The content of the invention
In view of the drawbacks described above of prior art, embodiment of the present invention is there is provided a kind of real-time approval system of loan and sideMethod, can effectively solve the problem that the problem that current loan examining risk is high and loan examination ＆ approval efficiency is low.
Specifically, embodiment of the present invention provides a kind of real-time approval system of loan, and it includes：
The pre- credit module of amount, for after client submits application materials to, being entered for client according to the contribution degree information of clientThe pre- credit of row amount is processed, and obtains the comprehensive credit line information of the client；
Whether internal control list checks module, including by the identity three elements information of matching client, checking the clientIn control list, if starting monitoring and early warning；If it was not then performing next module；
Reference information checks module, for extracting reference information in people's row reference information and row, is rejected loans according to self-definingRule, judges whether to reject loans；
The regular library module of examination ＆ approval, for the Rule Information in initial rules storehouse and the volume of predefined training dataCode, obtains initial neutral net, and the examination ＆ approval rule after being updated is extracted from the initial neutral net, further according to describedWhether the regular and described client's application materials of examination ＆ approval after renewal, automatic decision rejects loans；
Scoring card module, for being scored according to client's application materials, people's row reference information, credit card, according toPredefined calculating classifying rules, obtains client's scoring and examination ＆ approval advisory result of providing a loan.
Correspondingly, embodiment of the present invention additionally provides a kind of real-time measures and procedures for the examination and approval of loan, and it includes：
After client submits application materials to, the pre- credit of amount is carried out for client according to the contribution degree information of client and is processed, obtainedTo the comprehensive credit line information of the client；
By the identity three elements information for matching client, the client is checked whether in internal control list, if startedMonitoring and early warning；If it was not then performing next step；
Reference information in people's row reference information and row is extracted, according to self-defining rule of rejecting loans, judges whether to reject loans；
The coding of Rule Information and predefined training data in initial rules storehouse, obtains initial neutral net,And the examination ＆ approval rule after being updated is extracted from the initial neutral net, further according to the examination ＆ approval rule after the renewal and instituteClient's application materials are stated, whether automatic decision rejects loans；
According to client's application materials, people's row reference information, credit card scoring, classify according to predefined calculatingRule, obtains client's scoring and examination ＆ approval advisory result of providing a loan.
By adopting embodiment of the present invention, can effectively solve the problem that current loan examining risk is high and loan examination ＆ approval efficiency is lowProblem, so as to reduce loan examining risk and improve loan examination ＆ approval efficiency effect.
Embodiment 1：
The system includes：
The pre- credit module 100 of amount, for client submit to application materials after, according to the contribution degree information of client be clientCarry out the pre- credit of amount to process, obtain the comprehensive credit line information of the client；Wherein, the pre- credit module of amount is according to clientContribution degree be that assets information, storage housing loan, payroll credit and common reserve fund pay information and carry out the pre- credit of amount for client.
Internal control list checks module 200, for by the identity three elements information of matching client, checking that whether the client existsIn internal control list, if starting monitoring and early warning；If it was not then performing next module；Wherein, to customer credit line pre-grantedAfter letter, client can apply for the amount of the loan less than the pre- accrediting amount.After client's proposition loan application, visitor is first checked forWhether family is in internal control list.For the client of the artificial presence potential risk for finding under risk warning model or line, need to includeInternal control list simultaneously carries out corresponding operating.Internal control list data form is as follows：
Retail loan examines in real time component when client proposes loan application with payment on account of credit, and all needing will by client identity threeElement is matched with internal control list, such as finds client in internal control list, and is judged according to client's access suggestion in internal control list,Directly show the page of rejecting loans, payment reports an error or automatic flow is to next step.For the manual internal control list for importing, importing personnel shouldThe reason for " importing reason " hurdle should write client exactly and list internal control list in, in case the consulting of the client that rejects loans to this part is carried out in the futureReply.For internal control list be judged as terminate, then client application and pay retail loan when, page prompts report an error；ForAmount is adjusted and reduced in system judgement, then client passes through the volume that rule is calculated when amount is exported in application retail loan by systemThe superior coefficient in degree basis；In loan payment, system is exported to be judged to prop up client by system when can draw amount in originalWith taking advantage of a coefficient on the basis of amount.The assignment of the coefficient is arranged as parameter unification.
Reference information checks module 300, for extracting reference information in people's row reference information and row, is refused according to self-definingRule is borrowed, judges whether to reject loans；Wherein, need to agree to that loan approving authority extracts its people's row reference letter when client submits loan application toBreath.For the client not rejected loans by internal control list, its reference information is checked, if the overdue record of reference inquiry times, personal loan,Credit card record overdue with quasi- credit card meets any one rule of rejecting loans, then directly reject loans.All of standard in rule of rejecting loansParameter is each configured to, Mobile state adjustment can be entered according to indexs such as the fraction defectives of loan.
The regular library module 400 of examination ＆ approval, for the Rule Information in initial rules storehouse and predefined training dataCoding, obtains initial neutral net, and the examination ＆ approval rule after being updated is extracted from the initial neutral net, further according to instituteThe regular and described client's application materials of examination ＆ approval after updating are stated, whether automatic decision rejects loans；Wherein, the loan application in client leads toAfter crossing the inspection of internal control list and people's row reference, decide whether to make a loan using existing examination ＆ approval rule.In general,Credit agency can set up initial examination ＆ approval rule base.Initial rules storehouse be usually present knowledge it is incomplete or inconsistent the problems such as, it is difficultTo adapt to the change of actual credit environment.In view of larger to the difficulty that it carries out manual modification according to actual credit environment, andIn conventional loan permit business, credit agency have accumulated the mass data in terms of retail loan.It is in the best mannerKnowledge refining is carried out to original rule base using these data, can preferably be solved the above problems.
Scoring card module 500, for according to client's application materials, people's row reference information, credit card scoring, pressingAccording to predefined calculating classifying rules, client's scoring is obtained and examination ＆ approval advisory result of providing a loan.Wherein, the card module that scores utilizes systemRelevant information in incoming client's application materials, people's row reference information, credit card behavior scoring and customer information system, according toThe exclusion policy of setting, calculate standard, hard policy, special screening policy and screening policy, (approval of being classified to client automaticallyApplication client, refuse an application client and manually examine client), while export each participate in scoring client score value and system recommendationsAs a result.In addition, when the card module that scores scoring and advisory result to agree to signing housing loan when, may be selected it is autonomous pay, agreementPay and the form such as consumption and payment.
In the present embodiment, during client's initiation loan application reference information is checked with internal control list first, soIt is afterwards that assets information, storage housing loan, payroll credit and common reserve fund pay information and carry out amount for client according to the contribution degree of clientPre- credit, examines on this basis according to examination ＆ approval rule base to the loan application, and the appraisal result of final scorecard is determinedFixed loan application of whether letting pass, while specific score value determines that the loan pays in which way client.In addition, rightRetail loan examination ＆ approval rule base utilizes nerual network technique, and using the historical data of retail loan examination ＆ approval refinement is carried out；Using loanDetect that dangerous client is included internal control name menu manager by early warning mechanism afterwards.By adopting embodiment of the present invention, can effectively solve the problem thatThe problem that current loan examining risk is high and loan examination ＆ approval efficiency is low, examines so as to reducing loan examining risk and improving loanCriticize the effect of efficiency.
Embodiment 5：
In another embodiment of the invention, the system in addition to above-mentioned processing mode, the predefined meterPoint counting rule-like includes：Self-defined some variate-values, and client's scoring is calculated according to predefined computing formula, and pressThe client is classified according to predefined criteria for classification.It is specific as follows：
1st, Rating Model desired data is applied for
Scoring needs province information in customer data, account interest, whether native, family according to importance and discriminationThe data message of front yard total liability, occupation, sex, highest educational background, marital status, age totally nine aspects.
2nd, structure's variable
On this basis, to six of which data message：Province information, account interest, family's total liability, occupation, highestEducational background, age carry out the construction of variable.Make is as follows：
(1) province information：The passing business credit standing in area according to belonging to client, economic development situation considersThe many factors such as the economic development situation of each province and managerial skills, by province three classes are divided.
(2) account interest：The Demand Deposit Accounts interest sum of nearest 1 year marks off 5 when applying providing a loan according to clientClass.
| Sequence number | Interest amount |
| 1 | [0,5] |
| 2 | [5,20) |
| 3 | [20,50) |
| 4 | [50,100) |
| 5 | [100,+∞) |
| 6 | No effects account |
(3) family's total liability：Family's total liability amount of money according to client carries out following classification.
(4) occupation：According to the professional nature of client, five classes are classified as.
(5) highest educational background：Received an education situation according to client, highest educational background is divided into three classes.
| Sequence number | Highest educational background |
| 1 | It is more than postgraduate |
| 2 | Undergraduate course |
| 3 | Junior college's special secondary school |
| 4 | Other |
(6) age：The age of client is divided into into four classes.
| Sequence number | Age |
| 1 | 18-30 |
| 2 | 30-35 |
| 3 | 35-50 |
| 4 | More than 50 |
3rd, variable replacement
By structure's variable, sex, marital status are added, obtain 12 variables required for scoring：Need according to model,According to the different values of each variable, take its correspondence WOE values (evidence weight, Weight of Evidence) and be replaced, haveBody replacement values are as follows：
Variable one：Province information
Variable two：Account interest
| Property Name | Code | WOE |
| [0,5) | [0,5) | A |
| [5,20) | [5,20) | B |
| [20,50) | [20,50) | C |
| [50,100) | [50,100) | D |
| [100,+∞) | [100,+∞) | E |
| No effects account | No effects account | F |
Variable three：Whether native
| Property Name | Code | WOE |
| It is no | 0 | A |
| It is | 1 | B |
Variable four：Family's total liability
| Property Name | Code | WOE |
| Nothing and unknown | Nothing and unknown | A |
| Less than 70000) | Less than 70000 | B |
| [7-10 ten thousand) | 7-10 ten thousand | C |
| [10-16 ten thousand) | 10-16 ten thousand | D |
| [more than 160,000 | More than 160000 | E |
Variable five：Occupation
Variable six：Sex
Variable seven：Highest educational background
| Property Name | Code | WOE |
| It is more than postgraduate | 10 | A |
| Undergraduate course | 20 | B |
| Junior college's special secondary school | 30,40 | C |
| Other | Other | D |
Variable eight：Marital status
| Property Name | Code | WOE |
| It is unmarried | 1 | A |
| It is married to have children | 5 | B |
| It is married to have no children | 6 | C |
| Other | 3,4,9 | D |
Variable nine：Age
| Property Name | Code | WOE |
| 18-30 | [18,30] | A |
| 30-35 | (30,35] | B |
| 35-50 | (35,50] | C |
| More than 50 | >50 | D |
According to the variate-value obtained after above variable replacement, the scorecard score value of client can be calculated.The most final review of clientCard score value is divided to be scoreadjust.System calculates the application score value of each application client according to the computing formula of setting, thenScore value is collected according to certain standard, 20 fraction levels can be divided into altogether.
After completing score value calculating, to applying for that score value is sorted out, system is according to access score value set in advance scoringDelimit as five score value intervals, i.e.,：High score area client, middle score value area client, low score value area client, artificial judgment area client andRefusal score value area client.Score value judges that form is as follows：
It is located at that different score values are interval according to appraisal result, the loan application of client is done respectively and directly passed through, manually examinedCriticize the process with directly refusal.For the loan application for directly passing through proceeds to module of making loans of opening an account, for the loan of artificial examination ＆ approvalApplication proceeded to and examined under line, and remaining loan application is directly rejected loans.Simultaneously different score values intervals are located at according to appraisal result, it is determined thatThe means of payment of customer lending.
Fig. 4 is a kind of schematic flow sheet of the real-time measures and procedures for the examination and approval of loan according to embodiment of the present invention.With reference to Fig. 4, instituteThe system of stating includes：
Step S1, after client submits application materials to, the pre- credit of amount is carried out according to the contribution degree information of client for clientProcess, obtain the comprehensive credit line information of the client；Wherein, in the step according to client contribution degree be assets information,Storage housing loan, payroll credit and common reserve fund pay information carries out the pre- credit of amount for client.
Step S2, by the identity three elements information for matching client, checks the client whether in internal control list, ifThen starting monitoring and early warning；If it was not then performing next step；Wherein, after to the pre- credit of customer credit line, client can be withThe amount of the loan of the application less than the pre- accrediting amount.After client's proposition loan application, whether client is first checked in internal control nameDan Zhong.For the client of the artificial presence potential risk for finding under risk warning model or line, internal control list need to be included and carried outCorresponding operating.Internal control list data form is as follows：
Retail loan examines in real time component when client proposes loan application with payment on account of credit, and all needing will by client identity threeElement is matched with internal control list, such as finds client in internal control list, and is judged according to client's access suggestion in internal control list,Directly show the page of rejecting loans, payment reports an error or automatic flow is to next step.For the manual internal control list for importing, importing personnel shouldThe reason for " importing reason " hurdle should write client exactly and list internal control list in, in case the consulting of the client that rejects loans to this part is carried out in the futureReply.For internal control list be judged as terminate, then client application and pay retail loan when, page prompts report an error；ForAmount is adjusted and reduced in system judgement, then client passes through the volume that rule is calculated when amount is exported in application retail loan by systemThe superior coefficient in degree basis；In loan payment, system is exported to be judged to prop up client by system when can draw amount in originalWith taking advantage of a coefficient on the basis of amount.The assignment of the coefficient is arranged as parameter unification.
Step S3, extracts reference information in people's row reference information and row, according to self-defining rule of rejecting loans, judges whether to refuseBorrow；Wherein, need to agree to that loan approving authority extracts its people's row reference information when client submits loan application to.For not by internal control nameThe client for singly rejecting loans, checks its reference information, if the overdue record of reference inquiry times, personal loan, credit card and quasi- credit cardOverdue record meets any one rule of rejecting loans, then directly reject loans.All of standard is each configured to parameter in rule of rejecting loans, can be withThe indexs such as the fraction defective according to loan enter Mobile state adjustment.
Step S4, the coding of Rule Information and predefined training data in initial rules storehouse, obtains initial godJing networks, and the examination ＆ approval rule after being updated is extracted from the initial neutral net, further according to the examination ＆ approval after the renewalWhether regular and described client's application materials, automatic decision rejects loans；Wherein, the loan application in client passes through internal control list and peopleAfter the inspection of row reference, decide whether to make a loan using existing examination ＆ approval rule.In general, credit agency can set upInitial examination ＆ approval rule base.Initial rules storehouse be usually present knowledge it is incomplete or inconsistent the problems such as, it is difficult to adapt to actual creditThe change of environment.In view of larger to the difficulty that it carries out manual modification according to actual credit environment, and examine in conventional loanIn batch traffic, credit agency have accumulated the mass data in terms of retail loan.It is in the best manner using these data pairOriginal rule base carries out knowledge refining, can preferably solve the above problems.
Step S5, according to client's application materials, people's row reference information, credit card scoring, according to predefinedClassifying rules is calculated, client's scoring is obtained and examination ＆ approval advisory result of providing a loan.Wherein, using the client Shen that system is incoming in the stepRelevant information that please be in data, people's row reference information, credit card behavior scoring and customer information system, according to the exclusion political affairs of settingPlan, calculating standard, hard policy, special screening policy and screening policy, client is classified automatically (approval is applied for client, is refusedApplication client and manually examination ＆ approval client absolutely), while exporting each score value and system recommendations result for participating in scoring client.
In the present embodiment, during client's initiation loan application reference information is checked with internal control list first, soIt is afterwards that assets information, storage housing loan, payroll credit and common reserve fund pay information and carry out amount for client according to the contribution degree of clientPre- credit, examines on this basis according to examination ＆ approval rule base to the loan application, and the appraisal result of final scorecard is determinedFixed loan application of whether letting pass, while specific score value determines that the loan pays in which way client.In addition, rightRetail loan examination ＆ approval rule base utilizes nerual network technique, and using the historical data of retail loan examination ＆ approval refinement is carried out；Using loanDetect that dangerous client is included internal control name menu manager by early warning mechanism afterwards.By adopting embodiment of the present invention, can effectively solve the problem thatThe problem that current loan examining risk is high and loan examination ＆ approval efficiency is low, examines so as to reducing loan examining risk and improving loanCriticize the effect of efficiency.
In another embodiment of the present invention, the system is in addition to above-mentioned processing mode, wherein, the step S1 bagInclude：
Step S11, according to the AUM values in the contribution degree information, obtains credit line A；Wherein, with the nearly N number of moon, (N isPositive integer more than or equal to 1) the moon average daily AUM values be radix, consider the stability and Long-term change trend feelings of client's AUM valuesCondition, provides client's AUM value credit lines A.
Step S12, according to storage housing loan collateral value and adjustment factor, obtains credit line B；Wherein, with stock buildingsLoan collateral value is radix, considers mortgage rate coefficient, house property value-added coefficient, city adjustment factor and stock buildings credit balance volumeDeng providing client's storage housing loan credit line B.
Step S13, the client's annual income and survival phase in payroll credit data, obtains credit line C；Wherein, rootAccording to the client's annual income in payroll credit data, it is considered to the survival phase of payroll credit and stability, client's payroll credit letter is givenUse amount C.
Step S14, according to common reserve fund paying volume and pay coefficient in data is paid, and calculates expectations of customer annual income,And time and account balance are paid according to common reserve fund, obtain credit line D；Wherein, paying in data is paid according to common reserve fundVolume with pay coefficient, calculate the expected annual income of client, consider time span and account balance that common reserve fund is paid, giveGo out client's common reserve fund credit line D.
In another embodiment of the invention, methods described in addition to above-mentioned processing mode, wherein, step S1 is also wrappedInclude：
Amount calculating is carried out to credit line A, credit line B, credit line C and credit line D, is obtained and is exportedComprehensive credit line；
Wherein, the computation rule of the amount calculating is：Comprehensive credit line=Max (credit line A, credit line B,Credit line C, credit line D) * adjustment factors.
Adjustment factor is, when there is multiple indexs, suitably to increase amount.When the credit line of 4 dimensions of presence, thenAdjustment factor is 1.3；When such as there is the credit line of 3 dimensions, then adjustment factor is 1.2；There is the credit line of 2 dimensionsWhen, then adjustment factor is 1.1；The credit line of 1 dimension is only deposited, then adjustment factor is 1.0.
In a further embodiment of the present invention, the system in addition to above-mentioned processing mode, wherein, it is described from it is described justThe examination ＆ approval rule after being updated is extracted in beginning neutral net to be included：Using pruning algorithms and SubsetII algorithms, to described firstBeginning neutral net carries out simplifying process, obtains the rule of the examination ＆ approval after the renewal.
Initially set up the initial rules storehouse of retail loan examination ＆ approval：
Knowledge refining is carried out to it using retail loan history examination ＆ approval data.History examination ＆ approval data only have two kinds of conclusions：TogetherAnticipate and provide a loan and disagree loan.Simultaneously refinement is carried out to it using retail loan history overdue data.Examining overdue loanBatch conclusion is set to and disagrees loan, and the examination ＆ approval conclusion of the normal loan refunded is set to agreement and provides a loan.To Shen in history examination ＆ approval dataPlease person's information and loan application conclusion carry out encoding respectively it is as follows：
The coding situation of Rule Information and training data in initial rules storehouse, obtains an initial neutral net(as shown in Figure 3).
The structure of neutral net is adjusted using history loan examination ＆ approval data, when neutral net can be by all trainingData correctly after classification, that is, reach expected training effect.Then beta pruning is carried out to network structure using pruning algorithms, is removedUnessential connection or node, to simplify the structure of neutral net.Finally adopt nerve net of the SubsetII algorithms from after pruningNew examination ＆ approval rule is extracted in network.Refinement can be with rule of simplification storehouse, the complexity of reduction rule, while rule can also be improvedThe reasoning accuracy rate in storehouse.Can in time be adjusted with refund data according to recent personal loan examination ＆ approval using new examination ＆ approval ruleIt is whole, effectively raise the accuracy of retail loan examination ＆ approval.
In another embodiment of the invention, the system in addition to above-mentioned processing mode, the predefined meterPoint counting rule-like includes：Self-defined some variate-values, and client's scoring is calculated according to predefined computing formula, and pressThe client is classified according to predefined criteria for classification.It is specific as follows：
1st, Rating Model desired data is applied for
Scoring needs province information in customer data, account interest, whether native, family according to importance and discriminationThe data message of front yard total liability, occupation, sex, highest educational background, marital status, age totally nine aspects.
2nd, structure's variable
On this basis, to six of which data message：Province information, account interest, family's total liability, occupation, highestEducational background, age carry out the construction of variable.Make is as follows：
(1) province information：The passing business credit standing in area according to belonging to client, economic development situation considersThe many factors such as the economic development situation of each province and managerial skills, by province three classes are divided.
(2) account interest：The Demand Deposit Accounts interest sum of nearest 1 year marks off 5 when applying providing a loan according to clientClass.
| Sequence number | Interest amount |
| 1 | [0,5] |
| 2 | [5,20) |
| 3 | [20,50) |
| 4 | [50,100) |
| 5 | [100,+∞) |
| 6 | No effects account |
(3) family's total liability：Family's total liability amount of money according to client carries out following classification.
(4) occupation：According to the professional nature of client, five classes are classified as.
(5) highest educational background：Received an education situation according to client, highest educational background is divided into three classes.
| Sequence number | Highest educational background |
| 1 | It is more than postgraduate |
| 2 | Undergraduate course |
| 3 | Junior college's special secondary school |
| 4 | Other |
(6) age：The age of client is divided into into four classes.
| Sequence number | Age |
| 1 | 18-30 |
| 2 | 30-35 |
| 3 | 35-50 |
| 4 | More than 50 |
3rd, variable replacement
By structure's variable, sex, marital status are added, obtain 12 variables required for scoring：Need according to model,According to the different values of each variable, take its correspondence woe value and be replaced, concrete replacement values are as follows：
Variable one：Province information
Variable two：Account interest
| Property Name | Code | WOE |
| [0,5) | [0,5) | A |
| [5,20) | [5,20) | B |
| [20,50) | [20,50) | C |
| [50,100) | [50,100) | D |
| [100,+∞) | [100,+∞) | E |
| No effects account | No effects account | F |
Variable three：Whether native
| Property Name | Code | WOE |
| It is no | 0 | A |
| It is | 1 | B |
Variable four：Family's total liability
| Property Name | Code | WOE |
| Nothing and unknown | Nothing and unknown | A |
| Less than 70000) | Less than 70000 | B |
| [7-10 ten thousand) | 7-10 ten thousand | C |
| [10-16 ten thousand) | 10-16 ten thousand | D |
| [more than 160,000 | More than 160000 | E |
Variable five：Occupation
Variable six：Sex
| Property Name | Code | WOE |
| Female | 2 | A |
| Man | 1 | B |
Variable seven：Highest educational background
| Property Name | Code | WOE |
| It is more than postgraduate | 10 | A |
| Undergraduate course | 20 | B |
| Junior college's special secondary school | 30,40 | C |
| Other | Other | D |
Variable eight：Marital status
| Property Name | Code | WOE |
| It is unmarried | 1 | A |
| It is married to have children | 5 | B |
| It is married to have no children | 6 | C |
| Other | 3,4,9 | D |
Variable nine：Age
According to the variate-value obtained after above variable replacement, the scorecard score value of client can be calculated.The most final review of clientCard score value is divided to be scoreadjust.System calculates the application score value of each application client according to the computing formula of setting, thenScore value is collected according to certain standard, 20 fraction levels can be divided into altogether.
After completing score value calculating, to applying for that score value is sorted out, system is according to access score value set in advance scoringDelimit as five score value intervals, i.e.,：High score area client, middle score value area client, low score value area client, artificial judgment area client andRefusal score value area client.Score value judges that form is as follows：
It is located at that different score values are interval according to appraisal result, the loan application of client is done respectively and directly passed through, manually examinedCriticize the process with directly refusal.For the loan application for directly passing through proceeds to module of making loans of opening an account, for the loan of artificial examination ＆ approvalApplication proceeded to and examined under line, and remaining loan application is directly rejected loans.Simultaneously different score values intervals are located at according to appraisal result, it is determined thatThe means of payment of customer lending.
It should be noted that each embodiment of the real-time measures and procedures for the examination and approval of above-mentioned loan and the real-time approval system of the loanCorresponding technology contents it is completely the same, in order to avoid repeat, this is not repeated here.
Through the above description of the embodiments, those skilled in the art can be understood that the present invention can be byThe mode of software combined with hardware platform is realizing.Based on such understanding, technical scheme makes tribute to background technologyThat what is offered can be embodied in whole or in part in the form of software product, and the computer software product can be stored in storage and be situated betweenIn matter, such as ROM/RAM, magnetic disc, CD, including some instructions use is so that a computer equipment (can be individual calculusMachine, server, either network equipment etc.) perform method described in some parts of each embodiment of the invention or embodiment.
It will be appreciated by those skilled in the art that disclosed above is only embodiments of the present invention, certainly can notThe interest field of the present invention is limited with this, the equivalent variations made according to embodiment of the present invention still belong to the claims in the present inventionThe scope for being covered.