CN105931153A - Indirect questionnaire assessment method based on neural network prediction analysis model - Google Patents
Indirect questionnaire assessment method based on neural network prediction analysis model Download PDFInfo
- Publication number
- CN105931153A CN105931153A CN201610229841.9A CN201610229841A CN105931153A CN 105931153 A CN105931153 A CN 105931153A CN 201610229841 A CN201610229841 A CN 201610229841A CN 105931153 A CN105931153 A CN 105931153A
- Authority
- CN
- China
- Prior art keywords
- neural network
- output
- requirements
- layer
- hidden layer
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 33
- 230000007935 neutral effect Effects 0.000 claims description 29
- 238000011156 evaluation Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 12
- 230000004913 activation Effects 0.000 claims description 6
- 210000002569 neuron Anatomy 0.000 claims description 6
- 238000012360 testing method Methods 0.000 claims description 5
- 238000005538 encapsulation Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 claims description 3
- 238000011835 investigation Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000000644 propagated effect Effects 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims 1
- 230000014509 gene expression Effects 0.000 description 3
- 238000013459 approach Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses an indirect questionnaire assessment method based on a neural network prediction analysis model, and relates to the field of automation. The method comprises that questionnaire related data is extracted; the extracted questionnaire related data is preprocessed to obtain real and qualified data; a BP neural network assessment model is constructed based on the real and qualified data; the model is trained to obtain a more effective and reasonable assessment model; knowledge obtained via the analysis model and professional knowledge are packaged in a decision service to execute a decision; and the decision service is used to determine the achievement degree of graduation requirements. The method of the invention is wide in coverage range, high in efficiency, and capable of autonomous analysis and decision and assessing learning results of students objectively.
Description
Technical field
The present invention relates to autoknowledge field, devise a kind of questionnaire survey based on Neural Network Prediction model
Indirect assessment method.
Background technology
Since in March, 2015 Ministry of Education releases " Engineering Education Professional Certification Standards " equivalent with international essence, the whole nation
Ge great colleges and universities are all actively promoting Engineering Education Professional Certification work.
Also there are different evaluations of programme in each colleges and universities for different requirements for graduations, are mostly divided into diagnosis type evaluation, formation type
Evaluating, total junction type evaluation, owing to can relatively accurately quantify, existing Engineering Education Professional Certification scheme is mostly course and examines
Core this scheme of achievement analytic process.The development such as knowledge that student can have been formed by this scheme, ability, attitude are made reasonably
Evaluate, but be excessively inclined to total marks of the examination and cause open wretched insufficiency.
For solving the problems referred to above, it is indirect that the present invention devises a kind of questionnaire survey based on Neural Network Prediction model
Appraisal procedure, the method is versatile and flexible, have qualitative, the advantage that quantitatively combines of opening.
Summary of the invention
In order to overcome the deficiency of current appraisal procedure, the present invention devises a kind of based on Neural Network Prediction model
Questionnaire survey indirect assessment method, it specifically includes following steps:
1) questionnaire survey related data is extracted;
2) the questionnaire survey related data extracted is carried out pretreatment, it is thus achieved that true and qualified data;
3) based on true and qualified data construct BP neutral net assessment models;
4) training pattern is to obtain more efficient and rational assessment models;
5) by by analyzing the knowledge of model acquisition and combining in the decision service that expertise is packaged in execution decision-making;
6) decision service is utilized to judge requirements for graduation degree of reaching.
Requirements for graduation degree of the reaching indirect assessment method of above-mentioned knowledge based automatization, it is characterised in that: described step 1)
Including:
The object of questionnaire survey includes: employing unit, graduate, graduating student;
Investigation content includes that two: one is the approval degree of interviewee's ability every to requirements for graduation importance;Two is graduate
These abilities show and reaches situation.
Above-mentioned questionnaire survey indirect assessment method based on Neural Network Prediction model, it is characterised in that: described step
Rapid 2) farther include:
It is standardized above-mentioned data acquired processing:
Formulate a requirements for graduation filled in for interviewee according to above index and reach condition survey table, each metrics evaluation result
Be divided into " do not admit/the most dissatisfied " " do not admit/the most dissatisfied " " typically admit/the most satisfied " " substantially recognize
With/be satisfied in the main " " very approval/the most satisfied " five grades;
Each index can be marked by each interviewee, and score value is the five-grade marking system, and the index system constituted by 12 indexs is commented
Estimating requirements for graduation degree of reaching, the jth index of i-th interviewee is cij, then this index being standardized processes formula as follows:
(1)
Wherein, xijIt is cijNormal data;It it is the most standardized jth index meansigma methods;SjIt it is the most standardized jth
The standard deviation of individual index;
(2)
(3)
Wherein, M is the number of evaluator, if standardized data are still above 1, then this item data is entered as 1.
Questionnaire survey indirect assessment method based on Neural Network Prediction model according to claim 1, its
Being characterised by, described step A103 farther includes:
To requirements for graduation degree of reaching evaluation problem, can be regarded as inputting (requirements for graduation degree of reaching evaluation index) (right to output
The graduation final assessment result of degree of reaching) nonlinear mapping, therefore, have employed 3 layers of BP neural network structure, i.e. input layer, hidden
Containing layer, output layer;
Concrete modeling process is made up of following 4 steps:
1) determination of input layer:
Requirements for graduation degree of reaching evaluation index is divided into 12,12 evaluation indexes as the input of BP neural network input layer, because of
This, the input layer number of BP neutral net is the most correspondingly defined as 12;
2) determination of output layer node:
Due to requirements for graduation degree of reaching assessment result only one of which, therefore the output layer of network only sets 1 output node;
3) determination of hidden layer node:
Up to the present, how to select optimal hidden layer node number and be still a problem anxious to be resolved.If we select
Very little, then the convergence rate that can make whole neutral net is slack-off for the node in hidden layer selected, and is difficult to convergence, on the contrary, if I
The node in hidden layer that selects too many, then the topological structure that can cause neutral net is complicated, iterative learning time computationally intensive, by mistake
The problems such as difference is the most optimal, implying node in addition the most also can increase the training time;
At present, the empirical equation of relatively common determination node in hidden layer amount has:
(4)
Wherein, p is hidden layer node number, and m is input layer number, and n is output layer node number,Value is more than 1
And less than 10;
By testing one by one, obtaining optimal node in hidden layer is 7, i.e. p=7;
4) activation primitive is selected:
Owing to, in training data sample set, the desired output of assessment result all falls within [0,1] district after normalized
In, therefore, the activation primitive on BP neutral net hidden layer unit and output layer unit can be all taken as Sigmoid by us
Function, functional form is:
(5)
In this model:
Input vector be X=(X1, X2 ... Xm), m is the number of input layer;
Hidden layer be output as H=(H1, H2 ... Hp), p is the number of hidden layer node;
Model is actual is output as Y=(y);
D=(d) represents training sample desired output;
Input layer unit i to the weights of hidden layer unit j is;
Hidden layer unit j to the weights of output layer unit is;
Introduce x0=-1, can be that hidden layer neuron introducing threshold values is;
Introduce h0=-1, can be that output layer neuron introducing threshold values is;
For hidden layer, there is an equation below:
(6)
Wherein, j=1,2 ... p;
For output layer, there is an equation below:
(7).
Questionnaire survey indirect assessment method based on Neural Network Prediction model according to claim 1, its
Being characterised by, described step A104 farther includes:
Input vector X is substituted into repeatedly BP neural network model, and specify correspondence correctly export result, first calculate and actually enter
Obtain calculating error, reverse propagated error in neutral net with the difference of correct output, then calculate each weight to error
Contribution, and be adjusted connecting weights on this basis, neutral net constantly oneself adjust weights and make error
It is little, until it all can get correct output to all of input.
Questionnaire survey indirect assessment method based on Neural Network Prediction model according to claim 1, its
Being characterised by, described step A105 farther includes:
The knowledge of encapsulation is deployed in the decision service controlled by business process management system by Business Rules Management System.
Questionnaire survey indirect assessment method based on Neural Network Prediction model according to claim 1, its
Being characterised by, described step A106 farther includes:
The last output valve of BP neutral net assessment models is the value of requirements for graduation degree of reaching.
The method have technical effect that: the present invention collects questionnaire survey data by questionnaire method, adjust based on questionnaire
Look into data construct BP neutral net assessment models, and carry out indirect assessment graduate's requirements for graduation degree of reaching with this assessment models;
Devise the decision process of requirements for graduation degree of reaching, and construct the decision-making management circulation framework of requirements for graduation degree of reaching;Use
" operation flow-decision service-professional knowledge-analysis model-data-operation flow " this circulate approach, it is achieved thereby that
The automatization of indirect assessment requirements for graduation degree of reaching.
Accompanying drawing explanation
Fig. 1 is the implementing procedure figure of the present invention;
Fig. 2 is the BP neutral net assessment models figure of the present invention;
Fig. 3 is the decision-making management circulation Organization Chart of requirements for graduation degree of reaching in the present invention;
Fig. 4 is the decision flow diagram of requirements for graduation degree of reaching in the present invention;
Detailed description of the invention
Below in conjunction with the accompanying drawings technical scheme is described in further detail.It should be noted that explanation below
That be merely exemplary rather than in order to limit the scope of the present invention and its application.
Project implementation mode
As it is shown in figure 1, a kind of questionnaire survey indirect assessment method based on Neural Network Prediction model, detailed description of the invention
As follows:
1) questionnaire survey related data is extracted;
2) the questionnaire survey related data extracted is carried out pretreatment, it is thus achieved that true and qualified data;
3) based on true and qualified data construct BP neutral net assessment models;
4) training pattern is to obtain more efficient and rational assessment models;
5) by by analyzing the knowledge of model acquisition and combining in the decision service that expertise is packaged in execution decision-making;
6) decision service is utilized to judge requirements for graduation degree of reaching.
Above-mentioned questionnaire survey indirect assessment method based on Neural Network Prediction model, it is characterised in that: described step
Rapid 1) including:
The object of questionnaire survey includes: employing unit, graduate, graduating student;
Investigation content includes that two: one is the approval degree of interviewee's ability every to requirements for graduation importance;Two is graduate
These abilities show and reaches situation.
Above-mentioned questionnaire survey indirect assessment method based on Neural Network Prediction model, it is characterised in that: described step
Rapid 2) farther include:
It is standardized above-mentioned data acquired processing:
Formulate a requirements for graduation filled in for interviewee according to above index and reach condition survey table, each metrics evaluation result
Be divided into " do not admit/the most dissatisfied " " do not admit/the most dissatisfied " " typically admit/the most satisfied " " substantially recognize
With/be satisfied in the main " " very approval/the most satisfied " five grades;
Each index can be marked by each interviewee, and score value is the five-grade marking system, and the index system constituted by 12 indexs is commented
Estimating requirements for graduation degree of reaching, the jth index of i-th interviewee is cij, then this index being standardized processes formula as follows:
(1)
Wherein, xijIt is cijNormal data;It it is the most standardized jth index meansigma methods;SjIt it is the most standardized jth
The standard deviation of individual index;
(2)
(3)
Wherein, M is the number of evaluator, if standardized data are still above 1, then this item data is entered as 1.
Questionnaire survey indirect assessment method based on Neural Network Prediction model according to claim 1, its
Being characterised by, described step A103 farther includes:
To requirements for graduation degree of reaching evaluation problem, can be regarded as inputting (requirements for graduation degree of reaching evaluation index) (right to output
The graduation final assessment result of degree of reaching) nonlinear mapping, therefore, have employed 3 layers of BP neural network structure, i.e. input layer, hidden
Containing layer, output layer;
Concrete modeling process is made up of following 4 steps:
1) determination of input layer:
Requirements for graduation degree of reaching evaluation index is divided into 12,12 evaluation indexes as the input of BP neural network input layer, because of
This, the input layer number of BP neutral net is the most correspondingly defined as 12;
2) determination of output layer node:
Due to requirements for graduation degree of reaching assessment result only one of which, therefore the output layer of network only sets 1 output node;
3) determination of hidden layer node:
Up to the present, how to select optimal hidden layer node number and be still a problem anxious to be resolved.If we select
Very little, then the convergence rate that can make whole neutral net is slack-off for the node in hidden layer selected, and is difficult to convergence, on the contrary, if I
The node in hidden layer that selects too many, then the topological structure that can cause neutral net is complicated, iterative learning time computationally intensive, by mistake
The problems such as difference is the most optimal, implying node in addition the most also can increase the training time;
At present, the empirical equation of relatively common determination node in hidden layer amount has:
(4)
Wherein, p is hidden layer node number, and m is input layer number, and n is output layer node number,Value is more than 1
And less than 10;
By testing one by one, obtaining optimal node in hidden layer is 7, i.e. p=7;
4) activation primitive is selected:
Owing to, in training data sample set, the desired output of assessment result all falls within [0,1] district after normalized
In, therefore, the activation primitive on BP neutral net hidden layer unit and output layer unit can be all taken as Sigmoid by us
Function, functional form is:
(5)
In this model:
Input vector be X=(X1, X2 ... Xm), m is the number of input layer;
Hidden layer be output as H=(H1, H2 ... Hp), p is the number of hidden layer node;
Model is actual is output as Y=(y);
D=(d) represents training sample desired output;
Input layer unit i to the weights of hidden layer unit j is;
Hidden layer unit j to the weights of output layer unit is;
Introduce x0=-1, can be that hidden layer neuron introducing threshold values is;
Introduce h0=-1, can be that output layer neuron introducing threshold values is;
For hidden layer, there is an equation below:
(6)
Wherein, j=1,2 ... p;
For output layer, there is an equation below:
(7).
Questionnaire survey indirect assessment method based on Neural Network Prediction model according to claim 1, its
Being characterised by, described step A104 farther includes:
Input vector X is substituted into repeatedly BP neural network model, and specify correspondence correctly export result, first calculate and actually enter
Obtain calculating error, reverse propagated error in neutral net with the difference of correct output, then calculate each weight to error
Contribution, and be adjusted connecting weights on this basis, neutral net constantly oneself adjust weights and make error
It is little, until it all can get correct output to all of input.
Questionnaire survey indirect assessment method based on Neural Network Prediction model according to claim 1, its
Being characterised by, described step A105 farther includes:
The knowledge of encapsulation is deployed in the decision service controlled by business process management system by Business Rules Management System.
Questionnaire survey indirect assessment method based on Neural Network Prediction model according to claim 1, its
Being characterised by, described step A106 farther includes:
The last output valve of BP neutral net assessment models is the value of requirements for graduation degree of reaching.
2. embodiment
The present embodiment is as a example by the requirements for graduation degree of reaching of Automation Specialty is assessed.In order to the tool of the present invention is further illustrated
Body implementation process, provides 12 questionnaires to employing unit altogether, regains effective questionnaire 10 parts.The Index Content investigated is as shown in table 1, right
12 indexs in table 1, some index weights are relatively big, and have is the least, so, it is determined by weight coefficient and embodies a finger
Difference between mark, the weight coefficient of the present embodiment is determined by veteran expert data information by inquiry.
TableAssessment indicator system
Requirements for graduation degree of recognition and reach 1 ~ 5 point of expression of situation, wherein 1 point of expression do not admit/the most dissatisfied, 2 points
Represent do not admit/the most dissatisfied, 3 points represent general approval/the most satisfied, and 4 points of expressions are substantially admitted/are satisfied in the main, 5
Point represent very approval/the most satisfied.Table 2 is requirements for graduation degree of recognition application form, and table 3 reaches condition survey for requirements for graduation
Table, table 4 is for requirements for graduation degree of recognition and reaches situation summary sheet.
Table 2 requirements for graduation degree of recognition application form
Table 3 requirements for graduation reaches condition survey table
Table 4 requirements for graduation degree of recognition and reach situation summary sheet
Requirements for graduation degree of recognition application form and requirements for graduation are reached the data in condition survey table be analyzed and at standardization
Reason, as shown in Table 5,6.
Data (degree of recognition) after table 5 standardization
Data (reaching situation) after table 6 standardization
Each requirements for graduation is all to be passed judgment on by employing unit, and wherein approval angle value accounts for the 40% of integrated standardization value, reaches case values
Account for the 60% of integrated standardization value, as shown in table 7.
Data (10 assessment samples) after table 7 integrated standardization
Integrated standardization value can choose above-mentioned 10 assessment samples as the input value of BP neutral net assessment models, such as table 6,
And split data into two parts, and choose 8 samples above as training sample, 4 samples below, as test sample, are learned
Habit precision is, after 995 times are trained, its assessment result is shown in Table 8, after the assessment result of 4 test samples and desired output
Assessment result is shown in Table 9.
Table 8 BP neutral net assessment result
Table 9 BP neutral net assessment result and desired output Comparative result
As can be seen from Table 9, assessment result with desired output result closely, illustrates that this assessment result error is less.
Claims (7)
1. a questionnaire survey indirect assessment method based on Neural Network Prediction model, it is characterised in that described based on
The questionnaire survey indirect assessment method of Neural Network Prediction model includes:
Step S101, extracts questionnaire survey related data;
Step S102, carries out pretreatment to the questionnaire survey related data extracted, it is thus achieved that true and qualified data;
Step S103, based on true and qualified data construct BP neutral net assessment models;
Step S104, training pattern is to obtain more efficient and rational assessment models;
Step S105, by by analyze model obtain knowledge and combine expertise be packaged in perform decision-making decision service
In;
Step S106, utilizes decision service to judge requirements for graduation degree of reaching.
Questionnaire survey indirect assessment method based on Neural Network Prediction model the most according to claim 1, it is special
Levying and be, described questionnaire survey related data includes:
The object of questionnaire survey includes: employing unit, graduate, graduating student;
Investigation content includes that two: one is the approval degree of interviewee's ability every to requirements for graduation importance;Two is graduate
These abilities show and reaches situation.
Questionnaire survey indirect assessment method based on Neural Network Prediction model the most according to claim 1, it is special
Levying and be, described step S102 farther includes:
It is standardized above-mentioned data acquired processing:
Formulate a requirements for graduation filled in for interviewee according to above index and reach condition survey table, each index evaluation result
Be divided into " do not admit/the most dissatisfied " " do not admit/the most dissatisfied " " typically admit/the most satisfied " " substantially recognize
With/be satisfied in the main " " very approval/the most satisfied " five grades;
Each index can be marked by each interviewee, and score value is the five-grade marking system, and the index system constituted by 12 indexs is commented
Estimating requirements for graduation degree of reaching, the jth index of i-th interviewee is cij, then this index being standardized processes formula as follows:
Wherein, xijIt is cijNormal data;It it is the most standardized jth index meansigma methods;SjIt it is the most standardized jth
The standard deviation of index;
Wherein, M is the number of evaluator, if standardized data are still above 1, then this item data is entered as 1.
Questionnaire survey indirect assessment method based on Neural Network Prediction model the most according to claim 1, it is special
Levying and be, described step S103 farther includes:
To requirements for graduation degree of reaching evaluation problem, can be regarded as inputting (requirements for graduation degree of reaching evaluation index) (right to output
The graduation final assessment result of degree of reaching) nonlinear mapping, therefore, have employed 3 layers of BP neural network structure, i.e. input layer, hidden
Containing layer, output layer;
Concrete modeling process is made up of following 4 steps:
1) determination of input layer:
Requirements for graduation degree of reaching evaluation index is divided into 12,12 evaluation indexes as the input of BP neural network input layer, because of
This, the input layer number of BP neutral net is the most correspondingly defined as 12;
2) determination of output layer node:
Due to requirements for graduation degree of reaching assessment result only one of which, therefore the output layer of network only sets 1 output node;
3) determination of hidden layer node:
Up to the present, how to select optimal hidden layer node number and be still a problem anxious to be resolved, if we select
Very little, then the convergence rate that can make whole neutral net is slack-off for the node in hidden layer selected, and is difficult to convergence, on the contrary, if I
The node in hidden layer that selects too many, then the topological structure that can cause neutral net is complicated, iterative learning time computationally intensive, by mistake
The problems such as difference is the most optimal, implying node in addition the most also can increase the training time;
At present, the empirical equation of relatively common determination node in hidden layer amount has:
Wherein, p is hidden layer node number, and m is input layer number, and n is output layer node number,Value more than 1 and
Less than 10;
By testing one by one, obtaining optimal node in hidden layer is 7, i.e. p=7;
4) activation primitive is selected:
Owing to, in training data sample set, the desired output of assessment result all falls within [0,1] district after normalized
In, therefore, the activation primitive on BP neutral net hidden layer unit and output layer unit can be all taken as Sigmoid by us
Function, functional form is:
In this model:
Input vector be X=(X1, X2 ... Xm), m is the number of input layer;
Hidden layer be output as H=(H1, H2 ... Hp), p is the number of hidden layer node;
Model is actual is output as Y=(y);
D=(d) represents training sample desired output;
Input layer unit i to the weights of hidden layer unit j is;
Hidden layer unit j to the weights of output layer unit is;
Introduce x0=-1, can be that hidden layer neuron introducing threshold values is;
Introduce h0=-1, can be that output layer neuron introducing threshold values is;
For hidden layer, there is an equation below:
Wherein, j=1,2 ... p;
For output layer, there is an equation below:
。
Questionnaire survey indirect assessment method based on Neural Network Prediction model the most according to claim 1, it is special
Levying and be, described step S104 farther includes:
Input vector X is substituted into repeatedly BP neural network model, and specify correspondence correctly export result, first calculate and actually enter
Obtain calculating error, reverse propagated error in neutral net with the difference of correct output, then calculate each weight to error
Contribution, and be adjusted connecting weights on this basis, neutral net constantly oneself adjust weights and make error
It is little, until it all can get correct output to all of input.
Questionnaire survey indirect assessment method based on Neural Network Prediction model the most according to claim 1, it is special
Levying and be, described step S105 farther includes:
The knowledge of encapsulation is deployed in the decision service controlled by business process management system by Business Rules Management System.
Questionnaire survey indirect assessment method based on Neural Network Prediction model the most according to claim 1, it is special
Levying and be, described step S106 farther includes:
The last output valve of BP neutral net assessment models is the value of requirements for graduation degree of reaching.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610229841.9A CN105931153A (en) | 2016-04-14 | 2016-04-14 | Indirect questionnaire assessment method based on neural network prediction analysis model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610229841.9A CN105931153A (en) | 2016-04-14 | 2016-04-14 | Indirect questionnaire assessment method based on neural network prediction analysis model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105931153A true CN105931153A (en) | 2016-09-07 |
Family
ID=56838106
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610229841.9A Pending CN105931153A (en) | 2016-04-14 | 2016-04-14 | Indirect questionnaire assessment method based on neural network prediction analysis model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105931153A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107330309A (en) * | 2017-06-14 | 2017-11-07 | 广东网金控股股份有限公司 | A kind of security protection method and system based on neutral net |
CN108549987A (en) * | 2018-04-18 | 2018-09-18 | 河南理工大学 | A kind of Course Assessment method based on oriented ring analysis |
CN109186533A (en) * | 2018-07-13 | 2019-01-11 | 南京理工大学 | A kind of ground air defense radar shield angle calculation method based on BP neural network algorithm |
CN109388138A (en) * | 2017-08-08 | 2019-02-26 | 株式会社万都 | Automatic driving vehicle, automatic Pilot control device and automatic Pilot control method based on deep learning |
CN110222925A (en) * | 2019-04-24 | 2019-09-10 | 深圳证券交易所 | Performance quantization wire examination method, device and computer readable storage medium |
CN111834007A (en) * | 2020-07-14 | 2020-10-27 | 上海市第一妇婴保健院 | Data processing system and method for investigation of child developmental coordination disorder |
CN113052686A (en) * | 2021-04-30 | 2021-06-29 | 中国银行股份有限公司 | Data processing method and device |
CN117457234A (en) * | 2023-11-20 | 2024-01-26 | 四川互慧软件有限公司 | Snake bite grade prediction method and system based on neural network model |
-
2016
- 2016-04-14 CN CN201610229841.9A patent/CN105931153A/en active Pending
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107330309A (en) * | 2017-06-14 | 2017-11-07 | 广东网金控股股份有限公司 | A kind of security protection method and system based on neutral net |
CN109388138A (en) * | 2017-08-08 | 2019-02-26 | 株式会社万都 | Automatic driving vehicle, automatic Pilot control device and automatic Pilot control method based on deep learning |
CN109388138B (en) * | 2017-08-08 | 2024-05-14 | 汉拿科锐动电子股份公司 | Deep learning-based automatic driving vehicle, automatic driving control device and automatic driving control method |
CN108549987A (en) * | 2018-04-18 | 2018-09-18 | 河南理工大学 | A kind of Course Assessment method based on oriented ring analysis |
CN108549987B (en) * | 2018-04-18 | 2021-09-03 | 河南理工大学 | Course assessment method based on directed loop analysis |
CN109186533A (en) * | 2018-07-13 | 2019-01-11 | 南京理工大学 | A kind of ground air defense radar shield angle calculation method based on BP neural network algorithm |
CN110222925A (en) * | 2019-04-24 | 2019-09-10 | 深圳证券交易所 | Performance quantization wire examination method, device and computer readable storage medium |
CN110222925B (en) * | 2019-04-24 | 2022-04-08 | 深圳证券交易所 | Performance quantitative assessment method and device and computer readable storage medium |
CN111834007A (en) * | 2020-07-14 | 2020-10-27 | 上海市第一妇婴保健院 | Data processing system and method for investigation of child developmental coordination disorder |
CN113052686A (en) * | 2021-04-30 | 2021-06-29 | 中国银行股份有限公司 | Data processing method and device |
CN113052686B (en) * | 2021-04-30 | 2024-03-08 | 中国银行股份有限公司 | Data processing method and device |
CN117457234A (en) * | 2023-11-20 | 2024-01-26 | 四川互慧软件有限公司 | Snake bite grade prediction method and system based on neural network model |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105931153A (en) | Indirect questionnaire assessment method based on neural network prediction analysis model | |
CN103942993B (en) | A kind of self adaptation online exam system and method thereof based on IRT | |
CN104077227B (en) | The evaluation system of a kind of program source code based on B/S framework and method | |
Gadhavi et al. | Student final grade prediction based on linear regression | |
CN102222313A (en) | Urban evolution simulation structure cell model processing method based on kernel principal component analysis (KPCA) | |
CN113408957B (en) | Classroom teaching evaluation method based on combined weighting method | |
CN109272160A (en) | Score on Prediction system and prediction technique | |
Thunis et al. | A procedure for air quality models benchmarking | |
CN102591929A (en) | Library data processing system and data processing method thereof | |
Snyder et al. | Evaluation of integrated assessment model hindcast experiments: a case study of the GCAM 3.0 land use module | |
Mohamed et al. | Assessment method for course outcome and program outcome in Outcome Based Education (OBE) | |
CN105930272A (en) | Important statement-based branch coverage testing data generation method | |
CN105389646A (en) | Evaluation method based on variable-weight clustering method of gray whitening function | |
Chodakowska | The future of evaluation of lower secondary schools’ management | |
Scippacercola et al. | PRINCIPAL COMPONENT ANALYSIS TO RANKING TECHNICAL EFFICIENCIES THROUGH STOCHASTIC FRONTIER ANALYSIS AND DEA. | |
Münch et al. | Creating an advanced software engineering laboratory by combining empirical studies with process simulation | |
Zhang et al. | Multivariate discrete grey model base on dummy drivers | |
CN105069543A (en) | All perspective feedback evaluation assessment method based on grey clustering | |
CN109272233A (en) | A kind of employee's competency appraisal procedure closed based on type-2 fuzzy sets | |
Mostoli et al. | Using the Malmquist Index in evaluation process to enhance mathematical literacy in high school students | |
Huang et al. | RESEARCH ON INDEPENDENT COLLEGE TEACHERS’TEACHING ABILITY BASED ON FACTOR ANALYSIS IN SPSS | |
Imbar et al. | Development of an instrument to measure smart campus levels in Indonesian institutions of higher education | |
Roy et al. | Methodological problems in the formulation and validation of system dynamics models incorporating soft variables | |
Kim | Priority Analysis of PBL Evaluation Score using AHP | |
Alajbeg et al. | Evaluation of the impact of changing the number of test tasks on successful completion for automated LMS |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20160907 |