CN110458697A

CN110458697A - Method and apparatus for assessing risk

Info

Publication number: CN110458697A
Application number: CN201910762788.2A
Authority: CN
Inventors: 冯博豪
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2019-08-19
Filing date: 2019-08-19
Publication date: 2019-11-15

Abstract

The embodiment of the present disclosure is related to field of cloud calculation, discloses the method and apparatus for assessing risk.Method includes: the application information based on typing, determines the essential information of enterprise；Incidence relation based on enterprise's identification information in essential information and enterprise's identification information in company information library, generates the knowledge mapping of enterprise；Based on essential information, the business datum of enterprise and the public feelings information of enterprise are inquired；Essential information, the knowledge mapping of enterprise, business datum and public feelings information are inputted into risk control model, obtain the risk assessment information of enterprise.The method increase the efficiency and accuracy that obtain risk assessment information.

Description

Method and apparatus for assessing risk

Technical field

This disclosure relates to field of computer technology, and in particular to risk assessment technology field, more particularly, to assessment wind The method and apparatus of danger.

Background technique

With the development of economy, in the market enterprise is also more and more.The available external loan of these enterprises is enterprise The development in future is laid the groundwork, therefore very big for the demand of outside loan.For loan provider, enterprise's application loan It is the important component in loan provider's business source.As enterprise obtains the increase of the external business provided a loan, loan is provided Requirement of the side for the risk control that enterprise obtains loan is also higher and higher.How can accurately assess to enterprise and loan is provided Risk situation when money becomes the problem in loan provider's work.

Currently, mainly manually carrying out loan risk evaluation to enterprise by business personnel.

Summary of the invention

The embodiment of the present disclosure provides the method and apparatus for assessing risk.

In a first aspect, the embodiment of the present disclosure provides a kind of method for assessing risk, comprising: the application based on typing Information determines the essential information of enterprise；It is identified based on enterprise's identification information in essential information and the enterprise in company information library The incidence relation of information generates the knowledge mapping of enterprise；Based on essential information, the business datum of enterprise and the public sentiment of enterprise are inquired Information；Essential information, the knowledge mapping of enterprise, business datum and public feelings information are inputted into risk control model, obtain enterprise Risk assessment information.

In some embodiments, risk control model includes: risk relations model and risk score model；Risk assessment letter Breath includes: to establish the risk probability of service relation, the risk score and service prompts information of enterprise with enterprise；It will believe substantially Breath, the knowledge mapping of enterprise and business datum input risk control model, and the risk assessment information for obtaining enterprise includes: by base This information, the knowledge mapping of enterprise and public feelings information input risk relations model, obtain the risk that service relation is established with enterprise Probability；Business datum and public feelings information are inputted into risk score model, obtain the risk score and service prompts information of enterprise.

In some embodiments, method further includes at least one of following: risk relations model is to be dug based on RFM, NLP, figure The relational network model that pick technology obtains；Risk score model is multivariate regression models；Risk probability includes probability of cheating, promise breaking Probability and overdue probability；And business datum includes: the number of recruits, recruitment time, business circumstance and debt situation.

In some embodiments, based on the application information of typing, determine that the essential information of enterprise includes: the paper for identifying typing The application information of matter version, obtains recognition result；Recognition result is corrected using intelligent correction algorithm, obtains text sequence；Identification text Entity in this sequence obtains the solid data of shoes tagged；And the name solid data based on shoes tagged, export structure The essential information of enterprise.

In some embodiments, based on the application information of typing, determine that the essential information of enterprise includes: the Shen based on typing Please information, transfer reference information corresponding with the enterprise in application information；Wherein, reference information includes official information and/or goes through History information；Based on reference information, application information is verified；It is unverified in response to the result instruction presupposed information of verification, it presents Retract the prompt information of application；Meet blacklist rule in response to the enterprise in the result instruction application information of verification, will apply Blacklist is added in enterprise in information.

In some embodiments, it includes following at least one that the enterprise in the result instruction application information of verification, which meets blacklist, : there are the numbers of non-genuine information more than predetermined threshold in application information；History verifies number and exceeds preset threshold.

In some embodiments, method further include: by risk assessment information input risk control decision model, obtain risk The result of decision of control decision model output.

In some embodiments, risk control decision model is the rule model that serial or parallel multiple rules are formed.

In some embodiments, when risk assessment information includes: to establish risk probability, the enterprise of service relation with enterprise When risk score and service prompts information, multiple rules are included at least: if probability of cheating or Default Probability in risk probability Higher than predetermined probabilities, then the result of decision exported is to retract application；If the credit scoring in risk score is lower than default scoring, wind Income in the scoring of danger is higher than debt ratio threshold value, the then result of decision exported lower than the debt ratio in income threshold value or risk score To retract application.

In some embodiments, method further include: in response to receiving service instruction, establish service relation with enterprise, supervise Business datum of the enterprise of service relation after receiving service has been established in control；Based on the business datum received after servicing, optimize wind Dangerous Controlling model.

Second aspect, the embodiment of the present disclosure provide a kind of for assessing the device of the risk of enterprise, comprising: information determines Unit is configured to the application information based on typing, determines the essential information of enterprise；Map generation unit is configured to be based on The incidence relation of enterprise's identification information in enterprise's identification information and company information library in essential information, generates the knowledge of enterprise Map；Information query unit is configured to inquire the business datum of enterprise and the public feelings information of enterprise based on essential information；It comments Estimate output unit, is configured to essential information, the knowledge mapping of enterprise, business datum and public feelings information inputting risk control mould Type obtains the risk assessment information of enterprise.

In some embodiments, the risk control model assessed in output unit includes: that risk relations model and risk are commented Sub-model；The risk assessment information assessed in output unit includes: to establish the risk probability of service relation, the wind of enterprise with enterprise Danger scoring and service prompts information；Assessment output unit be further configured to: by essential information, enterprise knowledge mapping and Public feelings information inputs risk relations model, obtains the risk probability that service relation is established with enterprise；Business datum and public sentiment are believed Breath input risk score model, obtains the risk score and service prompts information of enterprise.

In some embodiments, device further includes at least one of following: risk relations model is to be dug based on RFM, NLP, figure The relational network model that pick technology obtains；Risk score model is multivariate regression models；Risk probability includes probability of cheating, promise breaking Probability and overdue probability；And business datum includes: the number of recruits, recruitment time, business circumstance and debt situation.

In some embodiments, information determination unit includes: result identification subelement, is configured to identify the papery of typing The application information of version, obtains recognition result；Modified result subelement is configured to using intelligent correction algorithm amendment identification knot Fruit obtains text sequence；Entity recognition subelement is configured to identify the entity in text sequence, obtains the entity of shoes tagged Data；And information exports subelement, is configured to the name solid data based on shoes tagged, the base of the enterprise of export structure This information.

In some embodiments, information determination unit includes: that information transfers subelement, is configured to the application based on typing Information transfers reference information corresponding with the enterprise in application information；Wherein, reference information includes official information and/or history Information；Information checking subelement is configured to verify application information based on reference information；Subelement is presented in prompt, is configured to It is unverified in response to the result instruction presupposed information of verification, the prompt information for retracting application is presented；Subelement is added in list, The enterprise being configured in response in the result instruction application information of verification meets blacklist rule, by the enterprise in application information Blacklist is added.

In some embodiments, the enterprise in result instruction application information verified in information determination unit meets blacklist Include at least one of the following: that there are the numbers of non-genuine information more than predetermined threshold in application information；History verification number exceeds Preset threshold.

In some embodiments, device further include: result determination unit is configured to risk assessment information input risk Control decision model obtains the result of decision of risk control decision model output.

In some embodiments, as a result the risk control decision model in determination unit is serial or parallel multiple regular shapes At rule model.

In some embodiments, when the risk assessment information in assessment output unit includes: to establish service relation with enterprise Risk probability, enterprise risk score and service prompts information when, as a result multiple rules in determination unit include at least: If the probability of cheating or Default Probability in risk probability are higher than predetermined probabilities, the result of decision exported is to retract application；If wind Credit scoring in the scoring of danger is lower than the income in default scoring, risk score lower than the debt in income threshold value or risk score Rate is higher than debt ratio threshold value, then the result of decision exported is to retract application.

In some embodiments, device further include: data monitoring unit is configured in response to receive service instruction, Service relation is established with enterprise, business datum of the enterprise of service relation after receiving service has been established in monitoring；Model optimization list Member is configured to optimize risk control model based on the business datum received after servicing.

The third aspect, the embodiment of the present disclosure provide a kind of electronic equipment/terminal/server, comprising: at one or more Manage device；Storage device, for storing one or more programs；When one or more programs are executed by one or more processors, So that one or more processors realize as above any method for assessing risk.

Fourth aspect, the embodiment of the present disclosure provide a kind of computer-readable medium, are stored thereon with computer program, should As above any method for assessing risk is realized when program is executed by processor.

The method and apparatus for assessing risk that the embodiment of the present disclosure provides, are primarily based on the application information of typing, really Determine the essential information of enterprise；Later, it is identified and is believed based on enterprise's identification information in essential information and the enterprise in company information library The incidence relation of breath generates the knowledge mapping of enterprise；Later, it is based on essential information, inquires business datum and the enterprise of enterprise Public feelings information；Finally, essential information, the knowledge mapping of enterprise, business datum and public feelings information are inputted risk control model, obtain To the risk assessment information of enterprise.In this course, since the input data of risk control model uses essential information, enterprise The data of the various dimensions such as knowledge mapping, business datum and the public feelings information of industry, so that the risk assessment of risk control model output The accuracy of information is higher, and obtains risk assessment information due to using risk control model to replace manually evaluating and testing, and mentions It is high to obtain the efficiency and accuracy of risk assessment information.

Detailed description of the invention

Non-limiting embodiment is described in detail referring to made by the following drawings by reading, the other feature of the disclosure, Objects and advantages will become more apparent upon:

Fig. 1 is that the disclosure can be applied to exemplary system architecture figure therein；

Fig. 2 is the flow diagram according to one embodiment of the method for assessing risk of the embodiment of the present disclosure；

Fig. 3 is an exemplary application scene according to the method for assessing risk of the embodiment of the present disclosure；

Fig. 4 is the flow diagram according to another embodiment of the method for assessing risk of the embodiment of the present disclosure；

Fig. 5 is the exemplary block diagram of one embodiment of the device for assessing risk of the disclosure；

Fig. 6 is adapted for the structural schematic diagram for the computer system for realizing the server of the embodiment of the present disclosure.

Specific embodiment

The disclosure is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to Convenient for description, part relevant to related invention is illustrated only in attached drawing.

It should be noted that in the absence of conflict, the feature in embodiment and embodiment in the disclosure can phase Mutually combination.The disclosure is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

Fig. 1 is shown can be using the method for assessing risk of the disclosure or the implementation of the device for assessing risk The exemplary system architecture 100 of example.

As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105. Network 104 between terminal device 101,102,103 and server 105 to provide the medium of communication link.Network 104 can be with Including various connection types, such as wired, wireless communication link or fiber optic cables etc..

User can be used terminal device 101,102,103 and be interacted by network 104 with server 105, to receive or send out Send message etc..Various telecommunication customer end applications, such as browser application, purchase can be installed on terminal device 101,102,103 Species application, searching class application, instant messaging tools, mailbox client, social platform software etc..

Terminal device 101,102,103 can be hardware, be also possible to software.When terminal device 101,102,103 is hard When part, it can be the various electronic equipments for supporting browser application, including but not limited to tablet computer, pocket computer on knee With desktop computer etc..When terminal device 101,102,103 is software, above-mentioned cited electronic equipment may be mounted at In.It may be implemented into for example for providing the multiple softwares or software module of Distributed Services, also may be implemented into single soft Part or software module.It is not specifically limited herein.

Server 105 can be to provide the server of various services, such as to carrying out on terminal device 101,102,103 Browser application provides the background server supported.Background server can carry out the data such as the request received analyzing etc. Reason, and processing result is fed back into terminal device.

It should be noted that server can be hardware, it is also possible to software.When server is hardware, may be implemented At the distributed server cluster that multiple servers form, individual server also may be implemented into.It, can when server is software To be implemented as that single software or software also for example may be implemented into for providing the multiple softwares or software module of Distributed Services Module.It is not specifically limited herein.

In practice, provided by the embodiment of the present disclosure for assess risk method can by terminal device 101,102, 103 and/or server 105,106 execute, the device for assessing risk also can be set in terminal device 101,102,103 And/or in server 105,106.

It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical.According to realization need It wants, can have any number of terminal device, network and server.

With continued reference to Fig. 2, Fig. 2 shows the processes according to one embodiment of the method for assessing risk of the disclosure 200.This be used for assess risk method the following steps are included:

Step 201, based on the application information of typing, the essential information of enterprise is determined.

In the present embodiment, for assessing the executing subject (such as terminal shown in FIG. 1 or server) of the method for risk The application information that user submits can be received by the human-computer interaction device of Local or Remote.

Specifically, electronic edition application information of the available user of above-mentioned executing subject from front end typing.Electronic edition application Information includes enterprise's essential information, the recent business circumstance of enterprise etc..Data input layer provides front end interface, for needing to apply Enterprise's typing its essential information.

Alternatively or additionally, the papery version application information of the available user's typing of above-mentioned executing subject.In typing Shen Please information when, papery application information specifically includes that legal person's perfect instrument, business license, Tax Registration Certificate, paper application forms etc..

In some optional implementations of the present embodiment, the application information based on typing determines the essential information of enterprise Include: the application information for identifying the papery version of typing, obtains recognition result；Recognition result is corrected using intelligent correction algorithm, is obtained To text sequence；It identifies the entity in text sequence, obtains the solid data of shoes tagged；And the name entity based on shoes tagged Data, the essential information of the enterprise of export structure.

It, can be using the prior art or the technology of future development for the application information of papery version in this implementation In the method for text is identified that the application does not limit this in picture for identification, for example, can identify skill using OCR The application information of art identification papery version.

It identifies obtained recognition result, usually there is more wrong word.Therefore, it is necessary to carry out at data to recognition result Reason, to obtain text sequence.Here data processing can correct the text in recognition result using intelligent correction algorithm, be Further analysis is prepared.Intelligent correction algorithm can be using wrong for correcting sentence in the prior art or the technology of future development Method accidentally, the application do not limit this.For example, tool (pycorrector) can be corrected using Chinese wrong word or followed Ring neural network (RNNLM) language model corrects the mistake in text.

In addition, it is directed to revised text sequence, it can be with binding entity recognizer, it is ensured that the correctness of Entity recognition. Here entity identification algorithms can be the algorithm of entity for identification in the technology of the prior art or future development, the application It does not limit this.For example, the identification of entity in text sequence can be carried out using name Entity recognition (NER) task.

Specifically, the name entity of identification may include:

(1) applicant company related entities such as apply for that title, title preapprove code, residence, production and operation, it is postal Coding, company's type etc.；

(2) personnel's related entities, such as legal representative, management, director, manager, financial administrator, shareholder promoter, connection Network person etc.；

(3) personal information association attributes entity, such as name, fixed-line telephone, mobile phone, identity card type, identification card number Code etc.；And other related entities.These entities can be examined for subsequent foundation and provide related fields term.

The method of the essential information of determination enterprise in this implementation, is corrected pair due to using intelligent correction algorithm The accuracy rate of the text sequence obtained according to recognition result can be improved in the recognition result of the application information of papery version；According to reality Body recognition result obtains the essential information of enterprise, improves the accuracy rate and efficiency of the essential information for obtaining enterprise.

In some optional implementations of the present embodiment, the application information based on typing determines the essential information of enterprise Include: the application information based on typing, transfers reference information corresponding with the enterprise in application information；Wherein, reference information packet Include official information and/or historical information；Based on reference information, application information is verified；In response to the default letter of result instruction of verification It ceases unverified, presents and retract the prompt information of application；Meet in response to the enterprise in the result instruction application information of verification Blacklist is added in enterprise in application information by blacklist rule.

It, can be to the application information of typing in order to verify whether the data in application information are correct in this implementation It is verified.For example, a verification is done to the CompanyAddress in application information, once the practical operation address of discovery company and official Register of company address in information is inconsistent, directly retracts application.This step can transfer the historical information number of application enterprise automatically According to, and with the information in application material carry out it is detailed compare, including business entity, enterprises registration capital etc. information.

It can be to avoid maliciously submitting in addition, being verified to the application information of typing.Can by the result of verification and in advance The blacklist of setting is compared, if result of the comparison indicate that verification result hit blacklist rule, will be in application information Enterprise be added blacklist.

Illustratively, it may include: in application information that the enterprise in the result instruction application information of verification, which meets blacklist, It is more than predetermined threshold there are the number of non-genuine information；History verifies number and exceeds preset threshold.

Specifically, the history of the available applicant of above-mentioned executing subject verifies number, and by history verification number and in advance If threshold value comparison.If it is more than preset threshold that history, which verifies number, blacklist is added in the enterprise customer of application, and export black List increasing person's prompt information.

Alternatively or additionally, the essential information that above-mentioned executing subject can verify out applicant's offer is not real information Number, and by the number of non-genuine information compared with predetermined threshold, if the number of non-genuine information is more than predetermined threshold, Blacklist is added in the enterprise customer of application, and exports blacklist increasing person's prompt information.

The method of the essential information of determination enterprise in this implementation, due to having carried out school to the data in application information It tests, the authenticity and accuracy of the essential information of identified enterprise can be improved, bad application and malice is avoided to apply.

Step 202, the pass based on enterprise's identification information in essential information and enterprise's identification information in company information library Connection relationship generates the knowledge mapping of enterprise.

In the present embodiment, above-mentioned executing subject can be inquired according to enterprise's identification information in the essential information of enterprise There is in company information library with enterprise's identification information in essential information enterprise's identification information of incidence relation.Later, above-mentioned to hold Row main body can have incidence relation according to enterprise's identification information in essential information and enterprise's identification information in essential information Enterprise's identification information and incidence relation between the two, generate the knowledge mapping of enterprise.

Here enterprise's identification information, can be identified enterprise, so that the external world identifies the enterprise.Specifically, enterprise Identification information may include at least one of following: enterprise name, business entity and enterprise trademark etc..

In a specific example, it can inquire in company information library and be somebody's turn to do according to the enterprise name in essential information Incidence relation between the associated enterprise name of enterprise name and these enterprises.Meanwhile it can be looked into from company information library Ask the legal person in essential information and its holding other companies.

Above-mentioned executing subject is obtaining enterprise's identification information in essential information and enterprise's mark letter in company information library After the incidence relation of breath, the data preparation that can be will acquire is at intuitive Company Knowledge map.For example, can be by each enterprise Essential information and the essential information of each legal person be organized into Company Knowledge map.

It should be appreciated that Company Knowledge map can be presented on interactive interface after obtaining Company Knowledge map, and By the information output in Company Knowledge map.

Step 203, it is based on essential information, inquires the business datum of enterprise and the public feelings information of enterprise.

In the present embodiment, the business datum of enterprise and the public feelings information of enterprise can be obtained from various approach.For example, can To obtain the business datum of enterprise and the public feelings information of enterprise from approach such as the network platform, database, third parties.

The business datum of enterprise refers to the data that can indicate the service operation situation of enterprise.For example, the business number of enterprise According at least may include: the number of recruits, recruitment time, business circumstance, debt situation etc..

The public feelings information of enterprise refers to the reflection form for the public sentiment of enterprise.For example, for enterprise media information, Government information etc..

Step 204, essential information, the knowledge mapping of enterprise, business datum and public feelings information are inputted into risk control model, Obtain the risk assessment information of enterprise.

In the present embodiment, risk control model mainly assesses the case where enterprise, and score output best friend to it Mutual interface is used for next step decision.

Here risk control model, can be rule-based model, and rule therein is to be counted according to historical data Obtained rule.Alternatively or additionally, risk control model can be the model based on machine learning algorithm, should be based on machine The model of learning algorithm is obtained based on the study to historical data.The application does not limit this.

After the knowledge mapping of essential information, enterprise, business datum and public feelings information are inputted risk control model, risk Controlling model can analyze the data of input, and the data based on input estimate the corresponding risk assessment of data of input Information.

In some optional implementations of the present embodiment, risk control model includes: that risk relations model and risk are commented Sub-model；Risk assessment information includes: to establish the risk probability of service relation, the risk score of enterprise and service with enterprise to mention Show information；Essential information, the knowledge mapping of enterprise and business datum are inputted into risk control model, the risk for obtaining enterprise is commented Estimating information includes: that essential information, the knowledge mapping of enterprise and public feelings information are inputted risk relations model, obtains and establishes with enterprise The risk probability of service relation；Business datum and public feelings information are inputted into risk score model, obtain the risk score of enterprise with And service prompts information.

In this implementation, knowledge mapping and carriage of the risk relations model for essential information, enterprise based on input Feelings information estimates the risk probability that service relation is established with enterprise.Risk probability can indicate to establish service relation with enterprise Risk.

Herein, risk relations model can be relational network model.Relational network is a kind of data structure based on figure, By node and Bian Zucheng.Each node on behalf individual, each edge are the relationship between individual and individual.Relational network will not Same individual links together according to its relationship, to provide the ability from the angle analysis problem of " relationship ".Relational network Be conducive to be identified from normal behaviour to abnormal clique's fraud.

Application scenarios of the relational network in anti-fraud, are broadly divided into two kinds of situations of monitor model and unsupervised model.Institute The monitor model of meaning refers under the premise of known " good " and " bad " label, attempts from historical data, excavates fraud group The characteristic feature and behavior pattern of partner, so as to effectively identify fraud clique.Although monitor model is in forecasting accuracy On have good performance, still, in actual conditions, " good " and the label of " bad " are often hardly resulted in.Therefore, believe in no label When breath, unsupervised model analysis also becomes particularly important.

Risk score model is used for the business datum based on enterprise, estimates the risk score and output service prompts of enterprise Information.Wherein, risk score can indicate the whole scoring of enterprise, and service prompts information can indicate the service provided to enterprise Grade.

Herein, risk score model can be simple multivariate regression models and be also possible to neural net regression model. In a specific example, risk score model can be the linear regression model that training obtain.In linear regression model (LRM) The value of dependent variable is risk score.Independent variable is business datum and public feelings information.Specifically, business datum may include: profit Growth rate, the number of recruits growth rate, the number of recruits, corporate debt rate etc..Model can be simple multivariate regression models can also To be neural net regression model

Risk control model in this implementation, by the way that risk relations model and risk are arranged in risk control model Rating Model, can be using the determining risk probability for establishing service relation with enterprise of risk relations model, and uses risk score Model obtains the risk score and service prompts information of enterprise, so that different input datas is directed to, using different models Special treatment is carried out, the accuracy rate of the result of output is improved.

In some optional implementations of the present embodiment, the method for assessing risk of the disclosure further include with down toward One item missing: risk relations model is the relational network model obtained based on RFM, NLP, figure digging technology；Risk score model is Multivariate regression models；Risk probability includes probability of cheating, Default Probability and overdue probability；And business datum includes: recruitment people Number, recruitment time, business circumstance and debt situation.

In this implementation, RFM model is customer relationship management model, and NLP is Natural Language Processing Models, and figure excavates Technology, which refers to the process of, finds from mass data using graph model and lifts useful knowledge and information.It is being based on RFM, NLP, figure After digging technology constructs relational network model, essential information, the knowledge mapping of enterprise and business datum can be inputted and be closed It is network model, to obtain establishing the probability of cheating of service relation, Default Probability with enterprise and exceeding for relational network model output Phase probability.

Multivariate regression models can make the number of recruits, recruitment time, business circumstance and the debt situation in business datum For independent variable, using the risk score of enterprise as dependent variable, and service prompts information is generated based on dependent variable.

The method for assessing risk of disclosure above-described embodiment, so that the risk assessment letter of risk control model output The accuracy of breath is higher, and improves to obtain the efficiency and accuracy of risk assessment information.

Below in conjunction with Fig. 3, the exemplary application scene of the method for assessing risk of the disclosure is described.

As shown in figure 3, Fig. 3 shows an exemplary application field of the method for assessing risk according to the disclosure Scape.

As shown in figure 3, the method 300 for assessing risk is run in electronic equipment 310, may include:

Firstly, the application information 301 based on typing, determines the essential information 302 of enterprise；

Based on the incidence relation 305 of the entity 304 in the entity 303 and company information library in essential information 302, enterprise is generated The knowledge mapping 306 of industry；

Based on essential information 302, the business datum 307 of enterprise and the public feelings information 308 of enterprise are inquired；

Essential information 302, the knowledge mapping 306 of enterprise, business datum 307 and public feelings information 308 are inputted into risk control Simulation 309 obtains the risk assessment information 310 of enterprise.

It should be appreciated that for assessing the application scenarios of the method for risk shown in above-mentioned Fig. 3, only for for commenting Estimate the exemplary description of the method for risk, does not represent the restriction to this method.For example, each step shown in above-mentioned Fig. 3, It can be further using the implementation method of more details.It can also be further increased other for commenting on the basis of above-mentioned Fig. 3 The step of estimating risk.

Another embodiment of the method for assessing risk according to the disclosure is shown with further reference to Fig. 4, Fig. 4 Schematic flow chart.

As shown in figure 4, the method 400 for being used to assess risk of the present embodiment, may comprise steps of:

Step 401, based on the application information of typing, the essential information of enterprise is determined.

Step 402, the incidence relation based on the entity in the entity and company information library in essential information, generates enterprise Knowledge mapping.

Step 403, it is based on essential information, inquires the business datum of enterprise and the public feelings information of enterprise.

Step 404, essential information, the knowledge mapping of enterprise, business datum and public feelings information are inputted into risk control model, Obtain the risk assessment information of enterprise.

It should be appreciated that above-mentioned steps 401 to step 404 respectively with the step 201 in embodiment shown in Fig. 2 to step 204 is corresponding.Therefore, same for operation and feature described in step 201 to step 204 in above-mentioned embodiment shown in Fig. 2 Sample is suitable for step 401 to step 404, and details are not described herein.

Step 405, by risk assessment information input risk control decision model, the output of risk control decision model is obtained The result of decision.

In the present embodiment, risk control decision model can carry out intelligent decision according to risk assessment information, and defeated The result of decision out.The entire decision process of risk control decision model is completed by model, is participated in without artificial.Decision is completed Afterwards, the result of decision can be output to interactive interface by risk control decision model, for artificial reference.

Risk control decision model is the rule model formed by multiple rules by mode in series or in parallel.

For example, the rule that risk control decision model includes, can be enterprise's probability of cheating that risk relations model obtains Or Default Probability is not above predetermined probabilities.Enterprise's probability of cheating that risk control decision model is obtained in risk relations model Or when being higher than predetermined probabilities of Default Probability, it is believed that enterprise is inclined to stronger fraud, the result of decision of output are as follows: it is recommended that Retract application.

In another example the Else Rule that risk control decision model includes can also include commenting for the credit in risk score Divide, take in the restriction of situations such as situation, debt ratio.When credit scoring is lower than default scoring, income is lower than income threshold value or negative When debt rate is higher than debt ratio threshold value, the result of decision of output are as follows: it is recommended that retracting application.

It is understood that these rules can be by manually setting via interactive interface.

The result of decision can be presented in interactive interface after generating the result of decision by risk control decision model.Into one Step ground, can also be presented in interactive interface for the application information for generating the result of decision or intermediate data.

Optional step 406 establishes service relation with enterprise, monitoring has been established service and closes in response to receiving service instruction Business datum of the enterprise of system after receiving service.

In the present embodiment, above-mentioned executing subject can be directly to suggest establishing service relation with enterprise in the result of decision When, it is directly indicated the result of decision as service, establishes service relation with enterprise.Alternatively, above-mentioned executing subject can will determine Plan result is presented to the user, and is received the service for establishing service relation with enterprise that user is inputted based on the result of decision and is indicated, with enterprise Industry establishes service relation.

Above-mentioned executing subject can monitor after establishing service relation with enterprise and the enterprise of service relation have been established connect Business datum after being serviced, in case subsequent adjust the service provided to enterprise for the business datum received after servicing.

Optional step 407 optimizes risk control model based on the business datum received after servicing.

In the present embodiment, business of the enterprise of service relation after receiving service has been established in monitoring in above-mentioned executing subject After data, the business datum of monitoring can be analyzed, so as to adjusting and optimizing risk control model.

The method for assessing risk in embodiment in disclosure Fig. 4, shown in Fig. 2 for assessing risk On the basis of method, risk assessment information input risk control decision model can be obtained into the output of risk control decision model The result of decision, compared with manual examination and verification in the prior art, reduce auditor repeated redundancy labour, improve assessment The working efficiency of risk, and approval process can be shortened, it avoids the difference due to auditor's level and leads to auditing standards Different, phenomena such as approval results differ greatly generation.In some embodiments, due to increasing optional step 406 and optional Step 407, the business datum that can monitor enterprise is reduced to enterprise and provides service but can not withdraw the risk of service revenue.

With further reference to Fig. 5, as the realization to method shown in above-mentioned each figure, the embodiment of the present disclosure provides one kind and is used for One embodiment of the device of risk is assessed, the Installation practice is corresponding with embodiment of the method shown in Fig. 2-Fig. 4, the dress Setting specifically can be applied to include in the device of publishing side and server-side.

As shown in figure 5, the present embodiment may include: information determination unit 510, quilt for assessing the device 500 of risk It is configured to the application information based on typing, determines the essential information of enterprise；Map generation unit 520 is configured to based on basic The incidence relation of enterprise's identification information in enterprise's identification information and company information library in information, generates the knowledge graph of enterprise Spectrum；Information query unit 530 is configured to inquire the business datum of enterprise and the public feelings information of enterprise based on essential information；It comments Estimate output unit 540, is configured to essential information, the knowledge mapping of enterprise, business datum and public feelings information inputting risk control Simulation obtains the risk assessment information of enterprise.

In some optional implementations of the present embodiment, the risk control model assessed in output unit 540 includes: wind Dangerous relational model and risk score model；Risk assessment information in assessment output unit 540 includes: to establish service pass with enterprise The risk probability of system, the risk score of enterprise and service prompts information；Assessment output unit 540 is further configured to: will Essential information, the knowledge mapping of enterprise and public feelings information input risk relations model, obtain the wind that service relation is established with enterprise Dangerous probability；Business datum and public feelings information are inputted into risk score model, obtain the risk score and service prompts letter of enterprise Breath.

In some optional implementations of the present embodiment, device 500 further includes at least one of following: risk relations model For the relational network model obtained based on RFM, NLP, figure digging technology；Risk score model is multivariate regression models；Risk is general Rate includes probability of cheating, Default Probability and overdue probability；And business datum includes: the number of recruits, recruitment time, business circumstance With debt situation.

In some optional implementations of the present embodiment, information determination unit 510 includes (not shown): result is known Small pin for the case unit is configured to identify the application information of the papery version of typing, obtains recognition result；Modified result subelement is matched It is set to and recognition result is corrected using intelligent correction algorithm, obtain text sequence；Entity recognition subelement is configured to identify text Entity in sequence obtains the solid data of shoes tagged；And information exports subelement, is configured to the name based on shoes tagged Solid data, the essential information of the enterprise of export structure.

In some optional implementations of the present embodiment, information determination unit 510 includes (not shown): information tune Subelement is taken, the application information based on typing is configured to, transfers reference information corresponding with the enterprise in application information；Its In, reference information includes official information and/or historical information；Information checking subelement is configured to based on reference information, verification Application information；Subelement is presented in prompt, and the result instruction presupposed information for being configured in response to verification is unverified, and presentation is moved back Return the prompt information of application；Subelement is added in list, the enterprise being configured in response in the result instruction application information of verification Meet blacklist rule, blacklist is added in the enterprise in application information.

In some optional implementations of the present embodiment, the result verified in information determination unit 510 indicates letter of application Enterprise in breath meets blacklist and includes at least one of the following: that there are the numbers of non-genuine information more than predetermined threshold in application information Value；History verifies number and exceeds preset threshold.

In some optional implementations of the present embodiment, device further include: result determination unit 550, be configured to by Risk assessment information input risk control decision model obtains the result of decision of risk control decision model output.

In some optional implementations of the present embodiment, as a result the risk control decision model in determination unit 550 is The rule model that serial or parallel multiple rules are formed.

In some optional implementations of the present embodiment, when the risk assessment information in assessment output unit 540 includes: With enterprise establish the risk probability of service relation, enterprise risk score and service prompts information when, as a result determination unit 550 In multiple rules include at least: if probability of cheating or Default Probability in risk probability are higher than predetermined probabilities, export certainly Plan result is to retract application；If the credit scoring in risk score is lower than the income in default scoring, risk score lower than income Debt ratio in threshold value or risk score is higher than debt ratio threshold value, then the result of decision exported is to retract application.

In some optional implementations of the present embodiment, device further include: data monitoring unit 560 is configured to ring Ying Yu receives service instruction, establishes service relation with enterprise, the enterprise of service relation has been established after receiving service in monitoring Business datum；Model optimization unit 570 is configured to optimize risk control model based on the business datum received after servicing.

It should be appreciated that each unit recorded in device 500 and each step recorded in the method for reference Fig. 2-Fig. 4 description It is rapid corresponding.Device 500 and each list wherein included are equally applicable to above with respect to the operation and feature of method description as a result, Member, details are not described herein.

Below with reference to Fig. 6, it illustrates the electronic equipment that is suitable for being used to realize embodiment of the disclosure, (example is as shown in figure 1 Server or terminal device) 600 structural schematic diagram.Terminal device in embodiment of the disclosure can include but is not limited to all Such as laptop, desktop computer.Terminal device/server shown in Fig. 6 is only an example, should not be to the disclosure Embodiment function and use scope bring any restrictions.

As shown in fig. 6, electronic equipment 600 may include processing unit (such as central processing unit, graphics processor etc.) 601, random access can be loaded into according to the program being stored in read-only memory (ROM) 602 or from storage device 608 Program in memory (RAM) 603 and execute various movements appropriate and processing.In RAM 603, it is also stored with electronic equipment Various programs and data needed for 600 operations.Processing unit 601, ROM 602 and RAM 603 pass through the phase each other of bus 604 Even.Input/output (I/O) interface 605 is also connected to bus 604.

In general, following device can connect to I/O interface 605: including such as touch screen, touch tablet, keyboard, mouse, taking the photograph As the input unit 606 of head, microphone, accelerometer, gyroscope etc.；Including such as liquid crystal display (LCD), loudspeaker, vibration The output device 607 of dynamic device etc.；Storage device 608 including such as tape, hard disk etc.；And communication device 609.Communication device 609, which can permit electronic equipment 600, is wirelessly or non-wirelessly communicated with other equipment to exchange data.Although Fig. 6 shows tool There is the electronic equipment 600 of various devices, it should be understood that being not required for implementing or having all devices shown.It can be with Alternatively implement or have more or fewer devices.Each box shown in Fig. 6 can represent a device, can also root According to needing to represent multiple devices.

Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium On computer program, which includes the program code for method shown in execution flow chart.In such reality It applies in example, which can be downloaded and installed from network by communication device 609, or from storage device 608 It is mounted, or is mounted from ROM 602.When the computer program is executed by processing unit 601, the implementation of the disclosure is executed The above-mentioned function of being limited in the method for example.It should be noted that computer-readable medium described in embodiment of the disclosure can be with It is computer-readable signal media or computer readable storage medium either the two any combination.It is computer-readable Storage medium for example may be-but not limited to-the system of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, device or Device, or any above combination.The more specific example of computer readable storage medium can include but is not limited to: have The electrical connection of one or more conducting wires, portable computer diskette, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD- ROM), light storage device, magnetic memory device or above-mentioned any appropriate combination.In embodiment of the disclosure, computer Readable storage medium storing program for executing can be any tangible medium for including or store program, which can be commanded execution system, device Either device use or in connection.And in embodiment of the disclosure, computer-readable signal media may include In a base band or as the data-signal that carrier wave a part is propagated, wherein carrying computer-readable program code.It is this The data-signal of propagation can take various forms, including but not limited to electromagnetic signal, optical signal or above-mentioned any appropriate Combination.Computer-readable signal media can also be any computer-readable medium other than computer readable storage medium, should Computer-readable signal media can send, propagate or transmit for by instruction execution system, device or device use or Person's program in connection.The program code for including on computer-readable medium can transmit with any suitable medium, Including but not limited to: electric wire, optical cable, RF (radio frequency) etc. or above-mentioned any appropriate combination.

Above-mentioned computer-readable medium can be included in above-mentioned electronic equipment；It is also possible to individualism, and not It is fitted into the electronic equipment.Above-mentioned computer-readable medium carries one or more program, when said one or more When a program is executed by the electronic equipment, so that the electronic equipment: the application information based on typing determines the basic letter of enterprise Breath；Incidence relation based on enterprise's identification information in essential information and enterprise's identification information in company information library, generates enterprise The knowledge mapping of industry；Based on essential information, the business datum of enterprise and the public feelings information of enterprise are inquired；By essential information, enterprise Knowledge mapping, business datum and public feelings information input risk control model, obtain the risk assessment information of enterprise.

The behaviour for executing embodiment of the disclosure can be write with one or more programming languages or combinations thereof The computer program code of work, described program design language include object oriented program language-such as Java, Smalltalk, C++ further include conventional procedural programming language-such as " C " language or similar program design language Speech.Program code can be executed fully on the user computer, partly be executed on the user computer, as an independence Software package execute, part on the user computer part execute on the remote computer or completely in remote computer or It is executed on server.In situations involving remote computers, remote computer can pass through the network of any kind --- packet Include local area network (LAN) or wide area network (WAN) --- it is connected to subscriber computer, or, it may be connected to outer computer (such as It is connected using ISP by internet).

Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the disclosure, method and computer journey The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction Combination realize.

Being described in unit involved in the embodiment of the present disclosure can be realized by way of software, can also be by hard The mode of part is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor packet Include information determination unit, map generation unit, information query unit and assessment output unit.Wherein, the title of these units exists The restriction to the unit itself is not constituted in the case of certain, for example, information determination unit is also described as " based on typing Application information, determine the unit of the essential information of enterprise ".

Above description is only the preferred embodiment of the disclosure and the explanation to institute's application technology principle.Those skilled in the art Member is it should be appreciated that invention scope involved in the disclosure, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic Scheme, while should also cover in the case where not departing from foregoing invention design, it is carried out by above-mentioned technical characteristic or its equivalent feature Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed in the disclosure Can technical characteristic replaced mutually and the technical solution that is formed.

Claims

1. a kind of method for assessing the risk of enterprise, comprising:

Application information based on typing determines the essential information of enterprise；

Incidence relation based on enterprise's identification information in the essential information and enterprise's identification information in company information library, it is raw At the knowledge mapping of enterprise；

Based on the essential information, the business datum of the enterprise and the public feelings information of the enterprise are inquired；

The essential information, the knowledge mapping of the enterprise, the business datum and the public feelings information are inputted into risk control Model obtains the risk assessment information of enterprise.

2. according to the method described in claim 1, wherein, the risk control model includes: that risk relations model and risk are commented Sub-model；

The risk assessment information include: with the enterprise establish the risk probability of service relation, enterprise risk score and Service prompts information；

It is described that the essential information, the knowledge mapping of enterprise and the business datum are inputted into risk control model, it is looked forward to The risk assessment information of industry includes: that the essential information, the knowledge mapping of the enterprise and the public feelings information are inputted risk Relational model obtains the risk probability that service relation is established with the enterprise；The business datum and the public feelings information is defeated Enter risk Rating Model, obtains the risk score and service prompts information of the enterprise.

3. according to the method described in claim 2, wherein, the method also includes at least one of following:

The risk relations model is the relational network model obtained based on RFM, NLP, figure digging technology；

The risk score model is multivariate regression models；

The risk probability includes probability of cheating, Default Probability and overdue probability；And

The business datum includes: the number of recruits, recruitment time, business circumstance and debt situation.

4. according to the method described in claim 1, wherein, the application information based on typing determines the essential information of enterprise Include:

The application information for identifying the papery version of typing, obtains recognition result；

Recognition result is corrected using intelligent correction algorithm, obtains text sequence；

It identifies the entity in the text sequence, obtains the solid data of shoes tagged；And

Name solid data based on the shoes tagged, the essential information of the enterprise of export structure.

5. method according to any of claims 1-4, wherein the application information based on typing determines enterprise Essential information include:

Application information based on typing transfers reference information corresponding with the enterprise in application information；Wherein, the reference information Including official information and/or historical information；

Based on the reference information, the application information is verified；

It is unverified in response to the result instruction presupposed information of verification, the prompt information for retracting application is presented；

Indicate that the enterprise in the application information meets blacklist rule in response to the result of verification, it will be in the application information Blacklist is added in enterprise.

6. according to the method described in claim 5, wherein, the result of the verification indicates that the enterprise in the application information meets Blacklist includes at least one of the following:

There are the numbers of non-genuine information more than predetermined threshold in application information；

History verifies number and exceeds preset threshold.

7. method described in -6 any one according to claim 1, wherein the method also includes:

By the risk assessment information input risk control decision model, the decision knot of risk control decision model output is obtained Fruit.

8. according to the method described in claim 7, wherein, the risk control decision model is serial or parallel multiple regular shapes At rule model.

9. according to the method described in claim 8, wherein, when the risk assessment information includes: and the enterprise establishes and services When the risk probability of relationship, the risk score of enterprise and service prompts information, the multiple rule is included at least:

If the probability of cheating or Default Probability in the risk probability are higher than predetermined probabilities, the result of decision exported is to retract Shen Please；

If credit scoring in the risk score lower than the income in default scoring, the risk score lower than income threshold value or Debt ratio in the risk score is higher than debt ratio threshold value, then the result of decision exported is to retract application.

10. method described in -9 any one according to claim 1, wherein the method also includes:

In response to receiving service instruction, service relation is established with the enterprise, the enterprise that service relation has been established in monitoring is connecing Business datum after being serviced；

Receive the business datum after servicing based on described, optimizes the risk control model.

11. a kind of for assessing the device of the risk of enterprise, comprising:

Information determination unit is configured to the application information based on typing, determines the essential information of enterprise；

Map generation unit is configured to based on enterprise's identification information in the essential information and the enterprise in company information library The incidence relation of identification information generates the knowledge mapping of enterprise；

Information query unit, be configured to inquire based on the essential information enterprise business datum and the enterprise Public feelings information；

Output unit is assessed, is configured to the essential information, the knowledge mapping of the enterprise, the business datum and described Public feelings information inputs risk control model, obtains the risk assessment information of enterprise.

12. device according to claim 11, wherein the risk control model packet in the assessment output unit It includes: risk relations model and risk score model；The risk assessment information in the assessment output unit include: with it is described Enterprise establishes the risk probability of service relation, the risk score of enterprise and service prompts information；

The assessment output unit is further configured to: by the essential information, the knowledge mapping of the enterprise and the carriage Feelings information input risk relations model obtains the risk probability that service relation is established with the enterprise；By the business datum and The public feelings information inputs risk score model, obtains the risk score and service prompts information of the enterprise.

13. device according to claim 12, wherein described device further includes at least one of following:

The risk score model is multivariate regression models；

14. device according to claim 11, wherein the information determination unit includes:

As a result it identifies subelement, is configured to identify the application information of the papery version of typing, obtains recognition result；

Modified result subelement is configured to correct recognition result using intelligent correction algorithm, obtains text sequence；

Entity recognition subelement is configured to identify the entity in the text sequence, obtains the solid data of shoes tagged；And

Information exports subelement, is configured to the name solid data based on the shoes tagged, the base of the enterprise of export structure This information.

15. device described in 1-14 any one according to claim 1, wherein the information determination unit includes:

Information transfers subelement, is configured to the application information based on typing, transfers base corresponding with the enterprise in application information Calibration information；Wherein, the reference information includes official information and/or historical information；

Information checking subelement is configured to verify the application information based on the reference information；

Subelement is presented in prompt, and the result instruction presupposed information for being configured in response to verification is unverified, and Shen is retracted in presentation Prompt information please；

Subelement is added in list, and the result for being configured in response to verification indicates that the enterprise in the application information meets blacklist Blacklist is added in enterprise in the application information by rule.

16. device according to claim 15, wherein described in the result instruction of verification described in the information determination unit Enterprise in application information meets blacklist and includes at least one of the following:

History verifies number and exceeds preset threshold.

17. device described in 1-16 any one according to claim 1, wherein described device further include:

As a result determination unit is configured to the risk assessment information input risk control decision model obtaining risk control The result of decision of decision model output.

18. device according to claim 17, wherein the risk control decision model in the result determination unit is string The rule models that company or multiple rules in parallel are formed.

19. device according to claim 18, wherein the risk assessment packet in the assessment output unit Include: with the enterprise establish the risk probability of service relation, enterprise risk score and service prompts information when, the result Multiple rules in determination unit include at least:

20. device described in 1-19 any one according to claim 1, wherein described device further include:

Data monitoring unit is configured in response to receive service instruction, establishes service relation with the enterprise, monitor built Business datum of the enterprise of vertical service relation after receiving service；

Model optimization unit is configured to receive the business datum after servicing based on described, optimizes the risk control model.

21. a kind of electronic equipment/terminal/server, comprising:

One or more processors；

Storage device, for storing one or more programs；

When one or more of programs are executed by one or more of processors, so that one or more of processors are real The now method as described in any in claim 1-10.

22. a kind of computer-readable medium, is stored thereon with computer program, such as right is realized when which is executed by processor It is required that any method in 1-10.