CN110533992A - A kind of learning effect prediction technique and system - Google Patents
A kind of learning effect prediction technique and system Download PDFInfo
- Publication number
- CN110533992A CN110533992A CN201910822061.9A CN201910822061A CN110533992A CN 110533992 A CN110533992 A CN 110533992A CN 201910822061 A CN201910822061 A CN 201910822061A CN 110533992 A CN110533992 A CN 110533992A
- Authority
- CN
- China
- Prior art keywords
- learning
- history
- learning state
- learner
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G09—EDUCATION; CRYPTOGRAPHY; DISPLAY; ADVERTISING; SEALS
- G09B—EDUCATIONAL OR DEMONSTRATION APPLIANCES; APPLIANCES FOR TEACHING, OR COMMUNICATING WITH, THE BLIND, DEAF OR MUTE; MODELS; PLANETARIA; GLOBES; MAPS; DIAGRAMS
- G09B19/00—Teaching not covered by other main groups of this subclass
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The embodiment of the present application provides a kind of learning effect prediction technique and system, is related to IT application in education sector technical field.This method comprises: obtaining the expression data of learner's current generation and parsing learning state and the learning state record total time of learner's current generation according to the expression data of current generation, the classification of learning state includes active learning state and negative learning state；According to the active learning state parameter for obtaining the learning state of current generation and learning state record total time the current generation；Using the learning effect of the active learning state parameter of current generation and preset learning effect prediction model prediction next stage, and calculate the study idea index of history learning power index and current generation；Solve that existing learning effect prediction technique subjectivity is strong, the problem of prediction result inaccuracy.
Description
Technical field
This application involves IT application in education sector technical fields, in particular to a kind of learning effect prediction technique and system.
Background technique
What existing learning effect prediction data source was concentrated mainly on learning management system record includes amount of reading, comment
Attitude towards study class data and school grade assessment data that the behavioral datas such as number, study duration and questionnaire survey obtain etc..
Although these data are able to reflect learning state to a certain extent, external learning behavior and study are still fallen within from type
The backtracking class data of attitude, and data subjectivity is too strong, so that reducing to following study at the confidence level for being prediction, can not make
Learner timely regularized learning algorithm strategy and learning Content in learning process.
Summary of the invention
The embodiment of the present application is designed to provide a kind of learning effect prediction technique and system, solves existing study
Effect prediction technique subjectivity is strong, the problem of prediction result inaccuracy.
The embodiment of the present application provides a kind of learning effect prediction technique, this method comprises:
It obtains the history expression data of learner's current generation and is parsed according to the history expression data of current generation and learnt
The learning state and learning state of person records total time, and the classification of the learning state includes active learning state and negative learning shape
State；
According to the active learning state for obtaining the learning state of current generation and learning state record total time the current generation
Parameter；
Utilize the active learning state parameter of current generation and preset learning effect prediction model prediction next stage
Practise effect.
During above-mentioned realization, current generation facial expression data of the learner on classroom is parsed, is identified
Active learning state and negative learning state of the learner in learning process out；According to active learning state parameter i.e. entire
Learning state records the accounting in total time, can reflect the history learning effect of learner's current generation, actively by these
Learning state parameter inputs the learning effect of preset learning effect prediction model i.e. predictable next stage.With history expression data
The history learning effect reflected is foundation, inputs preset learning effect prediction model, predicts learner's next stage
Effect is practised, the objectivity and accuracy of learning effect prediction are improved, to solve existing learning effect prediction technique master
Strong, the problem of prediction result inaccuracy of the property seen.
Further, the learning state and learning state of learner's current generation are parsed according to the expression data of current generation
Record total time, comprising:
The type of active learning state and continuing for every kind of active learning state are obtained according to the expression data of current generation
Time；
The type of negative learning state and continuing for every kind of negative learning state are obtained according to the expression data of current generation
Time；
Positive is calculated according to the duration of the type of the active learning state and every kind of active learning state
Practise state for time；
When calculating negative learning state according to the duration of the type of negative learning state and every kind of negative learning state
Between；
Study state recording total time is calculated according to active learning state for time and negative learning state for time.
During above-mentioned realization, according to expression data obtain active learning state type such as think deeply and understand and
The duration of every kind of active learning state, then active learning state for time is the total of the duration of every kind of active learning state
With；The type for obtaining negative learning state according to expression data is such as puzzled and bored and every kind of negative learning state continues
Time, then negative learning state for time is the summation of the duration of every kind of negative learning state；Learning state records total time
It is then the sum of active learning state for time and the negative learning state for time；Expression data comprehensively, objectively parse,
Be conducive to accurately obtain learning state of emotional change of the learner in learning process to embody learner.
Further, active learning state parameter includes active learning state accounting and active learning state ratio；Wherein,
Active learning state accounting is the ratio that active learning state for time and learning state record total time；Active learning state ratio
To be recorded in total time in learning state, the ratio of active learning state for time and negative learning state for time；
This method further includes the history learning power index that learner is calculated according to history learning achievement data, according to history
Practise the history learning power index that achievement data calculates learner, comprising:
The total number of persons of class where obtaining learner, in the class of place everyone history learning achievement data and everyone
History active learning state accounting；
Everyone history total performance is calculated according to everyone history learning achievement data；
Everyone the history total performance is normalized, obtain everyone normalization history it is comprehensive at
Achievement；
According to it is described everyone history active learning state accounting and the learner where class total number of persons, calculate
Class's history active learning state accounting average value；
According to it is described everyone normalization history total performance and the learner where class total number of persons, calculate class
History total performance average value；
The history active learning state accounting of learner and class's history active learning state accounting average value are compared
Compared with, the normalization history total performance of learner and class's history total performance average value are compared, comparison result is obtained,
History learning power index is determined according to comparison result.
During above-mentioned realization, everyone history learning achievement data includes that a certain course is of all categories in the class of place
History learning achievement and corresponding history learning achievement weight, the classification of the course such as performance of the test, interim achievement after class
With final grade etc.；According to everyone history is comprehensive in everyone available class of history learning achievement data in class
Achievement；Everyone normalization history total performance, class's history active learning state accounting average value and class are calculated again
Grade history total performance average value；By the history active learning state accounting of learner and class's history active learning state accounting
Average value is compared, and the normalization history total performance of learner and class's history total performance average value are compared,
I.e. the history active learning state accounting of comparative learning person, normalization history total performance are compared with class average value, really
The history learning power index for determining learner, for assessing the study idea of learner.
Further, it is predicted in the active learning state parameter and preset learning effect prediction model for utilizing the current generation
Before the step of learning effect of next stage, this method further includes constructing preset learning effect prediction model；It constructs preset
Learning effect prediction model, comprising:
Construct preset hundred-mark system learning effect prediction model；
Construct preset social estate system learning effect prediction model.
During above-mentioned realization, it is built with hundred-mark system learning effect prediction model and social estate system learning effect prediction mould
Type can from different angles predict the learning effect of learner, increase the alternative and suitable of prediction learning effect
With property, and improve the accuracy of learning effect prediction.
Further, preset hundred-mark system learning effect prediction model is constructed, comprising:
Obtain learner's history active learning state accounting and normalization history total performance；
According to learner's history active learning state accounting, normalization history total performance and the first regression equation, calculate
First regression coefficient of the first regression equation；
Hundred-mark system learning effect prediction model is obtained according to the first regression equation and the first regression coefficient.
During above-mentioned realization, using regression equation as prediction model according to history learning state accounting and next stage
Dependence between learning effect predicts next stage learning effect, so that more objective to the prediction of the learning effect of next stage
It sees.Using learner's history active learning state accounting as input parameter, normalization history total performance is defeated as output parameter
Enter the first regression equation calculation and go out corresponding first regression coefficient of the first regression equation, the first regression equation can be obtained, inputs
The history active learning state accounting of current generation can be obtained the learning effect prediction result of next stage, hundred-mark system study at
It imitates prediction model building simply, conveniently, and can preferably predict the learning effect of next stage.
Further, preset hundred-mark system learning effect prediction model is constructed, comprising:
Obtain the history active learning state ratio and normalization history total performance of learner；
According to the history active learning state ratio of learner, normalization history total performance and the second regression equation, meter
Calculate the second regression coefficient of second regression equation；
Hundred-mark system learning effect prediction model is obtained according to the second regression equation and the second regression coefficient.
During above-mentioned realization, using the history active learning state accounting of learner as input parameter, normalization is gone through
History total performance inputs the second regression equation calculation as output parameter and goes out corresponding second regression coefficient of the second regression equation, i.e.,
The learning effect of next stage can be obtained in available second regression equation, the history active learning state ratio for inputting the current generation
Prediction result, hundred-mark system learning effect prediction model building simply, conveniently, and can preferably predict the study of next stage
Effect.
Further, preset social estate system learning effect prediction model is constructed, comprising:
Establish the corresponding rank list of learning effect；
The social estate system history learning achievement normalization history total performance of learner converted in rank list；
It is patrolled according to the positive state accounting of the history of learner, social estate system history learning achievement and logistic regression equation, calculating
Collect the logistic regression coefficient of regression equation；
Social estate system learning effect prediction model is obtained according to logistic regression equation and logistic regression coefficient.
During above-mentioned realization, using the positive state accounting of the history of learner as input parameter, social estate system history
Achievement is practised as output parameter input logic regression equation, the estimation of logistic regression coefficient is carried out using maximum likelihood estimate, is adopted
The optimization that regression coefficient is carried out with gradient descent method is calculated logistic regression coefficient, logistic regression equation can be obtained, and will work as
The positive state accounting input logic regression equation of the history of last stage carries out the learning effect prediction result that the next stage is calculated.
The embodiment of the present application also provides a kind of learning effect forecasting system, which includes:
Expression data receives and parsing module, for obtaining the history expression data of learner's current generation and according to described
The learning state and learning state of history expression data parsing learner's current generation of current generation records total time；
Learning state processing module records total time according to the learning state and learning state of the current generation for root
Obtain active learning state parameter；
Learning effect prediction module, for using the current generation active learning state parameter and it is preset study at
Imitate the learning effect of prediction model prediction next stage.
During above-mentioned realization, expression data receives and the history expression number of parsing module parsing learner's current generation
According to the learning state and learning state for obtaining learner's current generation record total time；It is obtained using learning state processing module
Learning effect prediction module carries out active learning state parameter required when learning effect prediction, facilitates and is predicted using learning effect
The prediction of module progress learning effect.
Further, which further includes prediction model building module, and prediction model building module includes:
Hundred-mark system learning effect prediction model constructs module, for constructing preset hundred-mark system learning effect prediction model；
Social estate system learning effect prediction model constructs module, for constructing preset social estate system learning effect prediction model.
During above-mentioned realization, module is constructed by hundred-mark system learning effect prediction model and social estate system learning effect is pre-
The corresponding building hundred-mark system learning effect prediction model of model construction module and social estate system learning effect prediction model are surveyed, it can
To select different models to carry out learning effect prediction as needed, the convenience used is improved.
The embodiment of the present application also provides a kind of electronic equipment, and electronic equipment includes memory and processor, the memory
For storing computer program, processor runs computer program so that computer equipment executes the study in the embodiment of the present application
Effect prediction technique.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application will make below to required in the embodiment of the present application
Attached drawing is briefly described, it should be understood that the following drawings illustrates only some embodiments of the application, therefore should not be seen
Work is the restriction to range, for those of ordinary skill in the art, without creative efforts, can be with
Other relevant attached drawings are obtained according to these attached drawings.
Fig. 1 is a kind of flow chart of learning effect prediction technique provided by the embodiments of the present application；
Fig. 2 is the history learning power index provided by the embodiments of the present application that learner is calculated according to history learning achievement data
Flow diagram；
Fig. 3 is a kind of process signal for constructing preset hundred-mark system learning effect prediction model provided by the embodiments of the present application
Figure；
Fig. 4 is that another process for constructing preset hundred-mark system learning effect prediction model provided by the embodiments of the present application is shown
It is intended to；
Fig. 5 is a kind of process signal for constructing preset social estate system learning effect prediction model provided by the embodiments of the present application
Figure；
Fig. 6 is a kind of flow chart of the study idea index calculation method of current generation provided by the embodiments of the present application；
Fig. 7 is a kind of structural block diagram of learning effect forecasting system provided by the embodiments of the present application；
Fig. 8 is the structural block diagram of study idea index computing module provided by the embodiments of the present application；
Fig. 9 is the structural block diagram that a kind of hundred-mark system learning effect prediction model provided by the embodiments of the present application constructs module；
Figure 10 is the structural frames that another hundred-mark system learning effect prediction model provided by the embodiments of the present application constructs module
Figure；
Figure 11 is the structural block diagram that a kind of social estate system learning effect prediction model provided by the embodiments of the present application constructs module.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, technical solutions in the embodiments of the present application is described.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.Meanwhile the application's
In description, term " first ", " second " etc. are only used for distinguishing description, are not understood to indicate or imply relative importance.
Embodiment 1
With the deep development of mood and Cognitive Study, compared to for over behavior data, emotion index more meets study
The learning performance of person.In learning science field, emotional state is usually divided into two dimensions, i.e. the positive and negative mood of reflection emotion refers to
The arousal of number and description emotion excitation degree.Main carriers of the facial expression as emotional information, can be pushed away by facial expression
Emotional state in other people locating learning processes of breaking, to infer its learning effect.
Fig. 1 is please referred to, Fig. 1 is a kind of flow chart of learning effect prediction technique provided by the embodiments of the present application；It is exemplary,
This method can using computer equipment as carrier realize, method includes the following steps:
Step S100: obtaining the expression data of learner's current generation and parses study according to the expression data of current generation
The learning state and learning state of person's current generation records total time, and the classification of the learning state includes active learning state and disappears
Pole learning state.
Specifically, the learning state and learning state of learner are parsed in step S100 according to the expression data of current generation
Total time is recorded, can specifically include following steps:
The type of active learning state and continuing for every kind of active learning state are obtained according to the expression data of current generation
Time；
Exemplary, learning state is indicated with set S, and learning state includes active learning state and negative learning state, product
Pole learning state is expressed as SpWith negative learning state Sd, then S={ Sp, Sd}。
Wherein, active learning state includes h kind, is expressed as Sp={ Sp1, Sp2..., Sph}；Such as it can be absorbed, thinking
With understand etc. positive learning state.
It is exemplary, obtain the duration of every kind of active learning state, and by every kind of active learning state it is lasting when
Between be expressed as Δ ti(i=1 ..., h).
The type of negative learning state and continuing for every kind of negative learning state are obtained according to the expression data of current generation
Time；
Negative learning state includes q kind, is expressed as Sd={ Sd1, Sd2..., Sdq}；Such as can be for puzzlement, be sick of and
It vacillates and waits negative learnings state.
It is exemplary, obtain the duration of every kind of negative learning state, and by every kind of negative learning state it is lasting when
Between be expressed as Δ tj(j=1 ..., q).
When calculating active learning state according to the duration of the type of active learning state and every kind of active learning state
Between；
Exemplary, active learning state for time is expressed as Tsp, then
When calculating negative learning state according to the duration of the type of negative learning state and every kind of negative learning state
Between；
Exemplary, negative learning state for time is expressed as Tsd, then
Study state recording total time is calculated according to active learning state for time and the negative learning state for time.
Exemplary, learning state record General Schedule is shown as T, then T=Tsp+Tsd。
Expression data comprehensively, objectively parse, is conducive to accurately obtain mood of the learner in learning process
Variation is to embody the learning state of learner；And to calculate the history learning power index of learner and carrying out learning effect
Prediction is laid a good foundation.
Step S200: the positive of current generation is obtained according to the learning state of current generation and learning state record total time
Learning state parameter；
Specifically, active learning state parameter includes active learning state accounting and active learning state ratio；Wherein, product
Pole learning state accounting is the ratio that active learning state for time and learning state record total time, and active learning state ratio is
Within learning state record total time, the ratio of active learning state for time and negative learning state for time.
This method further includes the history learning power index that learner is calculated according to history learning achievement data, as shown in Fig. 2,
Fig. 2 is that the process of the history learning power index provided by the embodiments of the present application that learner is calculated according to history learning achievement data is shown
It is intended to, the history learning power index of learner is calculated according to history learning achievement data, comprising:
Step S211: the total number of persons of class where obtaining learner, everyone history learning achievement number in the class of place
According to the history active learning state accounting with everyone；
Exemplary, the total number of persons of class where the learner of acquisition is N.Everyone history learning achievement in the class of place
Data include performance of the test after class, unit performance of the test, interim achievement and the final grade totally 4 class achievement of any course m；Its
In, the other history learning achievement value of any sort can be expressed as GHi。
It is exemplary, if history learning achievement is using the representation method of social estate system or point system, need grade
System or the history learning achievement of point system are converted to the history learning achievement of hundred-mark system, specific method for transformation achievement as described below
Convert list:
As GH in social estate systemi=A+Or GH in point systemiWhen=5, then there is GH in hundred-mark systemi=100；
As GH in social estate systemiGH in=A or point systemi∈ [4.0,5.0) when, then there is GH in hundred-mark systemi=90；
As GH in social estate systemi=A-Or GH in point systemiWhen=3.7, then there is GH in hundred-mark systemi=85；
As GH in social estate systemi=B+Or GH in point systemiWhen=3.3, then there is GH in hundred-mark systemi=82；
As GH in social estate systemiGH in=B or point systemiWhen=3.0, then there is GH in hundred-mark systemi=78；
As GH in social estate systemi=B-Or GH in point systemiWhen=2.7, then there is GH in hundred-mark systemi=75；
As GH in social estate systemi=C+Or GH in point systemiWhen=2.3, then there is GH in hundred-mark systemi=72；
As GH in social estate systemiGH in=C or point systemiWhen=2.0, then there is GH in hundred-mark systemi=68；
As GH in social estate systemi=C-Or GH in point systemiWhen=1.5, then there is GH in hundred-mark systemi=64；
As GH in social estate systemiGH in=D or point systemiWhen [1.0,1.3] ∈, then there is GH in hundred-mark systemi=60；
As GH in social estate systemiGH in=F or point systemi=[0,1.0) when, then there is GH in hundred-mark systemi=0.
Calculate everyone history active learning state accounting, exemplary, the history active learning state of any learner k
Accounting can indicate are as follows: Sp(k)m=Tsp/T；
Step S212: everyone history total performance is calculated according to everyone history learning achievement data；
It is exemplary, everyone history total performance is expressed as GH (k)m, thenIts
In, n is history learning achievement total degree of all categories, wjIt is any learner k in course jth time history learning achievement weight,
GHiIt (j) is the jth of course the i-th class achievement time school grade value.
Step S213: being normalized everyone history total performance, obtains everyone normalization history
Total performance；
Exemplary, the history total performance minimum value of any learner k is expressed as MinGH (k)m, history total performance is most
Big value is expressed as MaxGH (k)m, the normalization history total performance of any learner k is expressed as GH (k)m1, thenGH(k)m1∈ [0,100].
Step S214: according to total people of class where everyone history active learning state accounting and the learner
Number calculates class's history active learning state accounting average value；
Class's history active learning state accounting average value is expressed asThen
Step S215: according to class's total number of persons, calculation squad where everyone normalization history total performance and learner
Grade history total performance average value；
Exemplary, class's history total performance average value is expressed asThen
Step S216: the history active learning state accounting of learner and class's history active learning state accounting are averaged
Value is compared, and the normalization history total performance of the learner and class's history total performance average value are compared
Compared with acquisition comparison result determines history learning power index according to comparison result.
Exemplary, study idea index includes that the grade of study idea index is expressed as by 4 levels according to comparison result
Determining study idea index is as follows:
WhenAndWhen, then history learning power is insufficient, is
WhenAndWhen, then history learning power is relatively low, is
WhenAndWhen, then history learning power charge less is sufficient, is
WhenAndWhen, then history learning power is sufficient, is
During above-mentioned realization, by the history active learning state accounting of learner and class's history active learning state
Accounting average value is compared, and the normalization history total performance of the learner and class's history total performance are averaged
Value is compared, and can learn level of the history learning state of learner in class, and using history learning power index come
The size of representative learning person's study idea.
History learning power index is for objectively responding level of effort, willpower size and ability of the learner in learning process
Level, therefore the study idea that history learning power index reflection learner's current generation can be used is horizontal.
Step S300: using under the active learning state parameter of current generation and the prediction of preset learning effect prediction model
The learning effect in stage.
It is pre- using the active learning state parameter of current generation and preset learning effect prediction model in step S300
Before the step of surveying the learning effect of next stage, this method further includes constructing preset learning effect prediction model；Building is default
Learning effect prediction model, comprising:
Step S400: preset hundred-mark system learning effect prediction model is constructed；
Step S500: preset social estate system learning effect prediction model is constructed.
Wherein, the hundred-mark system learning effect of any course m of any learner k can be expressed as A1(k)m, social estate system study
Effect can be expressed as A2(k)m。
During above-mentioned realization, mould is predicted by building hundred-mark system learning effect prediction model and social estate system learning effect
Type can from different angles predict the learning effect of learner, increase the alternative and suitable of prediction learning effect
With property, and improve the accuracy of learning effect prediction.
As one of embodiment, as shown in figure 3, Fig. 3 is a kind of building preset hundred provided by the embodiments of the present application
Divide the flow diagram of learning effect prediction model processed, constructing preset hundred-mark system learning effect prediction model includes following step
It is rapid:
Step S410: learner's history active learning state accounting and normalization history total performance are obtained；
It is exemplary, it can be seen from the above, any learner k is in TnHistory active learning state accounting in period is Sp
(k)m, normalization history total performance is GH (k)m1。
Step S411: according to the history active learning state accounting of learner, normalization history total performance and first time
Return equation, calculates the first regression coefficient of the first regression equation；
Exemplary, using history active learning state accounting as input parameter, normalization history total performance is as output
Parameter substitutes into the first regression equation, i.e. A1(k)m=a1*Sp(k)m+a2*Sp(k)m 2+a3+ε1；Wherein, A1(k)m=GH (k)m1, warp
A1, the value of a2, a3 can be acquired by crossing calculating；Wherein, a1, a2, a3 are the regression coefficient of the learning effect prediction model, 1 table of ε
Show random error.
Step S412: hundred-mark system learning effect prediction model is obtained according to the first regression equation and the first regression coefficient.
The value of the a1 that will be acquired, a2, a3 substitute into above-mentioned first regression equation, and hundred-mark system learning effect prediction mould can be obtained
Type, A1(k)m=a1*x+a2*x2+a3+ε1；Wherein, x indicates independent variable, indicates that history active learning state accounts in this model
Than.
When being predicted using learning effect of this model to a certain learner's next stage, by current slot Tn+1Interior
The history active learning state accounting of the learner is input in the model, obtained A1(k)mThe next stage as predicted
Practise effect.
As another embodiment, as shown in figure 4, Fig. 4 is another building preset hundred provided by the embodiments of the present application
Divide the flow diagram of learning effect prediction model processed, constructing preset hundred-mark system learning effect prediction model includes following step
It is rapid:
Step S420: the history active learning state ratio and normalization history total performance of learner are obtained；
It is exemplary, it can be seen from the above, obtaining any learner k in TnHistory active learning state ratio in period and
It normalizes history total performance GH (k)m1。
History active learning state ratio can be expressed as R (k)m, R (k)m=Sp(k)m/Sd(k)m；Sd(k)mIndicate any
The history active learning state accounting of learner k, i.e. Sd(k)m=Tsd/T。
Step S421: according to the history active learning state ratio of learner, normalization history total performance and second time
Return equation, calculates the second regression coefficient of second regression equation；
Exemplary, using history active learning state ratio as input parameter, normalization history total performance is as output
Parameter substitutes into the second regression equation, i.e. A1(k)m=b1*R (k)m+b2*R(k)m 2+b3+ε2；Wherein, A1(k)m=GH (k)m1, warp
B1, the value of b2, b3 can be acquired by crossing calculating；Wherein, b1, b2, b3 are the regression coefficient of the learning effect prediction model, and ε 2 is indicated
Random error.
Step S422: hundred-mark system learning effect prediction model is obtained according to the second regression equation and the second regression coefficient.
The value of the b1 that will be acquired, b2, b3 substitute into above-mentioned first regression equation, and hundred-mark system learning effect prediction mould can be obtained
Type, A1(k)m=b1*x+b2*x2+b3+ε2；Wherein, x indicates independent variable, and history active learning state ratio is indicated in this model
Rate.
When being predicted using learning effect of this model to a certain learner's next stage, by current slot Tn+1Interior
The history active learning state ratio of the learner is input in the model, obtained A1(k)mThe next stage as predicted
Practise effect.
As shown in figure 5, Fig. 5 is a kind of preset social estate system learning effect prediction model of building provided by the embodiments of the present application
Flow diagram, the preset social estate system learning effect prediction model of building in step S500, specifically includes the following steps:
Step S501: the corresponding rank list of school grade is established；
It is exemplary, the hundred-mark system in school grade is converted to social estate system, for example above-mentioned achievement of specific method for transformation turns
Change list, details are not described herein.
Step S502: by the normalization history total performance of learner be converted into the social estate system in rank list study at
Effect, wherein social estate system learning effect shares t grade, and j indicates any grade.
Specifically, by the normalization history total performance GH (k) of any learnerm1Conversion list on merit is converted to pair
The social estate system history learning achievement answered, social estate system history learning achievement are expressed as A2j。
Step S503: according to the positive state accounting of the history of learner, social estate system history learning achievement and logistic regression side
Journey, the logistic regression coefficient of calculating logic regression equation；
It is exemplary, by the positive state accounting S of the history of any learner kp(k)mWith corresponding social estate system history learning achievement
A2jInput logic regression equation.
Wherein, the logistic regression equation that any grade j occurs indicates are as follows: Logit (pj)=αj+ β x, wherein pjIndicate etc.
Cumulative probability before grade learning effect processed takes when j value, is embodied as:
Wherein, αjIt is the logistic regression coefficient of logistic regression equation with β, by A2jAnd Sp(k)mIt substitutes into logistic regression equation
A2(k)mAnd x, the estimation of logistic regression coefficient is carried out using maximum likelihood estimate, regression coefficient is carried out using gradient descent method
Optimization；Social estate system learning effect is successively pressed into different value horizontal segmentations into two classes, multiple two points of realization social estate systems study
The foundation of effect prediction model.
Step S504: social estate system learning effect prediction model is obtained according to logistic regression equation and logistic regression coefficient.
The logistic regression coefficient of the logistic regression equation acquired is substituted into above-mentioned logistic regression equation, social estate system can be obtained
Learning effect prediction model；Wherein, x indicates independent variable, and history active learning state accounting S is indicated in this modelp(k)m。
When being predicted using learning effect of this model to a certain learner's next stage, by current slot Tn+1Interior
The history active learning state accounting of the learner is input in the model, obtained A2(k)mThe next stage as predicted
Practise effect.
Optionally, this method can also include the study idea index and learning effect prediction result that display obtains, and specifically may be used
To include:
It is according to the time dimension for dividing learning effect and study idea index, by learning effect and with learning time length
Power index is practised to carry out horizontal and vertical relatively and showing according to short-term, mid-term, for a long time；
The variation tendency of the different phase learnt according to learning effect and study idea index two indices in learner and
Distribution situation with class where learner, provides recommendation service for learner.
By parsing to current generation facial expression data of the learner on classroom, identify that learner is learning
Active learning state and negative learning state in the process；It is i.e. total in entire learning state record according to active learning state parameter
Accounting in time can reflect the history learning effect of learner's current generation, these active learning state parameters are defeated
Enter the learning effect of preset learning effect prediction model i.e. predictable next stage.The history reflected with history expression data
Habit effect is foundation, inputs preset learning effect prediction model, predicts the learning effect of learner's next stage, improves
The objectivity and accuracy of effect prediction are practised, so that it is strong to solve existing learning effect prediction technique subjectivity, prediction result
The problem of inaccuracy.
The application obtains the study idea index of learner by the way that the facial expression of learner is resolved to a variety of learning states
With history learning effect, the learning effect of next stage can be predicted, facilitate the timely regularized learning algorithm strategy of learner, promote study effect
Fruit.
This method further includes the learning effect prediction result meter of the learning state and next stage according to learner's current generation
The study idea index of current generation is calculated, as shown in fig. 6, being a kind of study idea index of current generation provided by the embodiments of the present application
The flow chart of calculation method.It can specifically include following steps:
Step S601: the active learning state accounting S of the current generation of any learner k is calculatedp(k)m1And class works as
The active learning state accounting average value of last stage is expressed as
Step S602: the learning effect prediction result A of the learner is obtainedi(k)m, and calculate class's study of place class
Effect predicted mean voteWherein i=1 or 2, A1(k)mIndicate hundred-mark system learning effect, A2(k)mIndicate social estate system study
Effect, circular is as described in above-mentioned steps S400 and step S500, and details are not described herein.
Step S603: the study idea index of current generation is determined.
By the active learning state accounting of the active learning state accounting of the current generation of learner and class's current generation
Average value is compared；The learning effect prediction result of the learner and class's learning effect predicted mean vote are compared
Compared with acquisition comparison result determines the study idea index of current generation according to comparison result.
According to comparison result, the grade of the study idea index of current generation is expressed asDetermining current generation
It is as follows to practise power index:
WhenAndWhen, then the study idea of current generation is insufficient, is inWater
It is flat；
WhenAndWhen, then the study idea of current generation is relatively low, is inWater
It is flat；
WhenAndWhen, then the study idea charge less foot of current generation, is in
It is horizontal；
WhenAndWhen, then the study idea of current generation is sufficient, is inWater
It is flat.
Embodiment 2
The embodiment of the present application also provides a kind of learning effect forecasting system, as shown in fig. 7, Fig. 7 mentions for the embodiment of the present application
A kind of structural block diagram of the learning effect forecasting system supplied, the system include:
Expression data receives and parsing module 600, for obtaining the history expression data and basis of learner's current generation
The history expression data of the current generation parses the learning state of learner's current generation and when learning state records total
Between；
Learning state processing module 700 records always for root according to the learning state and learning state of the current generation
Time obtains active learning state parameter；
Learning effect prediction module 800, active learning state parameter and preset for the utilization current generation
Practise the learning effect of effect prediction model prediction next stage.
Optionally, the embodiment of the present application further includes study idea index computing module 500.As shown in figure 8, implementing for the application
The structural block diagram for the study idea index computing module 500 that example provides, study idea index computing module 500 include:
Parameter acquisition module 510, for obtaining the total number of persons of class where learner, everyone history in the class of place
The history active learning state accounting of school grade data and everyone；
History total performance computing module 520, for calculating everyone according to everyone history learning achievement data
History total performance；
History total performance computing module 530 is normalized, place is normalized for the history total performance to everyone
Reason, obtains everyone normalization history total performance；
History active learning state accounting mean value calculation module 540, for according to everyone history active learning shape
The total number of persons of state accounting and learner place class, calculates class's history active learning state accounting average value；
Class's history total performance mean value calculation module 550, it is comprehensive for everyone normalization history according to
Achievement and learner place class total number of persons, calculate class's history total performance average value；
History learning power index determining module 560, for going through the history active learning state accounting of learner and class
History active learning state accounting average value is compared, by the normalization history total performance of learner and class's history it is comprehensive at
Achievement average value is compared, and is obtained comparison result, is determined history learning power index according to comparison result.
Optionally, which further includes prediction model building module, and prediction model building module includes:
Hundred-mark system learning effect prediction model constructs module, for constructing preset hundred-mark system learning effect prediction model；
Social estate system learning effect prediction model constructs module 930, for constructing preset social estate system learning effect prediction mould
Type.
The application provides a kind of embodiment of hundred-mark system learning effect prediction model building module, as shown in figure 9, Fig. 9 is
A kind of structural block diagram of hundred-mark system learning effect prediction model building module 500 provided by the embodiments of the present application, is denoted as the 100th
Learning effect prediction model processed is divided to construct module 910, comprising:
First parameter acquisition submodule 911, for obtaining learner's history active learning state accounting and normalization
History total performance；
First regression coefficient computational submodule 912, for according to learner's history active learning state accounting, normalizing
Change history total performance and the first regression equation, calculates the first regression coefficient of the first regression equation；
First hundred-mark system learning effect prediction model constructs submodule 913, for according to the first regression equation and first time
Coefficient is returned to obtain hundred-mark system learning effect prediction model.
Optionally, the application also provides the embodiment of another hundred-mark system learning effect prediction model building module, such as schemes
Shown in 10, Figure 10 is the structural frames that another hundred-mark system learning effect prediction model provided by the embodiments of the present application constructs module
Figure is denoted as the second hundred-mark system learning effect prediction model building module 920, comprising:
Second parameter acquisition submodule 921, history active learning state ratio and normalization for obtaining learner are gone through
History total performance；
Second regression coefficient computational submodule 922, for according to the history active learning state ratio of the learner, return
One changes history total performance and the second regression equation, calculates the second regression coefficient of the second regression equation；
Second hundred-mark system learning effect prediction model constructs submodule 923, for according to second regression equation and the
Two regression coefficients obtain hundred-mark system learning effect prediction model.
Optionally, as shown in figure 11, Figure 11 is a kind of social estate system learning effect prediction model provided by the embodiments of the present application
The structural block diagram of module 930 is constructed, social estate system learning effect prediction model constructs module 930, comprising:
Rank list setting up submodule 931, for establishing the corresponding rank list of school grade；
Achievement converts submodule 932, for converting the normalization history total performance of learner in rank list
Social estate system history learning achievement；
Logistic regression coefficient computational submodule 933, for the positive state accounting of history, the social estate system history according to learner
School grade and logistic regression equation calculate the logistic regression coefficient of the logistic regression equation；
Social estate system learning effect prediction model constructs submodule 934, for according to logistic regression equation and logistic regression system
Number obtains social estate system learning effect prediction model.
The embodiment of the present application also provides a kind of electronic equipment, which includes memory and processor, memory
For storing computer program, processor runs the computer program so that the computer equipment executes the embodiment of the present application
Learning effect prediction technique in 1.
The embodiment of the present application also provides a kind of computer readable storage mediums, are stored with computer program instructions, on
When stating computer program instructions and being read and run by a processor, the learning effect prediction technique in the embodiment of the present application 1 is executed.
Embodiment 3
Present invention also provides an embodiments with above method progress learning effect prediction, specific as follows:
Step 1: parsing expression data:
Calculation processing is carried out via expression data, the active learning state parsed is divided into 3 kinds, Sp={ Sp1, Sp2,
Sp3}；Wherein, Sp1It is absorbed, Sp2For thinking, Sp3To understand；It parses obtained negative learning state and is divided into 3 kinds, Sd={ Sd1,
Sd2, Sd3}；Wherein, Sd1For puzzlement, Sd2It is bored, Sd3To vacillate.
Determine T1Period is 40 minutes；Sp1When 10 minutes a length of, the S of appearancep2When 6 minutes a length of, the S of appearancep3Occur
When it is 10 minutes a length of；Sd1When 6 minutes a length of, the S of appearanced2Do not occur, Sd3Appearance when it is 8 minutes a length of.
According to formula Sp(k)m=Tsp/ T, learner are 0.6 in the active learning state duration accounting of the course；Passiveness is learned
Habit state duration accounting value is 0.4；History active learning state ratio R1(k)m=0.6/0.4=1.5；
Determine T2Period is 40 minutes；Sp1When 20 minutes a length of, the S of appearancep3Appearance when it is 20 minutes a length of, then learn
Active learning state duration accounting value of the person in the period is 0.5, history active learning state ratio R2(k)m=1；
Determine T3Period is 40 minutes；Sp2When 20 minutes a length of, the S of appearancep3Appearance when it is 10 minutes a length of；Sd1Out
10 minutes a length of when existing, then active learning state duration accounting value of the learner in the period is 0.75, history active learning
State ratio R3(k)m=3.
Step 2: history total performance data calculation processing:
The history total performance of Chinese curriculum includes 2 types, tests GH1 and last posttest GH4, unit after class
Achievement weight is 20%, and final grade weight is 80%；Then T1Period, performance of the test was 80 points after class, and 1 time final grade is
90 points.T1Normalization history total performance is expressed as GH (k) in periodm1=88 points.
T2Period, performance of the test was 70 points after class, and 1 final grade is 90 points.Unit achievement weight is 20%, the end of term
Achievement weight is 80%；T2Normalization history total performance is expressed as GH (k) in periodm1=86 points.
T3Period, performance of the test was 90 points after class, and 1 final grade is 90 points.Unit achievement weight is 20%, the end of term
Achievement weight is 80%；Then T3Normalization history total performance is expressed as GH (k) in periodm1=90 points.
Step 3: calculate history learning power index:
T3 period class history active learning state accounting average valueT3 period class history is comprehensive
Achievement average valueIt is 85 points, is worked as according to computation ruleAndWhen, then history
Study idea is sufficient, is
Step 4: establish hundred-mark system learning effect prediction model:
By the history active learning state ratio and normalization history total performance input that multiple periods are in 40 minutes
Regression Equations hundred-mark system learning effect prediction model, prediction model are expressed as A1(k)m=b1*R (k)m+b2*R(k)m 2+b3+
ε2, wherein A1(k)m=GH (k)m1, b1, b2, b3 are the regression coefficient of learning effect prediction model, are carried out using least square method
Regression coefficient is estimated to obtain, b1=7.33, b2=-1.33, b3=80.00.
Step 5: prediction next stage learning effect:
By current slot T4Expression data parsed, Sp3When 10 minutes a length of, the S of appearancep1Appearance when it is a length of
17 minutes, Sd1When 6 minutes a length of, the S of appearanced2Appearance when it is 7 minutes a length of, active learning state duration accounting is calculated
With negative learning state duration accounting, R4(k)m=2, input learning effect prediction model, learning effect predicted value A1(k)m=89
Point.
It is specific as follows present invention also provides the embodiment that one shows operation result:
Learning effect classification is compared: if learning effect predicted value is hundred-mark system, the predicted value of learning effect being converted
For ordered categories.Assuming that being divided into three classifications, then have,
Work as A1(k)m∈ (80,100), it is outstanding for defining learning effect；
Work as A1(k)m∈ [60,80], it is up to standard for defining learning effect；
Work as A1(k)m∈[0,60)；It is not up to standard for defining learning effect.
The longitudinal comparison of history learning power index and current generation study idea index: any learner k exists in any course m
TnThe study idea index of period isIn Tn+1The study idea index of period isFoundationWithHierarchical location distance
It is compared, has:
WhenWhen one level, it is calculated as study idea index and declines；
WhenWhen two levels, it is too fast to be calculated as study idea index decreased；
WhenWhen three levels, it is calculated as study idea index and sharply declines；
WhenWhen one level, it is calculated as study idea index and remains unchanged；
WhenWhen one level, it is calculated as study idea index and rises gently；
WhenWhen a level, it is calculated as study idea index rapid increase；
WhenWhen three levels, it is calculated as study idea index high speed and rises；
It includes: using chart to the single class of learner that short-term study idea index and learning effect prediction, which are compared with display mode,
Study idea index and learning effect of the journey within the semester are shown, and are worked as using histogram or line chart longitudinal comparison learner
Preceding study idea index and history learning power index, using histogram or the current learning effect of line chart longitudinal comparison learner with go through
History learning effect closes chart and different colours using multiple groups and refers to current study idea index and the study idea of class where learner
Number average value is compared；Using different colours by the learning effect etc. of class where the learning effect of ordered categories and learner
Grade average value is compared；
It will include one that it includes: using Visual Chart with display mode that mid-term study idea index and learning effect prediction, which are compared,
The short-term study idea index and learning effect predicted value in term merge displaying, and learner's individual is presented using different colours
Study idea index variation trend；The study idea index variation of class where learner's individual and learner is presented using a variety of charts
Trend；The learning effect variation tendency of learner's individual is presented using different icons；Learner's individual is presented using a variety of charts
With the learning effect change of rank trend of class where learner；
It will include one that it includes: using Visual Chart with display mode that Term Learning power index and learning effect prediction, which are compared,
The short-term and mid-term study idea index and learning effect predicted value of academic year or more merges displaying, is presented using report longer
The study idea index variation trend and learning effect variation tendency of a period of time learner's individual；Using report, color and icon
Mixing present in longer period of time the study idea index variation trend of class where learner's individual and learner and learn at
Imitate variation tendency.
There is provided recommendation service for learner includes:
When the current learning effect predicted mean vote of the current learning effect predicted value of learner≤learner place class, clothes
Business is suggested based on encouraging；
When the current learning effect predicted mean vote of the current learning effect predicted value > learner place class of learner, clothes
Business is suggested based on praising；
When the grade of the current study idea exponential forecasting result of learner≤learner's history learning power index ranking, service
It is recommended that based on encouraging；
When grade > learner's history learning power index ranking of the current study idea exponential forecasting result of learner, service
It is recommended that based on praising.
In several embodiments provided herein, it should be understood that disclosed device and method can also pass through
Other modes are realized.The apparatus embodiments described above are merely exemplary, for example, flow chart and block diagram in attached drawing
Show the device of multiple embodiments according to the application, the architectural framework in the cards of method and computer program product,
Function and operation.In this regard, each box in flowchart or block diagram can represent the one of a module, section or code
Part, a part of the module, section or code, which includes that one or more is for implementing the specified logical function, to be held
Row instruction.It should also be noted that function marked in the box can also be to be different from some implementations as replacement
The sequence marked in attached drawing occurs.For example, two continuous boxes can actually be basically executed in parallel, they are sometimes
It can execute in the opposite order, this depends on the function involved.It is also noted that every in block diagram and or flow chart
The combination of box in a box and block diagram and or flow chart can use the dedicated base for executing defined function or movement
It realizes, or can realize using a combination of dedicated hardware and computer instructions in the system of hardware.
In addition, each functional module in each embodiment of the application can integrate one independent portion of formation together
Point, it is also possible to modules individualism, an independent part can also be integrated to form with two or more modules.
It, can be with if the function is realized and when sold or used as an independent product in the form of software function module
It is stored in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) execute each embodiment the method for the application all or part of the steps.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
The above description is only an example of the present application, the protection scope being not intended to limit this application, for ability
For the technical staff in domain, various changes and changes are possible in this application.Within the spirit and principles of this application, made
Any modification, equivalent substitution, improvement and etc. should be included within the scope of protection of this application.It should also be noted that similar label and
Letter indicates similar terms in following attached drawing, therefore, once it is defined in a certain Xiang Yi attached drawing, then in subsequent attached drawing
In do not need that it is further defined and explained.
The above, the only specific embodiment of the application, but the protection scope of the application is not limited thereto, it is any
Those familiar with the art within the technical scope of the present application, can easily think of the change or the replacement, and should all contain
Lid is within the scope of protection of this application.Therefore, the protection scope of the application shall be subject to the protection scope of the claim.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
Claims (10)
1. a kind of learning effect prediction technique, which is characterized in that the described method includes:
It obtains the expression data of learner's current generation and the learner is parsed according to the expression data of the current generation and work as
The learning state and learning state of last stage records total time, and the classification of the learning state includes active learning state and passiveness
Learning state；
According to the active learning state for obtaining the learning state of the current generation and learning state record total time the current generation
Parameter；
Utilize the active learning state parameter of the current generation and preset learning effect prediction model prediction next stage
Practise effect.
2. learning effect prediction technique according to claim 1, which is characterized in that the table according to the current generation
Feelings data parse learning state and the learning state record total time of the learner, comprising:
The type of active learning state and continuing for every kind of active learning state are obtained according to the expression data of the current generation
Time；
The type of negative learning state and continuing for every kind of negative learning state are obtained according to the expression data of the current generation
Time；
The active learning shape is calculated according to the duration of the type of the active learning state and every kind of active learning state
The state time；
The negative learning shape is calculated according to the duration of the type of the negative learning state and every kind of negative learning state
The state time；
The learning state, which is calculated, according to the active learning state for time and the negative learning state for time records total time.
3. learning effect prediction technique according to claim 2, which is characterized in that the active learning state parameter includes
Active learning state accounting and active learning state ratio；Wherein, active learning state accounting be active learning state for time with
The ratio of learning state record total time；Active learning state ratio is to record in total time in learning state, active learning shape
The ratio of state time and negative learning state for time；
The method also includes the history learning power index of the learner, the basis are calculated according to history learning achievement data
The history learning achievement data calculates the history learning power index of the learner, comprising:
The total number of persons of class where obtaining the learner, in the class of place everyone history learning achievement data and everyone
History active learning state accounting；
Everyone history total performance is calculated according to everyone history learning achievement data；
Everyone the history total performance is normalized, everyone normalization history total performance is obtained；
According to it is described everyone history active learning state accounting and the learner where class total number of persons, calculate class
History active learning state accounting average value；
According to it is described everyone normalization history total performance and the learner where class total number of persons, calculate class's history
Total performance average value；
By the history active learning state accounting of the learner and class's history active learning state accounting average value into
Row compares, and the normalization history total performance of the learner and class's history total performance average value are compared,
Comparison result is obtained, history learning power index is determined according to comparison result.
4. learning effect prediction technique according to claim 3, which is characterized in that utilizing the positive of the current generation
Before the step of learning effect of learning state parameter and preset learning effect prediction model prediction next stage, the method is also
Including constructing preset learning effect prediction model；The preset learning effect prediction model of building, comprising:
Construct preset hundred-mark system learning effect prediction model；
Construct preset social estate system learning effect prediction model.
5. learning effect prediction technique according to claim 4, which is characterized in that the preset hundred-mark system study of building
Effect prediction model, comprising:
Obtain learner's history active learning state accounting and normalization history total performance；
According to learner's history active learning state accounting, normalization history total performance and the first regression equation, calculate
First regression coefficient of first regression equation；
Hundred-mark system learning effect prediction model is obtained according to first regression equation and the first regression coefficient.
6. learning effect prediction technique according to claim 4, which is characterized in that the preset hundred-mark system study of building
Effect prediction model, comprising:
Obtain the history active learning state ratio and normalization history total performance of learner；
According to the history active learning state ratio of the learner, normalization history total performance and the second regression equation, meter
Calculate the second regression coefficient of second regression equation；
Hundred-mark system learning effect prediction model is obtained according to second regression equation and the second regression coefficient.
7. learning effect prediction technique according to claim 4, which is characterized in that the preset social estate system study of building
Effect prediction model, comprising:
Establish the corresponding rank list of school grade；
The social estate system history learning achievement normalization history total performance of learner converted in rank list；
According to the positive state accounting of the history of the learner, the social estate system history learning achievement and logistic regression equation, meter
Calculate the logistic regression coefficient of the logistic regression equation；
Social estate system learning effect prediction model is obtained according to the logistic regression equation and logistic regression coefficient.
8. a kind of learning effect forecasting system, which is characterized in that the system comprises:
Expression data receives and parsing module, for obtaining the history expression data of learner's current generation and according to described current
The history expression data in stage parses learning state and the learning state record total time of learner's current generation；
Learning state processing module is recorded according to the learning state and learning state of the current generation for root and is obtained total time
Active learning state parameter；
Learning effect prediction module, the active learning state parameter and preset learning effect for utilizing the current generation are pre-
Survey the learning effect of model prediction next stage.
9. learning effect forecasting system according to claim 8, which is characterized in that the system also includes prediction model structures
Block is modeled, the prediction model building module includes:
Hundred-mark system learning effect prediction model constructs module, for constructing preset hundred-mark system learning effect prediction model；
Social estate system learning effect prediction model constructs module, for constructing preset social estate system learning effect prediction model.
10. a kind of electronic equipment, which is characterized in that the electronic equipment includes memory and processor, and the memory is used
In storage computer program, the processor runs the computer program so that computer equipment execution is wanted according to right
Learning effect prediction technique described in asking any one of 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910822061.9A CN110533992A (en) | 2019-09-02 | 2019-09-02 | A kind of learning effect prediction technique and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910822061.9A CN110533992A (en) | 2019-09-02 | 2019-09-02 | A kind of learning effect prediction technique and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110533992A true CN110533992A (en) | 2019-12-03 |
Family
ID=68666075
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910822061.9A Pending CN110533992A (en) | 2019-09-02 | 2019-09-02 | A kind of learning effect prediction technique and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110533992A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110930781A (en) * | 2019-12-04 | 2020-03-27 | 广州云蝶科技有限公司 | Recording and broadcasting system |
| CN112053091A (en) * | 2020-09-28 | 2020-12-08 | 北京爱论答科技有限公司 | Data processing method and system based on learning operation |
| CN112348408A (en) * | 2020-11-30 | 2021-02-09 | 上海松鼠课堂人工智能科技有限公司 | Online teachers and materials training and evaluation method |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9082235B2 (en) * | 2011-07-12 | 2015-07-14 | Microsoft Technology Licensing, Llc | Using facial data for device authentication or subject identification |
| CN106157204A (en) * | 2016-06-20 | 2016-11-23 | 山东师范大学 | A kind of Forecasting The Scores method and system based on BP neural network model |
| CN107077711A (en) * | 2014-10-15 | 2017-08-18 | 欧姆龙株式会社 | Apparatus for evaluating characteristics, evaluating characteristics system, evaluating characteristics and evaluating characteristics program |
| CN107967572A (en) * | 2017-12-15 | 2018-04-27 | 华中师范大学 | A kind of intelligent server based on education big data |
| CN108399376A (en) * | 2018-02-07 | 2018-08-14 | 华中师范大学 | Student classroom learning interest intelligent analysis method and system |
| CN108898115A (en) * | 2018-07-03 | 2018-11-27 | 北京大米科技有限公司 | data processing method, storage medium and electronic equipment |
| CN109684984A (en) * | 2018-12-19 | 2019-04-26 | 山东旭兴网络科技有限公司 | A kind of expression recognition method suitable for classroom instruction |
| CN109711371A (en) * | 2018-12-29 | 2019-05-03 | 山东旭兴网络科技有限公司 | A kind of Estimating System of Classroom Teaching based on human facial expression recognition |
| CN109949187A (en) * | 2019-03-01 | 2019-06-28 | 北华大学 | A kind of novel Internet of Things teleeducation system and control method |
-
2019
- 2019-09-02 CN CN201910822061.9A patent/CN110533992A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9082235B2 (en) * | 2011-07-12 | 2015-07-14 | Microsoft Technology Licensing, Llc | Using facial data for device authentication or subject identification |
| CN107077711A (en) * | 2014-10-15 | 2017-08-18 | 欧姆龙株式会社 | Apparatus for evaluating characteristics, evaluating characteristics system, evaluating characteristics and evaluating characteristics program |
| CN106157204A (en) * | 2016-06-20 | 2016-11-23 | 山东师范大学 | A kind of Forecasting The Scores method and system based on BP neural network model |
| CN107967572A (en) * | 2017-12-15 | 2018-04-27 | 华中师范大学 | A kind of intelligent server based on education big data |
| CN108399376A (en) * | 2018-02-07 | 2018-08-14 | 华中师范大学 | Student classroom learning interest intelligent analysis method and system |
| CN108898115A (en) * | 2018-07-03 | 2018-11-27 | 北京大米科技有限公司 | data processing method, storage medium and electronic equipment |
| CN109684984A (en) * | 2018-12-19 | 2019-04-26 | 山东旭兴网络科技有限公司 | A kind of expression recognition method suitable for classroom instruction |
| CN109711371A (en) * | 2018-12-29 | 2019-05-03 | 山东旭兴网络科技有限公司 | A kind of Estimating System of Classroom Teaching based on human facial expression recognition |
| CN109949187A (en) * | 2019-03-01 | 2019-06-28 | 北华大学 | A kind of novel Internet of Things teleeducation system and control method |
Non-Patent Citations (1)
| Title |
|---|
| 赖祥伟: "面向用户行为模型的软件可用性研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110930781A (en) * | 2019-12-04 | 2020-03-27 | 广州云蝶科技有限公司 | Recording and broadcasting system |
| CN110930781B (en) * | 2019-12-04 | 2022-11-22 | 广州云蝶科技有限公司 | Recording and broadcasting system |
| CN112053091A (en) * | 2020-09-28 | 2020-12-08 | 北京爱论答科技有限公司 | Data processing method and system based on learning operation |
| CN112348408A (en) * | 2020-11-30 | 2021-02-09 | 上海松鼠课堂人工智能科技有限公司 | Online teachers and materials training and evaluation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Recommendation system for adaptive learning | |
| Zhang et al. | A learning style classification approach based on deep belief network for large-scale online education | |
| Junker et al. | Cognitive assessment models with few assumptions, and connections with nonparametric item response theory | |
| CN110533992A (en) | A kind of learning effect prediction technique and system | |
| Li et al. | Deep reinforcement learning for adaptive learning systems | |
| Lin et al. | Incorporating student response time and tutor instructional interventions into student modeling | |
| Intisar et al. | Classification of online judge programmers based on rule extraction from self organizing feature map | |
| CN113626499A (en) | Student portrait mining implementation method based on big data warehouse technology | |
| Intisar et al. | Cluster analysis to estimate the difficulty of programming problems | |
| CN114021722A (en) | Attention knowledge tracking method integrating cognitive portrayal | |
| US11416558B2 (en) | System and method for recommending personalized content using contextualized knowledge base | |
| Ghorbani et al. | Learners grouping in e-learning environment using evolutionary fuzzy clustering approach | |
| Rao et al. | Evolution and prediction of radical multi-dimensional e-learning system with cluster based data mining techniques | |
| ALSaad et al. | Unsupervised Approach for Modeling Content Structures of MOOCs. | |
| Hasibuan et al. | Prediction learning style based on prior knowledge for personalized learning | |
| Dai et al. | An improved deep model for knowledge tracing and question-difficulty discovery | |
| de Baker et al. | Chapter X: educational data mining and learning analytics | |
| Ndognkon Manga et al. | An Approach for Non-deterministic and Automatic Detection of Learning Styles with Deep Belief Net | |
| Fernández et al. | Exploring approaches to educational data mining and learning analytics, to measure the level of acquisition of student's learning outcome | |
| Singh et al. | An innovative approach of progressive feedback via artificial neural networks | |
| Pandey et al. | A Practical Approach for Machine Learning and Deep Learning Algorithms: Tools and Techniques Using MATLAB and Python | |
| KR102669921B1 (en) | Method, system and non-transitory computer-readable recording medium for providing information on user's conceptual understanding | |
| US20230087817A1 (en) | System and Methods for Educational and Psychological Modeling and Assessment | |
| Hamal et al. | Artificial Intelligent in Education. Sustainability 2022, 14, 2862 | |
| Wang et al. | Two-Way Neural Network Performance Prediction Model Based on Knowledge Evolution and Individual Similarity. |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191203 |
|
| RJ01 | Rejection of invention patent application after publication |