CN110046667A - A kind of Method of Teaching Appraisal based on deep neural network study score data pair - Google Patents
A kind of Method of Teaching Appraisal based on deep neural network study score data pair Download PDFInfo
- Publication number
- CN110046667A CN110046667A CN201910319686.3A CN201910319686A CN110046667A CN 110046667 A CN110046667 A CN 110046667A CN 201910319686 A CN201910319686 A CN 201910319686A CN 110046667 A CN110046667 A CN 110046667A
- Authority
- CN
- China
- Prior art keywords
- course
- teaching
- evaluation
- student
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 44
- 238000011156 evaluation Methods 0.000 claims abstract description 122
- 239000013598 vector Substances 0.000 claims abstract description 40
- 230000008569 process Effects 0.000 description 10
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000004913 activation Effects 0.000 description 4
- 238000007796 conventional method Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000010835 comparative analysis Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000009412 basement excavation Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000013604 expression vector Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003340 mental effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 210000004218 nerve net Anatomy 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- 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
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Tourism & Hospitality (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Strategic Management (AREA)
- Biophysics (AREA)
- Educational Technology (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Educational Administration (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Electrically Operated Instructional Devices (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention provides a kind of Method of Teaching Appraisal based on deep neural network study score data pair, it the described method comprises the following steps: using course as node, a students' needs figure is constructed by side of students' needs, then course is divided by the method for figure cluster by several class groups, course in same class group has similar curricula-variable student group, and the course of different class groups has different curricula-variable student groups;In each class group, teaching evaluation pair is generated to the teaching evaluation database of course according to student, the teaching evaluation is to being teaching evaluation result of the student to two subjects journey;By the evaluation pair of deep neural network learning teaching, the teaching evaluation for obtaining every subject indicates vector;The teaching evaluation for clustering every subject indicates vector, and a class group is divided into the grouping of several courses, then by counting the teaching evaluation pair between different course groupings, obtains course evaluation result to the end.More fair, just teaching evaluation result that the present invention can obtain.
Description
Technical field
The present invention relates to Method of Teaching Appraisal technical fields, more particularly to one kind based on deep neural network study scoring
The Method of Teaching Appraisal of data pair.
Background technique
According to " country in long-term educational reform and development planning outline (2010-2020) " spirit with " Ministry of Education about
The opinion of gerneral institutes of higher education's Undergraduate Teaching Evaluation work ", Regular Colleges quality of instruction is improved, prominent Discipline construction mentions
Teaching Quality Guarantee is risen, is the importance that Talent-cultivating Quality is continuously improved.Curriculum teaching guarantee is colleges and universities' religion
Educate an important content of one of most basic most crucial content of Teaching Quality Guarantee and Higher Educational Evaluation.
Effectively Method of Teaching Appraisal can be greatly facilitated the enthusiasm of teachers ' teaching, improve teaching efficiency.However, at present
Most assessments of teaching efficiency both at home and abroad are all to give a mark by student to the evaluation of current course, then do letter on this basis
Single analysis and excavation.These methods can be mainly divided into two classes.The first kind is the statistical with student to course score data
Teaching evaluation is carried out based on analysis, such methods are ranked up course evaluation by calculating the statistics such as average mark, variance.This
The information source of class method foundation is single, high to the reliability requirement of original score data.However, real data often noise
It is more, data are lack of standardization, such as to course scoring by total marks of the examination, student give a mark when mood and teacher to the stringent of class management
The influence of the factors such as degree, in addition, each student's scoring criterion disunity.Therefore, simply the subject is commented using student
Divide to evaluate the teaching level and quality for reacting course that the teaching efficiency of the course often can not be objective, just.
Second class method is to carry out teaching evaluation in the method for data mining, for example excavate using association rules method
Raw evaluation result and teacher the implicit associations relationship between the essential informations such as gender, educational background, age, academic title, but it is this kind of
Method generally can not effectively merge different types of feature, as the total marks of the examination information of student's scoring, student integrate
Evaluation.
In recent years, as the fast development of information technology and colleges and universities continue to increase information-based investment, big data is
Major university campus is come into, how really the effective big data for utilizing school instruction and generating in managing carries out education and instruction
Evaluation guarantees that the objective, just of teaching evaluation becomes an important problem in teaching evaluation research.Current course teaching evaluation
Be primarily present two big challenges: (1) different students causes different students to same course there are different Psychological Assessment standards
Scoring does not have comparativity, therefore all students of simple computation compare the average mark of the course evaluation and impart knowledge to students between different courses
As a result quality tends not to satisfactory;(2) a plurality of types of data can be generated in teaching process, as student performance, student give
Scoring, course types of course etc., how to comprehensively consider a plurality of types of data characteristicses to carry out course teaching evaluation scoring is
Problem.
Summary of the invention
The present invention against the above deficiency place, provide it is a kind of based on deep neural network study score data pair teaching comment
Valence method, to overcome existing method different students in scoring process to have different Psychological assessment standards and isomeric data special
Fusion rules problem is levied, so that teaching evaluation is more objective, just.
A kind of Method of Teaching Appraisal based on deep neural network study score data pair, comprising the following steps:
Using course as node, a students' needs figure is constructed by side of students' needs, then passes through the method handle of figure cluster
Course is divided into several class groups, and the course in same class group has similar curricula-variable student group, and the course of different class groups has not
Same curricula-variable student group;
In each class group, teaching evaluation pair is generated according to teaching evaluation database of the student to course, the teaching is commented
Valence is to being teaching evaluation result of the student to two subjects journey;
By the evaluation pair of deep neural network learning teaching, the teaching evaluation for obtaining every subject indicates vector;
The teaching evaluation for clustering every subject indicates vector, and a class group is divided into the grouping of several courses, is then passed through
The teaching evaluation pair between different course groupings is counted, course evaluation result to the end is obtained.
It is above-mentioned based on deep neural network study score data pair Method of Teaching Appraisal, wherein it is described with course be section
Point constructs a students' needs figure by side of students' needs, and course is then divided into several class groups' by the method for figure cluster
Step specifically includes:
Using course as node, if there is w students have taken as an elective course certain two subjects journey simultaneously, then there is one between two subjects journey
Side, the weight on side is w, in this way, all courses constitute a weighted graph, then all courses are drawn using figure clustering method
It is divided into several class groups.
The above-mentioned Method of Teaching Appraisal based on deep neural network study score data pair, wherein described in each class group
It is interior, it is specifically included according to the step that teaching evaluation database of the student to course generates teaching evaluation pair:
In each class group, is given to score to generate to oneself courses taken according to each student and be scored to < si,sj>, i.e., certain
Student is higher than course j to the scoring of course i, if certain student has taken as an elective course n subject, produces n × (n- for the student
1)/2 teaching scoring pair.
The above-mentioned Method of Teaching Appraisal based on deep neural network study score data pair, wherein described to pass through depth mind
Through e-learning teaching evaluation pair, obtains the teaching evaluation of every subject and is specifically included the step of indicating vector:
Two deep neural networks are inputted respectively according to the feature of the feature of student, teaching scoring centering two subjects journey, are led to
Overfitting, which predicts the student, indicates vector to the teaching scoring of this two subjects journey to learn the teaching evaluation of this two subjects.
The above-mentioned Method of Teaching Appraisal based on deep neural network study score data pair, wherein the every subject of cluster
The teaching evaluation of journey indicates vector, and a class group is divided into the grouping of several courses, is then grouped it by counting different courses
Between teaching evaluation pair, the step of obtaining course evaluation result to the end specifically includes:
For each class group, the course teaching evaluation in cluster class group indicates vector, obtains k course grouping, course religion
It learns evaluation and is finally divided into k grade, count the course teaching evaluation pair between different course groupings, determine each course grouping institute
The teaching evaluation grade of category.
The above-mentioned Method of Teaching Appraisal based on deep neural network study score data pair, wherein the method also includes:
The feature of student is inputted in a multilayer neural network first, study obtains a student characteristics coding;Class
The feature of journey teaching evaluation centering two subjects journey inputs a multilayer neural network with identical parameters respectively, respectively obtains two
A course teaching evaluation indicates vector;Then student characteristics vector is indicated that vector connects into two course teaching evaluations respectively
Two [student characteristics-course teaching evaluation indicates] vectors, input has a multilayer neural network of identical parameters respectively, in advance
Measure scoring of the student to this two subjects journey;Finally the height of prediction scoring is really scored just to input with this two subjects journey and be damaged
Function is lost, carrying out learned lesson teaching evaluation indicates vector.
The above-mentioned Method of Teaching Appraisal based on deep neural network study score data pair, wherein obtained course evaluation
As a result the course teaching evaluation result having the same in, in identical course evaluation grouping.
Method of Teaching Appraisal proposed by the present invention based on deep neural network study score data pair, utilizes depth nerve
Then the identical student of e-learning leads to the comparative evaluation of different courses so that the teaching evaluation vector of study to course indicates
Cluster course teaching evaluation vector is crossed to obtain the teaching evaluation grade of course.
Compared with prior art, the present invention is taught by the teaching evaluation data pair of study and using the fusion of deep learning method
The multiple types of data that process generates carrys out overall merit teaching efficiency grade, and result is more objective than conventional method, just,
It is particularly advantageous in that:
(1) due to allowing same student to compare the teaching efficiency of two different courses, student is easier to provide the answer of oneself,
Therefore, the present invention obtains course teaching evaluation expression vector to the comparative evaluation of different courses by learning same student,
As a result more objective.
(2) present invention devises a new deep neural network framework and carries out learning teaching evaluation data pair, at this
In structure, a variety of different features can be merged: such as student characteristics, student examination achievement, course teaching scoring, course types,
Carry out synthetic instruction evaluation.
Detailed description of the invention
Fig. 1 is the Method of Teaching Appraisal for learning score data pair based on deep neural network that one embodiment of the invention provides
Flow diagram;
Fig. 2 is student's elective exemplary relationship figure;
Fig. 3 is the flow diagram for generating course evaluation data pair;
Fig. 4 is the deep neural network frame diagram that course evaluation vector table dendrography is practised;
Fig. 5 is course teaching grading flow diagram;
Fig. 6 is that data are evaluated between course grouping to statistical flowsheet schematic diagram.
Specific embodiment
To facilitate the understanding of the present invention, a more comprehensive description of the invention is given in the following sections with reference to the relevant attached drawings.In attached drawing
Give several embodiments of the invention.But the invention can be realized in many different forms, however it is not limited to this paper institute
The embodiment of description.On the contrary, purpose of providing these embodiments is make it is more thorough and comprehensive to the disclosure.
Unless otherwise defined, all technical and scientific terms used herein and belong to technical field of the invention
The normally understood meaning of technical staff is identical.Term as used herein in the specification of the present invention is intended merely to description tool
The purpose of the embodiment of body, it is not intended that in the limitation present invention.Term " and or " used herein includes one or more phases
Any and all combinations of the listed item of pass.
Referring to Fig. 1, the embodiment provides a kind of religions based on deep neural network study score data pair
Evaluation method is learned, includes the following steps S101~S104:
S101 constructs a students' needs figure by side of students' needs, then passes through the side of figure cluster using course as node
Course is divided into several class groups by method, and the course in same class group has similar curricula-variable student group, and the course of different class groups has
There is different curricula-variable student groups;
Wherein, class all living creatures at, using course as node, if there is w students have taken as an elective course certain two subjects journey simultaneously, then two subjects
There is a line between journey, the weight on side is w, in this way, all courses constitute a weighted graph, then utilizes figure clustering method handle
All courses are divided into several class groups.
Specific implementation process is as follows: class all living creatures at purpose be that all courses of whole school are divided into several parts, each portion
It is divided into a class group, the comparison of quality of instruction is then carried out inside class group, because of two totally different type of course teachings
The standard of evaluation may be different, for example, literature course put together with computer studies carry out quality of instruction compared with possibility ratio
It is more difficult and have little significance.In order to generate class group, the present invention constructs one using course as node first, with taken as an elective course simultaneously certain two
Student's quantity of subject is weight, constructs a weighted-graph G={ V, E, W }, and V indicates that node collection, E indicate line set, W
Indicate the weight of E, as shown in Figure 2.
Class all living creatures at implementation in use the figure clustering algorithm based on level, i.e., first every subject as one
Then repeatedly following two step: independent cluster (1) calculates any two cluster Ci, CjBetween similarity Dij, similarity dijIt adopts
Use the average value of weight between two cluster interior joints as the similarity of two clusters, it may be assumed thatWherein wpq
For the weight of node p, q, | Ci|、|Cj| it is respectively cluster Ci, CjThe number of interior joint;(2) merging in existing cluster has maximum similar
Two clusters of degree;Two above step is repeated until the number of cluster reaches specified threshold values;Finally, according to formulaThe isolated node generated in previous step and the cluster with less node, which are merged into, has maximum with it
In the cluster of similarity, class group required for the present invention is obtained.
S102 generates teaching evaluation pair, the religion according to teaching evaluation database of the student to course in each class group
Evaluation is learned to being teaching evaluation result of the student to two subjects journey;
Wherein, score data pair is generated, in each class group, is given to score to oneself courses taken according to each student and be given birth to
At scoring to < si,sj>, i.e., certain student is higher than course j to the scoring of course i, if certain student has taken as an elective course n subject, is directed to
The student produces n × (n-1)/2 teaching scoring pair.
Specific implementation process is as follows: generate score data to be obtained for next step training deep neural network it is every
A branch of instruction in school scoring indicates vector.One score data is to a training sample for being neural network.Generate score data pair
Process reads student's courses taken as shown in figure 3, reading first from database, and any two subjects journey is formed one
Course pair, if two subjects journey belongs to the same class group, the course is to an effective score data pair is belonged to, then this
Score data is stored in corresponding storage location to information, and marks its affiliated class group.The corresponding score data of deposit includes student
The scoring of characteristic information, the characteristic information of two subjects journey and student to two subjects journey.In specific implementation of the invention, learn
Raw characteristic information includes the scoring to its courses completed of achievement and the life of class corresponding to student, the life courses completed
Deng;Curriculum characteristic includes the type of course, Assessment, affiliated institute, all scorings for taking as an elective course student to the course.
S103, by the evaluation pair of deep neural network learning teaching, the teaching evaluation for obtaining every subject indicates vector;
Wherein, course teaching evaluation indicates vector study, according to the spy of the feature of student, teaching scoring centering two subjects journey
Sign inputs two deep neural networks respectively, which is predicted to the teaching of this two subjects journey scoring by study come learn this two
The teaching evaluation of subject indicates vector.
Specific implementation process is as follows: the present invention devises the teaching evaluation that a new neural network framework carrys out learned lesson
Indicate vector.The frame by learning teaching evaluate to come obtain every subject teaching evaluation indicate vector.New nerve net
Network framework is as shown in figure 4, in the network, using student characteristics and two subjects journey feature as input.Wherein student characteristics are defeated
Enter student's feature coding network and obtain student characteristics coding, two subjects journey feature inputs the course with identical parameters respectively
Feature coding network respectively obtains the coding of two subjects journey.Wherein student characteristics coding network and curriculum characteristic coding network point
Not Wei 64 × 32 and 128 × 32 two-tier networks, and use relu function as activation primitive, before activation primitive, increase and batch returns
One changes layer.After obtaining the coding of student's coding and two subjects journey, student's coding is connected with two course codings respectively, structure
At two students-course coding vector, a scoring network is inputted, scoring of the student to the subject is obtained.Score net
The three-layer network that network is one 64 × 32 × 1, equally uses relu function as activation primitive, before activation primitive, increases and criticizes
Normalize layer.Finally the scoring of two subjects journey is subtracted each other, with true scoring together entrance loss function.In the present invention, using flat
Variance loss function, i.e.,Wherein n is course score data to number, gprdIt is poor to score for neural network forecast
Away from glbThe true difference of two squares away from.Then whole network is trained using backpropagation, obtains network parameter.
S104, the teaching evaluation for clustering every subject indicate vector, a class group are divided into the grouping of several courses, then
By counting the teaching evaluation pair between different course groupings, course evaluation result to the end is obtained.
Wherein, the course teaching evaluation having the same in the course evaluation result obtained, in identical course evaluation grouping
As a result.Course teaching grading, for each class group, the course teaching evaluation in cluster class group indicates vector, obtains k course
Grouping, course teaching evaluation are finally divided into k grade, count the course teaching evaluation pair between different course groupings, determine every
Teaching evaluation grade belonging to one course grouping.
Specific implementation process is as follows: after the teaching evaluation for obtaining all courses indicates vector, the present invention according to setting in advance
Fixed comments religion number of levels k, clusters to each class group, then counts the course scoring pair between different grouping, determines every
A course is grouped affiliated opinion rating, as shown in Figure 5.Religion of traditional k-means algorithm to course is used in the present embodiment
Learning evaluation indicates that vector is clustered.It needs to be grouped this k after obtaining k course grouping by cluster for each class group
Determine grade.In implementation process, the present invention applies for that one is initially 0 k × k array S and is used to store course and comment first
Point pair statistical result.The numerical value of array S the i-th row jth column indicates that the course scoring in i-th of course grouping is higher than jth
The score data of course scoring in a course grouping is to number.In array S, the element on diagonal line is 0.Then it opens
Beginning calculates the value of array S, and process is as shown in Figure 6;For each scoring pair in class group, two subjects journey is first determined whether
Whether belong to the same class group grouping to abandon if it is the grouping of same course or the scoring having the same of two subjects journey;Such as
Fruit is not same course grouping, it is assumed that two subjects journey i, j is grouped C respectively from coursei、CjIf the scoring of course i is than course j
Height, then the value of the i-th row jth column of data matrix S increases by 1, if the scoring of course j is higher than course i, the jth of data matrix S
The value that row i-th arranges increases by 1, until having counted score data pair all in class group.
After obtaining statistical result matrix S, by calculating the sum of every a line come but opinion rating.Row with maximum sum is opposite
The course grouping opinion rating highest answered, representative have most students to think the grouping of this course than other course group instruction matter
Amount is more preferable.The corresponding course grouping opinion rating of row with minimum sum is minimum, represents the teaching that less student thinks the group
It is good that quality is grouped than other courses.
Method of Teaching Appraisal based on deep neural network study score data pair of the invention, first building students' needs
Relational graph divides curricula-variable relational graph by figure clustering method, obtains class group.Then it is directed to each class group, is believed according to student
Cease library, teaching scoring library, curriculum information library, production teaching score data pair;Each teaching score data is learned comprising one
The scoring of raw information, two subjects journey information and the life to this two subjects journey.Then the present invention devises a new depth mind
The teaching evaluation for learning every subject through the network architecture indicates vector.Finally, indicating vector by cluster course teaching evaluation
Course in class group is divided into k group, by Statistics Course be grouped between teaching evaluation data to determining each course point
The grade of group.Method of Teaching Appraisal based on deep neural network study score data pair of the invention overcomes in conventional method
In there is result error caused by different mental evaluation criterion due to different students, while considering in evaluation procedure
Influence of raw total marks of the examination, student characteristics and the curriculum characteristic to teaching evaluation.Therefore, the present invention can be obtained with respect to conventional method
To more objective, just teaching evaluation result.
One embodiment of the present invention above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
Limitations on the scope of the patent of the present invention therefore cannot be interpreted as.It should be pointed out that for those of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to guarantor of the invention
Protect range.Therefore, the scope of protection of the patent of the invention shall be subject to the appended claims.
Claims (7)
1. a kind of Method of Teaching Appraisal based on deep neural network study score data pair, which is characterized in that the method packet
Include following steps:
Using course as node, a students' needs figure is constructed by side of students' needs, then by the method for figure cluster course
It is divided into several class groups, the course in same class group has similar curricula-variable student group, and the course of different class groups has different
Curricula-variable student group;
In each class group, teaching evaluation pair, the teaching evaluation pair are generated according to teaching evaluation database of the student to course
It is teaching evaluation result of the student to two subjects journey;
By the evaluation pair of deep neural network learning teaching, the teaching evaluation for obtaining every subject indicates vector;
The teaching evaluation for clustering every subject indicates vector, and a class group is divided into the grouping of several courses, then passes through statistics
Teaching evaluation pair between different course groupings, obtains course evaluation result to the end.
2. the Method of Teaching Appraisal according to claim 1 based on deep neural network study score data pair, feature
It is, it is described using course as node, a students' needs figure is constructed by side of students' needs, then passes through the method handle of figure cluster
Course is divided into the step of several class groups and specifically includes:
Using course as node, if there is w students have taken as an elective course certain two subjects journey simultaneously, then there are a line, side between two subjects journey
Weight be then w is divided into all courses using figure clustering method in this way, all course constitutes a weighted graph
Several class groups.
3. the Method of Teaching Appraisal according to claim 2 based on deep neural network study score data pair, feature
It is, described in each class group, the step for generating teaching evaluation pair according to teaching evaluation database of the student to course is specific
Include:
In each class group, is given to score to generate to oneself courses taken according to each student and be scored to < si,sj>, i.e. certain student
It is higher than course j to the scoring of course i, if certain student has taken as an elective course n subject, n × (n-1)/2 is produced for the student
Teaching scoring pair.
4. the Method of Teaching Appraisal according to claim 3 based on deep neural network study score data pair, feature
Be, it is described by deep neural network learning teaching evaluation pair, obtain every subject teaching evaluation indicate vector the step of
It specifically includes:
Two deep neural networks are inputted respectively according to the feature of the feature of student, teaching scoring centering two subjects journey, pass through
It practises and predicts that the student scores to the teaching of this two subjects journey to learn the teaching evaluation of this two subjects and indicate vector.
5. the Method of Teaching Appraisal according to claim 4 based on deep neural network study score data pair, feature
It is, the teaching evaluation of the every subject of cluster indicates vector, and a class group is divided into the grouping of several courses, is then passed through
The step of counting the teaching evaluation pair between different course groupings, obtaining course evaluation result to the end specifically includes:
For each class group, the course teaching evaluation in cluster class group indicates vector, obtains k course grouping, and course teaching is commented
Valence is finally divided into k grade, counts the course teaching evaluation pair between different course groupings, determines belonging to each course grouping
Teaching evaluation grade.
6. the Method of Teaching Appraisal according to claim 1 based on deep neural network study score data pair, feature
It is, the method also includes:
The feature of student is inputted in a multilayer neural network first, study obtains a student characteristics coding;Course is taught
The feature for learning evaluation centering two subjects journey inputs a multilayer neural network with identical parameters respectively, respectively obtains two classes
Journey teaching evaluation indicates vector;Then student characteristics vector is indicated that vector connects into two with two course teaching evaluations respectively
[student characteristics-course teaching evaluation indicates] vector, input has a multilayer neural network of identical parameters respectively, predicts
Scoring of the student to this two subjects journey;Finally height and this two subjects journey of prediction scoring are really scored height entrance loss letter
Number, carrying out learned lesson teaching evaluation indicates vector.
7. according to claim 1 to the teaching evaluation based on deep neural network study score data pair described in 6 any one
Method, the course teaching evaluation result having the same in obtained course evaluation result, in identical course evaluation grouping.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910319686.3A CN110046667B (en) | 2019-04-19 | 2019-04-19 | Teaching evaluation method based on deep neural network learning scoring data pair |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910319686.3A CN110046667B (en) | 2019-04-19 | 2019-04-19 | Teaching evaluation method based on deep neural network learning scoring data pair |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110046667A true CN110046667A (en) | 2019-07-23 |
CN110046667B CN110046667B (en) | 2022-08-12 |
Family
ID=67278153
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910319686.3A Active CN110046667B (en) | 2019-04-19 | 2019-04-19 | Teaching evaluation method based on deep neural network learning scoring data pair |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110046667B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110990583A (en) * | 2019-11-25 | 2020-04-10 | 中山大学 | Course map construction method based on network embedding |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001080153A1 (en) * | 2000-04-15 | 2001-10-25 | Kim Sang Su | Method and system for network-based education and training capable of evaluating learner"s faithfulness and computer-readable medium recording its program |
US20130322741A1 (en) * | 2012-06-05 | 2013-12-05 | DRVision Technologies LLC. | Teachable pattern scoring method |
CN106023015A (en) * | 2016-05-18 | 2016-10-12 | 腾讯科技(深圳)有限公司 | Course learning path recommending method and device |
CN106203811A (en) * | 2016-07-05 | 2016-12-07 | 上海电力学院 | A kind of Evaluation Method of Teaching Quality |
CN107180568A (en) * | 2017-06-30 | 2017-09-19 | 楼志刚 | A kind of Kernel-based methods control and the interactive learning platform and method of evaluation of result |
CN107590760A (en) * | 2017-11-01 | 2018-01-16 | 长春理工大学 | A kind of Teaching quality evaluation system based on big data |
CN108256102A (en) * | 2018-02-01 | 2018-07-06 | 厦门大学嘉庚学院 | A kind of Independent College Studentss based on cluster comment religion data analysing method |
CN109255028A (en) * | 2018-08-28 | 2019-01-22 | 西安交通大学 | Quality of instruction integrated evaluating method based on teaching evaluation data reliability |
CN109360458A (en) * | 2018-10-25 | 2019-02-19 | 重庆鲁班机器人技术研究院有限公司 | Interest assistant teaching method, device and robot |
CN109597937A (en) * | 2018-12-03 | 2019-04-09 | 华中师范大学 | Network courses recommended method and device |
-
2019
- 2019-04-19 CN CN201910319686.3A patent/CN110046667B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2001080153A1 (en) * | 2000-04-15 | 2001-10-25 | Kim Sang Su | Method and system for network-based education and training capable of evaluating learner"s faithfulness and computer-readable medium recording its program |
US20130322741A1 (en) * | 2012-06-05 | 2013-12-05 | DRVision Technologies LLC. | Teachable pattern scoring method |
CN106023015A (en) * | 2016-05-18 | 2016-10-12 | 腾讯科技(深圳)有限公司 | Course learning path recommending method and device |
CN106203811A (en) * | 2016-07-05 | 2016-12-07 | 上海电力学院 | A kind of Evaluation Method of Teaching Quality |
CN107180568A (en) * | 2017-06-30 | 2017-09-19 | 楼志刚 | A kind of Kernel-based methods control and the interactive learning platform and method of evaluation of result |
CN107590760A (en) * | 2017-11-01 | 2018-01-16 | 长春理工大学 | A kind of Teaching quality evaluation system based on big data |
CN108256102A (en) * | 2018-02-01 | 2018-07-06 | 厦门大学嘉庚学院 | A kind of Independent College Studentss based on cluster comment religion data analysing method |
CN109255028A (en) * | 2018-08-28 | 2019-01-22 | 西安交通大学 | Quality of instruction integrated evaluating method based on teaching evaluation data reliability |
CN109360458A (en) * | 2018-10-25 | 2019-02-19 | 重庆鲁班机器人技术研究院有限公司 | Interest assistant teaching method, device and robot |
CN109597937A (en) * | 2018-12-03 | 2019-04-09 | 华中师范大学 | Network courses recommended method and device |
Non-Patent Citations (8)
Title |
---|
HU ZHIJUN等: ""Teaching Evaluation and Diagnosis Based on BP Neural Network"", 《2009 INTERNATIONAL CONFERENCE ON MANAGEMENT AND SERVICE SCIENCE》 * |
PINGPING GU等: ""cluster analysis on the teaching evaluation data from college students"", 《2018 13TH INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE & EDUCATION》 * |
YANG H等: ""A Kind of Analytic Hierarchy Process and Neural Network Integrated Evaluation Method of Teaching"", 《2016 INTERNATIONAL CONFERENCE ON HUMANITY,EDUCATION AND SOCIAL SCIENCE》 * |
施蕙: ""课程建设质量评价体系的神经网络模型"", 《江苏技术师范学院学报》 * |
晏榆洋: ""基于秩和检验及非参数Bootstrap法的多指标统计综合评价模型"", 《四川文理学院学报》 * |
王佳利: ""大数据视角下高校网络教学平台应用的实践与反思"", 《中国现代教育装备》 * |
黄晓辉等: ""基于混合高斯模型融合历史评价信息的高校教学评价方法"", 《亚太教育》 * |
黄晓辉等: ""大数据环境下基于多维信息融合的高校教学评价"", 《亚太教育》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110990583A (en) * | 2019-11-25 | 2020-04-10 | 中山大学 | Course map construction method based on network embedding |
Also Published As
Publication number | Publication date |
---|---|
CN110046667B (en) | 2022-08-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Balschweid et al. | Integrating science in agricultural education: Attitudes of Indiana agricultural science and business teachers | |
CN107230174A (en) | A kind of network online interaction learning system and method | |
AU2020103529A4 (en) | An educational big data analysis method based on artificial intelligence | |
CN109858797A (en) | The various dimensions information analysis of the students method of knowledge based network exact on-line education system | |
CN108776942A (en) | A kind of method and system of the class's of walking elective course | |
CN103400160A (en) | Zero training sample behavior identification method | |
CN110232343A (en) | Children personalized behavioral statistics analysis system and method based on latent variable model | |
CN107392810A (en) | A kind of microteaching integrated evaluating method based on index weights study and behavior entropy | |
CN110046667A (en) | A kind of Method of Teaching Appraisal based on deep neural network study score data pair | |
CN115438152B (en) | Simple answer scoring method and system based on multi-neural network and knowledge graph | |
CN110189236A (en) | Alarming system method based on big data | |
CN113642880A (en) | Internet-based team training method and system | |
Glass et al. | Learning to produce knowledge: Reconstructing teacher preparation for urban schools | |
CN113935869A (en) | Student subjective and objective factor combined score prediction method and system | |
CN113256466A (en) | Smart campus construction sports teaching management method and system based on artificial intelligence and computer storage medium | |
Kong et al. | [Retracted] Construction and Application of Physical Education Teacher Literacy Evaluation System Based on Grey Relational Model | |
Chang | Design and Application of Evaluation Method for Civics Classroom Based on CRITIC Fuzzy Algorithm | |
Jha | A Systematic Study on Student Performance Prediction from the Perspective of Machine Learning and Data Mining Approaches | |
Huang et al. | An adaptive cellular genetic algorithm based on selection strategy for test sheet generation | |
Siemon | Teaching with the Big Ideas in Mathematics | |
Osah et al. | Spatial distribution pattern of lower basic schools for the implementation of lower basic education in Rivers State, Nigeria | |
Aldikanji et al. | Studying Academic Indicators within Virtual Learning Environment Using Educational Data Mining | |
Wu | A Study on Data-Driven Cluster Analysis of Teaching Quality in Civic and Political Education in Colleges and Universities | |
Nami et al. | Comparison of the ethical status of public and non-profit primary school students | |
Yang et al. | Analysis on Quality Evaluation of Animation Majors' Works from Perspective of DANP model |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |