---|
朱福勇 等: "基于图谱融合的人工智能司法数据库构建研究" * |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112131392A (en) * | 2020-08-01 | 2020-12-25 | 赛飞特工程技术集团有限公司 | Public health epidemic situation early warning method and system based on knowledge graph |
CN112598507A (en) * | 2020-12-25 | 2021-04-02 | 北京知因智慧科技有限公司 | Excessive credit granting risk prediction system and method based on knowledge graph |
CN112785414A (en) * | 2021-01-04 | 2021-05-11 | 上海海事大学 | Credit risk prediction method based on knowledge graph and ontology inference engine |
CN112785414B (en) * | 2021-01-04 | 2024-01-26 | 上海海事大学 | Credit risk prediction method based on knowledge graph and ontology inference engine |
CN112785418A (en) * | 2021-01-22 | 2021-05-11 | 深圳前海微众银行股份有限公司 | Credit risk modeling method, credit risk modeling device, credit risk modeling equipment and computer readable storage medium |
CN112785418B (en) * | 2021-01-22 | 2024-02-06 | 深圳前海微众银行股份有限公司 | Credit risk modeling method, apparatus, device and computer readable storage medium |
CN112529681A (en) * | 2021-02-08 | 2021-03-19 | 南京审计大学 | Credit risk transfer method based on credit subject correlation strength |
CN113610626A (en) * | 2021-07-26 | 2021-11-05 | 建信金融科技有限责任公司 | Bank credit risk identification knowledge graph construction method and device, computer equipment and computer readable storage medium |
CN115689720A (en) * | 2022-09-02 | 2023-02-03 | 睿智合创(北京)科技有限公司 | Newly-increased risk monitoring system based on risk attribution |
CN115269879A (en) * | 2022-09-05 | 2022-11-01 | 北京百度网讯科技有限公司 | Knowledge structure data generation method, data search method and risk warning method |
CN116452323A (en) * | 2023-04-28 | 2023-07-18 | 广州市良策网络技术有限公司 | Risk assessment method, system, equipment and storage medium |
CN116452323B (en) * | 2023-04-28 | 2024-01-30 | 广州市良策网络技术有限公司 | Risk assessment method, system, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN111383102B (en) | 2023-10-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111383102B (en) | Financial credit risk identification method, model construction method and device | |
CN109977151B (en) | Data analysis method and system | |
CN110060144B (en) | Method for training credit model, method, device, equipment and medium for evaluating credit | |
US8676726B2 (en) | Automatic variable creation for adaptive analytical models | |
CN110717816A (en) | Artificial intelligence technology-based global financial risk knowledge graph construction method | |
US8224790B2 (en) | Information processor arrangement | |
CA2481432A1 (en) | Processing mixed numeric and/or non-numeric data | |
CN106897930A (en) | A kind of method and device of credit evaluation | |
CN110400082B (en) | Abnormal transaction enterprise identification method and device | |
CN111538794B (en) | Data fusion method, device and equipment | |
Shukla et al. | Comparative analysis of ml algorithms & stream lit web application | |
Ziegler et al. | Injecting semantic background knowledge into neural networks using graph embeddings | |
Fernandez et al. | Counterfactual explanations for data-driven decisions | |
CN114782161A (en) | Method, device, storage medium and electronic device for identifying risky users | |
Muñoz-Cancino et al. | On the dynamics of credit history and social interaction features, and their impact on creditworthiness assessment performance | |
CN113657990A (en) | Ant-lion algorithm optimized NARX neural network risk prediction system and method | |
CN113918660A (en) | API asset management method and device, computer equipment and storage medium | |
CN115994331A (en) | Message sorting method and device based on decision tree | |
US20210073247A1 (en) | System and method for machine learning architecture for interdependence detection | |
CN114648406A (en) | User credit integral prediction method and device based on random forest | |
Bagui | An approach to mining crime patterns | |
Blue et al. | Proving yourself: addressing the refugee identity crisis with Bayesi-chain probability & digital footprints | |
CN111784495B (en) | Guarantee ring identification method and device, computer equipment and storage medium | |
CN115204457A (en) | Loan default risk prediction method based on graph attention network | |
CN113849580A (en) | Subject rating prediction method and device, electronic equipment 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 | ||
GR01 | Patent grant |