CN109359818A - A kind of pair of enterprise carries out the method and six rib star models of risk quantification assessment - Google Patents
A kind of pair of enterprise carries out the method and six rib star models of risk quantification assessment Download PDFInfo
- Publication number
- CN109359818A CN109359818A CN201811097238.5A CN201811097238A CN109359818A CN 109359818 A CN109359818 A CN 109359818A CN 201811097238 A CN201811097238 A CN 201811097238A CN 109359818 A CN109359818 A CN 109359818A
- Authority
- CN
- China
- Prior art keywords
- level index
- index
- enterprise
- company
- level
- 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.)
- Pending
Links
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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses methods and six rib star models that a kind of pair of enterprise carries out risk quantification assessment, including first class index, two-level index, three-level index, the two-level index is made of management state, Industry situation, innovation ability, team's correlation, conjunction regulation degree, social effectiveness；The three-level index is refinement index corresponding to the two-level index；The two-level index is calculated by corresponding three-level index by weighted average mode, and the first class index is calculated by all two-level index by weighted average mode.This model can carry out quantitative evaluation to the risk of enterprise, and first class index is the comprehensive score of enterprise, by first class index it can be concluded that the integration capability of enterprise；Two-level index can evaluate enterprise compared with other enterprises of the same trade in the performance of certain dimension, by the calculating of model, the Comprehensive Evaluation of enterprise is measured, obtains an objective appraisal standard, make financial institution and business risk model management all evidence-based, it is reasonable to look into.
Description
Technical field
The present invention relates to information security fields, and the method and six of risk quantification assessment is carried out more particularly to a kind of pair of enterprise
Rib star model.
Background technique
Business risk quantitative evaluation is the core component of Enterprise Risk Management, and the quantization of risk can be used to enterprise
Whether can occurrence risk predicted and managed.Now there is presently no a kind of model can the risk to enterprise completely measured
Change assessment, Risk-warning can not be thus provided for enterprise and cannot be compared with the integration capability of enterprise of the same trade.
Summary of the invention
In view of the above shortcomings of the prior art, the present invention provides a kind of pair of enterprise carry out risk quantification assessment method and
Six rib star models can be realized and carry out quantitative evaluation to the risk of enterprise, and quantized result as enterprise itself Risk-warning with
And the foundation with Enterprise Integrated ability of the same trade comparison.The six each dimension of rib star ＂ of ＂ is 0-100 point, point higher expression enterprise this
Item performance is better, otherwise performance is poorer.
In order to solve the above-mentioned technical problem, present invention employs the following technical solutions:
A kind of pair of enterprise carries out six rib star models of risk quantification assessment, including first class index, two-level index, three-level refer to
Mark, the two-level index are made of management state, Industry situation, innovation ability, team's correlation, conjunction regulation degree, social effectiveness；
The three-level index is refinement index corresponding to the two-level index；The first class index is by all two-level index by adding
Weight average mode is calculated.
As optimization, the management state include operation life in the three-level index, registered capital, acceptance of the bid number,
Acceptance of the bid total amount；
The Industry situation includes that small micro- business event unicity, the large and medium-sized enterprise's design in the three-level index are multiple
Professional, in the recent period advantageous policy quantity, in the recent period unfavorable policy quantity, affiliated enterprise's same industry ratio；
The innovation ability includes trade mark quantity in the three-level index, every year on average trade mark quantity, patent numbers, flat
Annual patent numbers, the last publication patent time, the last publication trade mark time；
Team's correlation includes the news quantity for being related to arrears of wages in the three-level index, is related to what legal person ran away
City numbers that news quantity, the quantity for being related to company management risk news, business are related to, team development speed, team are average
Wages；
The conjunction regulation degree include being listed in the three-level index break one's promise be performed list number, by financial institution rise
It tells number, be related to emphasis case and related to by number, affiliated company to tell situation；
The social effectiveness includes press exposure rate, affiliated company's total quantity, company's recruitment in the three-level index
Public praise, news combined influence power.
As optimization, the two-level index is calculated by corresponding three-level index by weighted average mode.
The method that a kind of pair of enterprise carries out risk quantification assessment, risk quantification assessment, institute are carried out based on six rib star models
Stating six rib star models includes first class index, two-level index, three-level index, and the two-level index is by management state, Industry situation, wound
New ability, team is related, closes regulation degree, social effectiveness composition；The three-level index is thin corresponding to the two-level index
Change index；The first class index is calculated by all two-level index by weighted average mode, and the method includes following
Step:
A. structural data is converted by data analysis technique by the three-level index, the structural data is carried out
Data cleansing makes the data format of the three-level index be unified into number format, the specific steps are as follows:
1) web crawlers engineer obtains all initial data of enterprise using crawler technology, by using to initial data
Natural language processing means parse the specific fields in the initial data, entity extracts, storage, to what is parsed
Structural data carries out ElasticSearch high-volume data loading；
2) the relevant field progress specification of the structural data of storage, statistics, grouping, sequence are obtained using SQL statement
The quantitative value of certain field；
B. each numeric formatted three-level index is normalized as the quantized value between 0 to 1；
C., the quantized value of the three-level index is calculated to the numerical value of the two-level index: E by weighted average modei
=100* ∑ (Aij*Wij)；Wherein i=1,2...6, j=1,2...n；EiFor the two-level index score, AijIt is described any
The quantized value of the corresponding three-level index of two-level index, WijFor the corresponding weight of corresponding j-th of three-level index of i-th of two-level index
Coefficient, n are the quantity of three-level index；；
D. the level-one is calculated by weighted average mode in the numerical value of all two-level index for being weighted and averaged and obtaining
The numerical value of index:
S=∑ (Ei*Vi)；Wherein i=1,2 ... 6, wherein S is the first class index score, EiFor the two-level index
Score, ViFor the corresponding weight coefficient of the two-level index；
The public affairs of " bad " that the company and N family for e. extracting " good " that the M family that bank or financial institution provide generally acknowledges generally acknowledge
The three-level index of department and by three-level quantification of targets be numerical value, as the input of the six ribs star model, by the way that the three-level is referred to
Marker tape enters the numerical value that first class index is calculated in the six ribs star model, and the scores binding test of all test companies is public
" good ", " bad " two tag along sort of department, test company is assessed using Receiver operating curve in the six ribs star mould
The numerical value of first class index in type passes through the first class index score of Receiver operating curve's test method binding test company
And true two tag along sort of test company, best score threshold value is obtained, by judging that Receiver operating curve obtains
Best rate of precision；M, N is the positive integer not less than 100；；
F. it when the rate of precision when test data on model is less than the first preset value, then needs according to two-level index and three-level
The significance level of index readjusts the weight of two-level index and three-level index, or adds or deletes the dependency rule word of enterprise
Section, more than the second preset value of model rate of precision, then completes the adjustment of the six ribs star model；
G. when the rate of precision when test data on the six ribs star model is greater than the first preset value, six rib star models are completed
Exploitation, the later period according to the feedback of user market, the regular weight of the two-level index to the six ribs star model and three-level index
It is adjusted, adaptation of the six ribs star model to more and more companies is continuously improved, enhances its generalized ability and accuracy；
H. the enterprise for carrying out risk quantification assessment is needed to substitute into the six rib star models that the exploitation is completed client, with output
Risk quantification assessment result.
Specific steps as optimization, in the step b are as follows: pass through the quantity and the three-level index of the three-level index
Quantized value between establish a Function Mapping relationship, to calculate the quantized value of all three-level indexs；Each three-level
Index has independent function, defines the three-level index as positive message or negative news；The three-level index of positive message is used
Sigmoid function is brought into, and the three-level index of negative news is brought into anti-Sigmoid function.
As optimization, the expression formula of the rate of precision is as follows:
(I-J)/I；
Wherein, I is the sum for bringing " good " company of six rib star models into, and J is " good " public affairs obtained by six rib star models
Department and reality are also the quantity of " good " company.
As optimization, the management state include operation life in the three-level index, registered capital, acceptance of the bid number,
Acceptance of the bid total amount；
The Industry situation includes that small micro- business event unicity, the large and medium-sized enterprise's design in the three-level index are multiple
Professional, in the recent period advantageous policy quantity, in the recent period unfavorable policy quantity, affiliated enterprise's same industry ratio；
The innovation ability includes trade mark quantity in the three-level index, every year on average trade mark quantity, patent numbers, flat
Annual patent numbers, the last publication patent time, the last publication trade mark time；
Team's correlation includes the news quantity for being related to arrears of wages in the three-level index, is related to what legal person ran away
City numbers that news quantity, the quantity for being related to company management risk news, business are related to, team development speed, team are average
Wages；
The conjunction regulation degree include being listed in the three-level index break one's promise be performed list number, by financial institution rise
It tells number, be related to emphasis case and related to by number, affiliated company to tell situation；
The social effectiveness includes press exposure rate, affiliated company's total quantity, company's recruitment in the three-level index
Public praise, news combined influence power.
As optimization, the three-level index of the front message has operation life, registered capital, acceptance of the bid number, the total gold of acceptance of the bid
Volume, small micro- business event unicity, large and medium-sized enterprise design multiple professionals, in the recent period advantageous policy quantity, affiliated enterprise colleague
Industry ratio, trade mark quantity, every year on average trade mark quantity, patent numbers, every year on average patent numbers, the last publication patent when
Between, the city numbers that are related to of the last publication trade mark time, business, team development speed, team be averaged wages, press exposure
Rate, affiliated company's total quantity, company recruit public praise, news combined influence power；The three-level index of the negative news has in the recent period not
Sharp policy quantity, the news quantity for being related to arrears of wages are related to news quantity that legal person runs away, are related to company management risk news
Quantity, be listed in break one's promise be performed list number, by financial institution prosecution number, be related to emphasis case by number, affiliated company
It relates to and tells situation.
As optimization, first preset value is 80%, and second preset value is 90%.
The beneficial effects of the present invention are:
This model can carry out quantitative evaluation to the risk of enterprise, and first class index is the comprehensive score of enterprise, pass through level-one
Index it can be concluded that enterprise integration capability；Two-level index can evaluate enterprise compared with other enterprises of the same trade a certain
The performance of dimension measures the Comprehensive Evaluation of enterprise, obtains an objective appraisal standard by the calculating of model, allows finance
Mechanism and business risk model management are all evidence-based, reasonable to look into.
Detailed description of the invention
Fig. 1 is the flow chart for the method that a kind of pair of enterprise of the present invention carries out risk quantification assessment.
Fig. 2 is the composed structure schematic diagram of six rib star models of the invention.
Fig. 3 is the embodiment for the method that a kind of pair of enterprise of the present invention carries out risk quantification assessment.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawing.
As shown in Fig. 2, a kind of pair of enterprise carries out six rib star models of risk quantification assessment, including first class index, second level refer to
Mark, three-level index, two-level index is by management state, Industry situation, innovation ability, team's correlation, conjunction regulation degree, social effectiveness
Composition；Three-level index is refinement index corresponding to two-level index；First class index passes through weighted average by all two-level index
Mode is calculated.
In the present embodiment, management state includes that operation life in three-level index, registered capital, acceptance of the bid number, acceptance of the bid are total
The amount of money；
Industry situation include small micro- business event unicity in three-level index, large and medium-sized enterprise design multiple professionals,
Advantageous policy quantity, in the recent period unfavorable policy quantity, affiliated enterprise's same industry ratio in the recent period；
Innovation ability includes trade mark quantity in three-level index, every year on average trade mark quantity, patent numbers, every year on average specially
Sharp quantity, the last publication patent time, the last publication trade mark time；
Team's correlation includes the news quantity for being related to arrears of wages in three-level index, is related to the news number that legal person runs away
It measures, the city numbers that the quantity, the business that are related to company management risk news is related to, team development speed, team are averaged wages；
Close regulation degree include being listed in three-level index break one's promise be performed list number, by financial institution prosecution number,
It is related to emphasis case and is related to by number, affiliated company to tell situation；
Social effectiveness includes press exposure rate, affiliated company's total quantity, company recruitment public praise, news in three-level index
Combined influence power.
Two-level index is calculated by corresponding three-level index by weighted average mode, and first class index is by all second levels
Index is calculated by weighted average mode.
As shown in Figure 1,3, the method that a kind of pair of enterprise carries out risk quantification assessment, carries out risk based on six rib star models
Quantitative evaluation, six rib star models include first class index, two-level index, three-level index, and two-level index is by management state, industry shape
Condition, innovation ability, team is related, closes regulation degree, social effectiveness composition；Three-level index is that refinement corresponding to two-level index refers to
Mark；First class index is calculated by all two-level index by weighted average mode, method the following steps are included:
A. structural data is converted by data analysis technique by three-level index, data cleansing is carried out to structural data
So that the data format of three-level index is unified into number format, retain 2 significant digits number, the specific steps are as follows:
1) web crawlers engineer obtains all initial data of enterprise using crawler technology, by using to initial data
Natural language processing means parse the specific fields in the initial data, entity extracts, storage, to what is parsed
Structural data carries out Elastic Search high-volume data loading, and ElasticSearch is the base developed with Java
In the search server of Lucene, is issued as the open source code under Apache license terms, be that enterprise-level is searched currently popular
Index is held up, it provides the full-text search engine of a distributed multi-user ability, is based on RESTful web interface, and design is used
In cloud computing, real-time search can be reached, stablized, it is reliably, quickly, easy to install and use；
2) the relevant field progress specification of the structural data of storage, statistics, grouping, sequence are obtained using SQL statement
The quantitative value of certain field；
B. each numeric formatted three-level index is normalized as the quantized value between 0 to 1, that is, passes through three
A Function Mapping relationship is established between the quantity of grade index and the quantized value of three-level index, to calculate all three-level indexs
Quantized value；Each three-level index has independent function, and defining three-level index is positive message or negative news；Positive message
Three-level index is brought into Sigmoid function, and the three-level index of negative news is brought into anti-Sigmoid function.
In the present embodiment, management state includes that operation life in three-level index, registered capital, acceptance of the bid number, acceptance of the bid are total
The amount of money；
Industry situation include small micro- business event unicity in three-level index, large and medium-sized enterprise design multiple professionals,
Advantageous policy quantity, in the recent period unfavorable policy quantity, affiliated enterprise's same industry ratio in the recent period；
Innovation ability includes trade mark quantity in three-level index, every year on average trade mark quantity, patent numbers, every year on average specially
Sharp quantity, the last publication patent time, the last publication trade mark time；
Team's correlation includes the news quantity for being related to arrears of wages in three-level index, is related to the news number that legal person runs away
It measures, the city numbers that the quantity, the business that are related to company management risk news is related to, team development speed, team are averaged wages；
Close regulation degree include being listed in three-level index break one's promise be performed list number, by financial institution prosecution number,
It is related to emphasis case and is related to by number, affiliated company to tell situation；
Social effectiveness includes press exposure rate, affiliated company's total quantity, company recruitment public praise, news in three-level index
Combined influence power.
Wherein, the three-level index of positive message has operation life, registered capital, acceptance of the bid number, acceptance of the bid total amount, small micro- enterprise
Industry business unicity, large and medium-sized enterprise design multiple professionals, in the recent period advantageous policy quantity, affiliated enterprise's same industry ratio, quotient
Mark quantity, every year on average trade mark quantity, patent numbers, every year on average patent numbers, the last publication patent time, nearest one
City numbers that secondary publication trade mark time, business are related to, team development speed, team be averaged wages, press exposure rate, association public affairs
Take charge of total quantity, company recruits public praise, news combined influence power；The three-level index of negative news has recent unfavorable policy quantity, relates to
And arrears of wages news quantity, be related to news quantity that legal person runs away, be related to company management risk news quantity, be listed in
Break one's promise be performed list number, by financial institution's prosecution number, be related to emphasis case and related to by number, affiliated company to tell situation；
C., the quantized value of three-level index is calculated to the numerical value of two-level index: E by weighted average modei=100* ∑
(Aij*Wij)；Wherein i=1,2...6, j=1,2...n；EiFor the two-level index score, AijFor any secondary index
The quantized value of corresponding three-level index, WijFor the corresponding weight coefficient of corresponding j-th of three-level index of i-th of two-level index, n is
The quantity of three-level index；；
D. first class index is calculated by weighted average mode in the numerical value of all two-level index for being weighted and averaged and obtaining
Numerical value:
S=∑ (Ei*Vi)；Wherein i=1,2 ... 6, wherein S is first class index score, EiIt is obtained for the two-level index
Point, ViFor the corresponding weight coefficient of the two-level index；
Wherein, Wj、ViCoefficient artificially carries out an initialization apportioning cost when constructing six rib star models first, point
Be exactly with principle it is preliminary different to the disturbance degree of an enterprise according to second level, three-level rule, by test model, continuous iteration
Change weight or update, addition rule, achievees the effect that Optimized model；
In the present embodiment, first class index score=management state * 10%+ Industry situation * 10%+ innovation ability * 20%+ group
Team correlation * 10%+ closes regulation degree * 25%+ social effectiveness * 25%；；
The public affairs of " bad " that the company and N family for e. extracting " good " that the M family that bank or financial institution provide generally acknowledges generally acknowledge
The three-level index of department and by three-level quantification of targets be numerical value, as the input of the six ribs star model, by the way that the three-level is referred to
Marker tape enters the numerical value that first class index is calculated in the six ribs star model, and the numerical value of first class index here is exactly the synthesis of company
Score, by " good " of the scores binding test company of all test companies, " bad " two tag along sort, using subject's work
Make indicatrix (Receiver Operating Characteristic curve, ROC) to assess test company described six
The numerical value of first class index on rib star model is referred to by the level-one of Receiver operating curve's test method binding test company
It marks score and tests true two tag along sort of company, obtain best score threshold value, by judging Receiver Operating Characteristics' song
Line curve obtains best rate of precision；M, N is the positive integer not less than 100；
F. it when the rate of precision when test data on model is less than the first preset value, then needs according to two-level index and three-level
The significance level of index readjusts the weight of two-level index and three-level index, or adds or deletes the dependency rule word of enterprise
Section, more than the second preset value of model rate of precision, then completes the adjustment of six rib star models；
G. when the rate of precision when test data on six rib star models is greater than the first preset value, opening for six rib star models is completed
Hair, later period are adjusted according to the feedback of user market, the periodically weight of the two-level index to six rib star models and three-level index,
Adaptation of the six rib star models to more and more companies is continuously improved, enhances its generalized ability and accuracy；
H. the enterprise that client needs to carry out risk quantification assessment is substituted into the six rib star models that exploitation is completed, to export risk
Quantitative evaluation result.
In the present embodiment, the expression formula of rate of precision is as follows:
(I-J)/I；
Wherein, I is the sum for bringing " good " company of six rib star models into, and J is " good " public affairs obtained by six rib star models
Department and reality are also the quantity of " good " company.
If having obtained 95 companies by model has been company, and having 90 in this 95 is in 100 generally acknowledged good public affairs
It takes charge of in list, then rate of precision is 90%.The working principle of the invention: each enterprise is carried out using big data processing technique complete
The summation in orientation is assessed, and calculates each enterprise's final score.Specific implementation can use business risk quantitative estimation method:
The multiple dimension scores situations for firstly evaluating enterprise, then integrate each dimension and calculate model score.Each dimension and synthesis
Score is 0-100 points, and it is better that the higher expression enterprise of score shows in a certain dimension, otherwise performance is poorer.
Finally, it should be noted that those skilled in the art various changes and modifications can be made to the invention without departing from
The spirit and scope of the present invention.In this way, if these modifications and changes of the present invention belongs to the claims in the present invention and its waits system
Within the scope of counting, then the present invention is also intended to encompass these modification and variations.
Claims (9)
1. the six rib star models that a kind of pair of enterprise carries out risk quantification assessment, which is characterized in that refer to including first class index, second level
Mark, three-level index, the two-level index is by management state, Industry situation, innovation ability, team's correlation, conjunction regulation degree, social shadow
Ring power composition；The three-level index is refinement index corresponding to the two-level index；The first class index is by all second levels
Index is calculated by weighted average mode.
2. the six rib star models that a kind of pair of enterprise according to claim 1 carries out risk quantification assessment, which is characterized in that
The management state includes operation life in the three-level index, registered capital, acceptance of the bid number, acceptance of the bid total amount；
The Industry situation includes small micro- business event unicity, the multiple industries of large and medium-sized enterprise's design in the three-level index
Property, advantageous policy quantity, in the recent period unfavorable policy quantity, affiliated enterprise's same industry ratio in the recent period；
The innovation ability includes trade mark quantity in the three-level index, every year on average trade mark quantity, patent numbers, average every
Year patent numbers, the last publication patent time, the last publication trade mark time；
Team's correlation includes the news quantity for being related to arrears of wages in the three-level index, is related to the news that legal person runs away
City numbers that quantity, the quantity for being related to company management risk news, business are related to, team development speed, team are averaged wages；
The conjunction regulation degree include being listed in the three-level index break one's promise be performed list number, by financial institution prosecute time
Number is related to emphasis case and is related to by number, affiliated company to tell situation；
The social effectiveness include press exposure rate in the three-level index, affiliated company's total quantity, company's recruitment public praise,
News combined influence power.
3. the six rib star models that a kind of pair of enterprise according to claim 2 carries out risk quantification assessment, which is characterized in that institute
It states two-level index and is calculated by corresponding three-level index by weighted average mode.
4. the method that a kind of pair of enterprise carries out risk quantification assessment, which is characterized in that carry out risk amount based on six rib star models
Change assessment, the six ribs star model includes first class index, two-level index, three-level index, and the two-level index is by management state, row
Industry situation, innovation ability, team is related, closes regulation degree, social effectiveness composition；The three-level index is the two-level index institute
Corresponding refinement index；The first class index is calculated by all two-level index by weighted average mode, the method
The following steps are included:
A. structural data is converted by data analysis technique by the three-level index, data is carried out to the structural data
Cleaning makes the data format of the three-level index be unified into number format, the specific steps are as follows:
1) web crawlers engineer obtains all initial data of enterprise using crawler technology, by using nature to initial data
Language Processing means parse the specific fields in the initial data, entity extracts, storage, to the structure parsed
Change data and carries out Elastic Search high-volume data loading；
2) certain word is obtained to the relevant field progress specification of the structural data of storage, statistics, grouping, sequence using SQL statement
The quantitative value of section；
B. each numeric formatted three-level index is normalized as the quantized value between 0 to 1；
C., the quantized value of the three-level index is calculated to the numerical value of the two-level index: E by weighted average modei=
100*∑(Aij*Wij)；Wherein i=1,2...6, j=1,2...n；EiFor the two-level index score, AijIt is described any two
The quantized value of the corresponding three-level index of grade index, WijFor the corresponding weight system of corresponding j-th of three-level index of i-th of two-level index
Number, n are the quantity of three-level index；
D. the first class index is calculated by weighted average mode in the numerical value of all two-level index for being weighted and averaged and obtaining
Numerical value: S=∑ (Ei*Vi)；Wherein i=1,2 ... 6, wherein S is the first class index score, EiFor the two-level index
Score, ViFor the corresponding weight coefficient of the two-level index；
The company of " bad " that the company and N family for e. extracting " good " that the M family that bank or financial institution provide generally acknowledges generally acknowledge
Three-level index and be numerical value by three-level quantification of targets, as the input of the six ribs star model, by by the three-level index band
Enter the numerical value that first class index is calculated in the six ribs star model, by the scores binding test company of all test companies
" good ", " bad " two tag along sort assesses test company using Receiver operating curve on the six ribs star model
First class index numerical value, first class index score and survey by Receiver operating curve's test method binding test company
True two tag along sort of examination company, obtains best score threshold value, by judging that Receiver operating curve obtains most preferably
Rate of precision；M, N is the positive integer not less than 100；
F. it when the rate of precision when test data on model is less than the first preset value, then needs according to two-level index and three-level index
Significance level readjust the weight of two-level index and three-level index, or add or delete the dependency rule field of enterprise, directly
To more than the second preset value of model rate of precision, then the adjustment of the six ribs star model is completed；
G. when the rate of precision when test data on the six ribs star model is greater than the first preset value, opening for six rib star models is completed
Hair, the later period, periodically the weight of the two-level index to the six ribs star model and three-level index carried out according to the feedback of user market
Adjustment is continuously improved adaptation of the six ribs star model to more and more companies, enhances its generalized ability and accuracy；
H. the enterprise for carrying out risk quantification assessment is needed to substitute into the six rib star models that the exploitation is completed client, to export risk
Quantitative evaluation result.
5. the method that a kind of pair of enterprise according to claim 4 carries out risk quantification assessment, which is characterized in that the step
Specific steps in b are as follows: by establishing a function between the quantity of the three-level index and the quantized value of the three-level index
Mapping relations, to calculate the quantized value of all three-level indexs；Each three-level index has an independent function, described in definition
Three-level index is positive message or negative news；The three-level index of positive message is brought into Sigmoid function, and the three of negative news
Grade index is brought into anti-Sigmoid function.
6. the method that a kind of pair of enterprise according to claim 4 carries out risk quantification assessment, which is characterized in that described accurate
The expression formula of rate is as follows: (I-J)/I；Wherein, I is the sum for bringing " good " company of six rib star models into, and J is to pass through six rib star moulds
" good " company that type obtains and the quantity that reality is also " good " company.
7. the method that a kind of pair of enterprise according to claim 4 carries out risk quantification assessment, which is characterized in that the operation
Situation includes operation life in the three-level index, registered capital, acceptance of the bid number, acceptance of the bid total amount；
The Industry situation includes small micro- business event unicity, the multiple industries of large and medium-sized enterprise's design in the three-level index
Property, advantageous policy quantity, in the recent period unfavorable policy quantity, affiliated enterprise's same industry ratio in the recent period；
The innovation ability includes trade mark quantity in the three-level index, every year on average trade mark quantity, patent numbers, average every
Year patent numbers, the last publication patent time, the last publication trade mark time；
Team's correlation includes the news quantity for being related to arrears of wages in the three-level index, is related to the news that legal person runs away
City numbers that quantity, the quantity for being related to company management risk news, business are related to, team development speed, team are averaged wages；
The conjunction regulation degree include being listed in the three-level index break one's promise be performed list number, by financial institution prosecute time
Number is related to emphasis case and is related to by number, affiliated company to tell situation；
The social effectiveness include press exposure rate in the three-level index, affiliated company's total quantity, company's recruitment public praise,
News combined influence power.
8. the method that a kind of pair of enterprise according to claim 5 carries out risk quantification assessment, which is characterized in that the front
The three-level index of message has operation life, registered capital, acceptance of the bid number, acceptance of the bid total amount, small micro- business event unicity, big-and-middle
Type enterprise designs multiple professionals, in the recent period advantageous policy quantity, affiliated enterprise's same industry ratio, trade mark quantity, every year on average quotient
Mark quantity, patent numbers, every year on average patent numbers, the last publication patent time, the last publication trade mark time, industry
Be averaged wages, press exposure rate, affiliated company's total quantity, the company of the city numbers being related to, team development speed, team that be engaged in recruits
Public praise, news combined influence power；The three-level index of the negative news has recent unfavorable policy quantity, is related to the new of arrears of wages
Quantity, the quantity for being related to news quantity that legal person runs away, being related to company management risk news are heard, is listed in break one's promise and is performed list
Number, by financial institution's prosecution number, be related to emphasis case and related to by number, affiliated company to tell situation.
9. the method that a kind of pair of enterprise according to claim 4 carries out risk quantification assessment, which is characterized in that described first
Preset value is 80%, and second preset value is 90%.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811097238.5A CN109359818A (en) | 2018-09-20 | 2018-09-20 | A kind of pair of enterprise carries out the method and six rib star models of risk quantification assessment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811097238.5A CN109359818A (en) | 2018-09-20 | 2018-09-20 | A kind of pair of enterprise carries out the method and six rib star models of risk quantification assessment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109359818A true CN109359818A (en) | 2019-02-19 |
Family
ID=65351385
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811097238.5A Pending CN109359818A (en) | 2018-09-20 | 2018-09-20 | A kind of pair of enterprise carries out the method and six rib star models of risk quantification assessment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109359818A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110008345A (en) * | 2019-04-17 | 2019-07-12 | 重庆天蓬网络有限公司 | Platform service firm industry data aggregate analysis method, device, medium and equipment |
| CN111754131A (en) * | 2020-06-30 | 2020-10-09 | 苏州朗动网络科技有限公司 | Enterprise information dynamic monitoring method, equipment and medium |
| CN111798102A (en) * | 2020-06-11 | 2020-10-20 | 中国建设银行股份有限公司 | Data supervision method, system, computer equipment and readable storage medium |
| CN111915155A (en) * | 2020-07-13 | 2020-11-10 | 上海冰鉴信息科技有限公司 | Small and micro enterprise risk level identification method and device and computer equipment |
| CN112070603A (en) * | 2020-09-11 | 2020-12-11 | 重庆誉存大数据科技有限公司 | Grading card model, configuration system thereof and grading processing method |
| CN112418601A (en) * | 2020-10-15 | 2021-02-26 | 重庆市科学技术研究院 | Policy matching method and system based on index set |
| CN112418600A (en) * | 2020-10-15 | 2021-02-26 | 重庆市科学技术研究院 | Enterprise policy scoring method and system based on index set |
| CN112561320A (en) * | 2020-12-14 | 2021-03-26 | 中国建设银行股份有限公司 | Training method of mechanism risk prediction model, mechanism risk prediction method and device |
| CN112613717A (en) * | 2020-12-17 | 2021-04-06 | 安徽兆尹信息科技股份有限公司 | Financial data processing method and storage medium |
| CN113505990A (en) * | 2021-07-08 | 2021-10-15 | 建信金融科技有限责任公司 | Enterprise risk assessment method and device, electronic equipment and storage medium |
| CN116777222A (en) * | 2023-07-06 | 2023-09-19 | 北京中知智慧科技有限公司 | Enterprise risk prevention dimension review system |
-
2018
- 2018-09-20 CN CN201811097238.5A patent/CN109359818A/en active Pending
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110008345A (en) * | 2019-04-17 | 2019-07-12 | 重庆天蓬网络有限公司 | Platform service firm industry data aggregate analysis method, device, medium and equipment |
| CN111798102A (en) * | 2020-06-11 | 2020-10-20 | 中国建设银行股份有限公司 | Data supervision method, system, computer equipment and readable storage medium |
| CN111754131A (en) * | 2020-06-30 | 2020-10-09 | 苏州朗动网络科技有限公司 | Enterprise information dynamic monitoring method, equipment and medium |
| CN111915155A (en) * | 2020-07-13 | 2020-11-10 | 上海冰鉴信息科技有限公司 | Small and micro enterprise risk level identification method and device and computer equipment |
| CN112070603A (en) * | 2020-09-11 | 2020-12-11 | 重庆誉存大数据科技有限公司 | Grading card model, configuration system thereof and grading processing method |
| CN112418601A (en) * | 2020-10-15 | 2021-02-26 | 重庆市科学技术研究院 | Policy matching method and system based on index set |
| CN112418600A (en) * | 2020-10-15 | 2021-02-26 | 重庆市科学技术研究院 | Enterprise policy scoring method and system based on index set |
| CN112561320A (en) * | 2020-12-14 | 2021-03-26 | 中国建设银行股份有限公司 | Training method of mechanism risk prediction model, mechanism risk prediction method and device |
| CN112613717A (en) * | 2020-12-17 | 2021-04-06 | 安徽兆尹信息科技股份有限公司 | Financial data processing method and storage medium |
| CN113505990A (en) * | 2021-07-08 | 2021-10-15 | 建信金融科技有限责任公司 | Enterprise risk assessment method and device, electronic equipment and storage medium |
| CN116777222A (en) * | 2023-07-06 | 2023-09-19 | 北京中知智慧科技有限公司 | Enterprise risk prevention dimension review system |
| CN116777222B (en) * | 2023-07-06 | 2024-04-02 | 北京中知智慧科技有限公司 | Enterprise risk prevention dimension review system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109359818A (en) | A kind of pair of enterprise carries out the method and six rib star models of risk quantification assessment | |
| TWI712981B (en) | Risk identification model training method, device and server | |
| US20240078475A1 (en) | Attributing reasons to predictive model scores with local mutual information | |
| Kumar et al. | Software development cost estimation using wavelet neural networks | |
| Couillard et al. | A comment on measuring the Hurst exponent of financial time series | |
| Fang | Intelligent online English teaching system based on SVM algorithm and complex network | |
| JP2004528657A5 (en) | ||
| CN107798604A (en) | Become a shareholder when selecting method and terminal device based on machine learning | |
| Gao | The use of machine learning combined with data mining technology in financial risk prevention | |
| Song et al. | An XGBoost algorithm for predicting purchasing behaviour on E-commerce platforms | |
| CN108572988A (en) | A kind of house property assessment data creation method and device | |
| Liao et al. | GRBMC: An effective crowdsourcing recommendation for workers groups | |
| Holló et al. | Assessing household credit risk: evidence from a household survey | |
| Liu et al. | Information gain with weight based decision tree for the employment forecasting of undergraduates | |
| Ting et al. | Design and implementation of intelligent accounting data analysis platform based on industrial cloud computing | |
| Rahman et al. | A Classification Based Model to Assess Customer Behavior in Banking Sector. | |
| Cai et al. | Towards correlated data trading for high-dimensional private data | |
| CN110070264A (en) | Contract evaluation method, platform, equipment and readable storage medium storing program for executing | |
| Xia et al. | Analysis and prediction of telecom customer churn based on machine learning | |
| CN109523185A (en) | The method, apparatus and storage medium of collection scorecard are generated based on artificial intelligence | |
| Qiang et al. | Relationship model between human resource management activities and performance based on LMBP algorithm | |
| TWM623354U (en) | investment recommendation system | |
| Patidar et al. | Quantile Regression Comprehensive in Machine Learning: A Review | |
| Lu et al. | Research on time series data prediction based on clustering algorithm-A case study of Yuebao | |
| CN116308158B (en) | National asset supervision and management system |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20191113 Address after: 400042 No.51 dapingzheng street, Yuzhong District, Chongqing Applicant after: CHONGIQNG TELECOM SYSTEM INTEGRATION CO., LTD. Applicant after: Chongqing Yu Yu Da Data Technology Co., Ltd. Address before: The middle section of Mount Huangshan road 401121 Chongqing city Yubei District No. 53 of No. 2 C block 9 layer kylin Applicant before: Chongqing Yu Yu Da Data Technology Co., Ltd. |
|
| TA01 | Transfer of patent application right | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190219 |
|
| WD01 | Invention patent application deemed withdrawn after publication |