CN110377814A - Topic recommended method, device and medium - Google Patents
Topic recommended method, device and medium Download PDFInfo
- Publication number
- CN110377814A CN110377814A CN201910468820.6A CN201910468820A CN110377814A CN 110377814 A CN110377814 A CN 110377814A CN 201910468820 A CN201910468820 A CN 201910468820A CN 110377814 A CN110377814 A CN 110377814A
- Authority
- CN
- China
- Prior art keywords
- topic
- knowledge point
- student side
- knowledge
- degree
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- 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
- G09B7/00—Electrically-operated teaching apparatus or devices working with questions and answers
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Educational Technology (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to data analyses, provide a kind of topic recommended method, comprising: acquisition student side history answer data predict student side to the first master degree of knowledge point by deep knowledge tracing model；Student side is obtained to the forgetting degree of knowledge point according to the learning time of knowledge point and forgetting curve；First master degree and forgetting degree of the student side to knowledge point, which are combined, predicts present student side to the second master degree of knowledge point；According to history answer data and knowledge the second master degree of point of student side, topic relevant information is extracted；Predict that student side was not doing the first accuracy on topic according to Bayesian model by topic relevant information；It is ranked up according to the first accuracy to not doing topic, the topic for choosing the first forward setting quantity that sorts pushes student side.The present invention also provides a kind of electronic device and storage mediums.The present invention recommends different topics for master degree of the different students to knowledge point.
Description
Technical field
The present invention relates to data analysis technique fields, more specifically, are related to a kind of topic recommended method, device and Jie
Matter.
Background technique
Existing mechanism in the industry can arrange various winter and summer vacation activities, in addition winter and summer vacation student's time is freer, school and family
Length is little to the control degree of student, lacks the product of control student study on the market.In addition, existing winter and summer vacation currently on the market
Exercise is occurred in the form of this by traditional exercise.The form that this unified purchase exercise is originally handed down to student has following lack
It falls into:
Be to the topic that different students recommend it is identical, the master degree difference of knowledge point can not be carried out according to different students
Topic is recommended, and the recommendation of different topics can not be carried out according to the different light current knowledge points of different students.In addition, can not be according to student
Topic situation is done, the grasp situation that student includes knowledge point to topic is objectively evaluated, recommendation topic cannot be more adjusted.
Summary of the invention
In view of the above problems, recommend not the object of the present invention is to provide a kind of for master degree of the different students to knowledge point
With topic recommended method, electronic device and the storage medium of topic.
To achieve the goals above, the present invention provides a kind of electronic device, and the electronic device includes memory and processing
Device, includes topic recommended program in the memory, realizes when the topic recommended program is executed by the processor and walks as follows
It is rapid:
Step S1 acquires the history answer data of student side, predicts student side to by deep knowledge tracing model
The first master degree gained knowledge a little；
Step S2 obtains student side to the forgetting degree of knowledge point according to the learning time of knowledge point and forgetting curve；
First master degree and forgetting degree of the student side to knowledge point are combined and predict present student side to Xue Zhi by step S3
Know the second master degree of point；
It is related to extract topic according to history answer data and knowledge the second master degree of point of student side by step S4
Information, the topic relevant information include item difficulty, topic information entropy and topic identification；
Step S5 predicts student side on the topic not done by the topic relevant information according to Bayesian model
The first accuracy；
Step S6 is ranked up the topic not done according to the first accuracy, chooses the first forward setting number that sorts
The topic of amount pushes student side.
In addition, to achieve the goals above, the present invention also provides a kind of topic recommended methods, comprising:
Step S1 acquires the history answer data of student side, predicts student side to by deep knowledge tracing model
The first master degree gained knowledge a little；
Step S2 obtains student side to the forgetting degree of knowledge point according to the learning time of knowledge point and forgetting curve；
First master degree and forgetting degree of the student side to knowledge point are combined and predict present student side to Xue Zhi by step S3
Know the second master degree of point；
It is related to extract topic according to history answer data and knowledge the second master degree of point of student side by step S4
Information, the topic relevant information include item difficulty, topic information entropy and topic identification；
Step S5 predicts student side on the topic not done by the topic relevant information according to Bayesian model
The first accuracy；
Step S6 is ranked up the topic not done according to the first accuracy, chooses the first forward setting number that sorts
The topic of amount pushes student side.
Preferably, after step S3, which comprises
Obtain the knowledge point weight of the knowledge point；
Knowledge point weight and the second master degree of knowledge point are combined by following formula, obtain the recommendation of knowledge point:
Wherein, TjFor the recommendation of j-th of knowledge point, WjFor the knowledge point weight of j-th of knowledge point,Know for j-th
Know the second master degree of point；
Knowledge point is ranked up according to the recommendation descending sequence, chooses the second forward setting number that sorts
The knowledge point of amount；
Extract the topic relevant information of the knowledge point for the second setting quantity chosen；
Predicted student in not doing comprising above-mentioned knowledge point according to Bayesian model by the topic relevant information
Topic on the first accuracy；
In the knowledge point of the second setting quantity of selection, topic was not done according to the first accuracy to each knowledge point
It is ranked up, the topic for choosing each knowledge point for the first forward setting quantity that sorts pushes student side.
In addition, after step S3, the method also includes:
Knowledge tree is constructed according to the knowledge hierarchy of school's teaching material, the knowledge tree includes multiple knowledge with set membership
Point；
Obtain the corresponding father node of new knowledge point that degree-of-difficulty factor in knowledge tree is not more than setting difficulty；
Whether the first master degree or/and the second master degree for judging the knowledge point of father node are greater than the set value, and filter out
The father node that one master degree or/and the second master degree are greater than the set value；
According to student side to the history answer data of the topic of the knowledge point of the father node comprising filtering out and to described
Second master degree of the knowledge point of father node extracts the topic relevant information of the topic；
Predicted student comprising above-mentioned father node and new knowledge according to Bayesian model by the topic relevant information
The first accuracy on the topic of point not done；
The topic that do not did comprising father node and new knowledge point is ranked up according to the first accuracy, it is forward to choose sequence
First setting quantity topic student side is pushed.
Preferably, step S4 includes:
Student side history answer is recorded into cuit reaction theory model, obtains item difficulty and topic identification,
In, item response theory model is constructed according to the following formula:
Wherein, P (θ) is the prediction probability that topic is answered questions by student side, and a is discrimination parameter, and D is constant 1.7, and b is topic
Mesh difficult parameters, c is conjecture parameter or pseudo- conjecture parameter, θ are the learning ability of student side；
Multiple topics have been done by multiple student sides to be trained item response theory model, obtain each ginseng of model
Numerical value.
Further, it is preferable that described the step of being trained to item response theory model includes:
Using the first master degree of the knowledge point of student side as the study energy of the student side for the topic for including the knowledge point
Power θ；
Initial assignment is carried out to parameter a, b and c in item response theory model, by the learning ability generation of above-mentioned student side
Enter item response theory model and obtains prediction probability P (θ)；
Using answer questions the quantity including the knowledge point topic divided by include all topic quantity in the knowledge point as learning
The root-mean-square error of destination probability and prediction probability is less than setting by the destination probability for the topic including the knowledge point caused trouble
Value is trained item response theory model as iterated conditional, obtains optimum model parameter a, b and c.
Preferably, after step S3, the method also includes:
Knowledge point of second master degree less than setting value is filtered out as weakness knowledge point；
According to student side to the history answer data of the topic comprising weakness knowledge point and to the weakness knowledge point
Second master degree extracts the topic relevant information of the topic；
By the topic relevant information predicted according to Bayesian model student comprising the weakness knowledge point not
The first accuracy on the topic done；
The topic that do not did comprising weakness knowledge point is ranked up according to the first accuracy, chooses forward first of sorting
The topic of setting quantity pushes student side.
Moreover it is preferred that in step s 4, the step of extracting the topic information entropy, includes:
Obtain the topic information entropy of the topic according to the following formula by the number that letter in topic and Chinese character occur
Wherein, Hs is the topic information entropy of a topic, piFor the number that i-th of letter or text occur in topic, IeFor
The information content of i-th of letter or text, n are the sum of letter and text in the topic.
Preferably, in step s 6, Error processing is carried out to the first accuracy, to the topic not done according to Error processing
The first accuracy afterwards is ranked up.
In addition, to achieve the goals above, the present invention also provides a kind of computer readable storage medium, the computer can
It reads in storage medium to include topic recommended program, when the topic recommended program is executed by processor, realizes that above-mentioned topic pushes away
The step of recommending method.
Topic recommended method, electronic device and computer readable storage medium of the present invention are for different student sides to knowing
The master degree for knowing point recommends the topic not done to it, and the forgetting degree of different time is combined when recommendation, so that recommending more
Accurately.In addition, being made the highest entitled student of most first accuracy according to the type that student does in entire term is calculated
Personalized recommendation is provided, to mitigate vacation burden, learning initiative is improved, allows and review and preview more efficiently.
Detailed description of the invention
Fig. 1 is the application environment schematic diagram of topic recommended method preferred embodiment of the present invention；
Fig. 2 is the module diagram of topic recommended program preferred embodiment in Fig. 1；
Fig. 3 is the flow chart of topic recommended method preferred embodiment of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
Hereinafter, specific embodiments of the present invention will be described in detail with reference to the accompanying drawings.
The present invention provides a kind of topic recommended method, is applied to a kind of electronic device 1.Shown in referring to Fig.1, for present invention topic
The application environment schematic diagram of mesh recommended method preferred embodiment.
In the present embodiment, electronic device 1 can be server, mobile phone, tablet computer, portable computer, desktop meter
Calculation machine etc. has the client terminals of calculation function.
Memory 11 includes the readable storage medium storing program for executing of at least one type.The readable storage medium storing program for executing of at least one type
It can be the non-volatile memory medium of such as flash memory, hard disk, multimedia card, card-type memory.In some embodiments, described can
Reading storage medium can be the internal storage unit of the electronic device 1, such as the hard disk of the electronic device 1.In other realities
It applies in example, the readable storage medium storing program for executing is also possible to the external memory of the electronic device 1, such as on the electronic device 1
The plug-in type hard disk of outfit, intelligent memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD)
Card, flash card (Flash Card) etc..
In the present embodiment, the readable storage medium storing program for executing of the memory 11 is installed on the electronic device commonly used in storage
1 topic recommended program 10 etc..The memory 11 can be also used for temporarily storing the number that has exported or will export
According to.
Processor 12 can be in some embodiments a central processing unit (Central Processing Unit,
CPU), microprocessor or other data processing chips, program code or processing data for being stored in run memory 11, example
Such as execute topic recommended program 10.
Network interface 13 optionally may include standard wireline interface and wireless interface (such as WI-FI interface), be commonly used in
Communication connection is established between the electronic device 1 and other electronic clients.
Communication bus 14 is for realizing the connection communication between these components.
Fig. 1 illustrates only the electronic device 1 with component 11-14, it should be understood that being not required for implementing all show
Component out, the implementation that can be substituted is more or less component.
Optionally, which can also include user interface, and user interface may include input unit such as keyboard
(Keyboard), speech input device such as microphone (microphone) etc. has the client of speech identifying function, voice
Output device such as sound equipment, earphone etc., optionally user interface can also include standard wireline interface and wireless interface.
Optionally, which can also include display, and display is referred to as display screen or display unit.
It can be light-emitting diode display, liquid crystal display, touch-control liquid crystal display and organic light emission in some embodiments
Diode (Organic Light-Emitting Diode, OLED) touches device etc..Display is for showing in the electronic apparatus 1
The information of processing and for showing visual user interface.
Optionally, which further includes touch sensor.It is touched provided by the touch sensor for user
The region for touching operation is known as touch area.In addition, touch sensor described here can be resistive touch sensor, capacitor
Formula touch sensor etc..Moreover, the touch sensor not only includes the touch sensor of contact, proximity may also comprise
Touch sensor etc..In addition, the touch sensor can be single sensor, or such as multiple biographies of array arrangement
Sensor.
Optionally, which can also include logic gates, and sensor, voicefrequency circuit etc. are no longer superfluous herein
It states.
In Installation practice shown in Fig. 1, as may include in a kind of memory 11 of computer storage medium behaviour
Make system and topic recommended program 10；Processor 12 is realized as follows when executing the topic recommended program 10 stored in memory 11
Step:
Step S1 acquires the history answer data of student side, passes through deep knowledge tracing model (Deep Knowledge
Tracing, DKT) student side is predicted to the first master degree of knowledge point；
Step S2 obtains student side to the forgetting degree of knowledge point according to the learning time of knowledge point and forgetting curve；
Step S3 combines student side to the first master degree of knowledge point (such as linear combination, be multiplied) with forgetting degree
Predict present student side to the second master degree of knowledge point；
It is related to extract topic according to history answer data and knowledge the second master degree of point of student side by step S4
Information, the topic relevant information include item difficulty, topic information entropy and topic identification (discrimination)；
Step S5 predicts student side on the topic not done by the topic relevant information according to Bayesian model
The first accuracy, first accuracy indicates that student side answers questions wrong probability to topic was not done；
Step S6 is ranked up the topic not done according to the first accuracy, chooses the first forward setting number that sorts
The topic of amount pushes student side.
Above-mentioned electronic device, which is based on forgetting curve and deep knowledge tracing model, can track student to knowledge point master degree,
Accurately the topic that do not did of knowledge point is recommended.
In other embodiments, the topic recommended program 10 can also be divided into one or more module, and one
Or multiple modules are stored in memory 11, and are executed by processor 12, to complete the present invention.The so-called module of the present invention
It is the series of computation machine program instruction section for referring to complete specific function.It is topic recommended program in Fig. 1 referring to shown in Fig. 2
The functional block diagram of 10 preferred embodiments.The topic recommended program 10 can be divided into acquisition module 110, the first master degree
Obtain module 120, forgetting degree obtains module 130, the second master degree obtains module 140, extraction module 150, the first accuracy and obtains
Module 160 and pushing module 170 are obtained, the acquisition module 110 acquires the history answer data of student side, first master degree
It obtains module 120 and passes through DKT model by the history answer data that acquisition module 110 acquires, predict student side to Xue Zhi
Know the first master degree of point, the forgetting degree obtains module 130 and obtains according to the learning time and forgetting curve of knowledge point
For student side to the forgetting degree of knowledge point, second master degree obtains what module 140 obtained the first master degree acquisition module 120
First master degree and forgetting degree obtain the forgetting degree that module 130 obtains and combine the present student side of prediction to the second of knowledge point
Master degree, the extraction module 150 according to the second master degree obtain module 140 obtain the second master degree of knowledge point and
The history answer data for the student side that acquisition module 110 acquires and, extract topic relevant information, first accuracy obtains
The topic relevant information that module 160 is extracted by extraction module 150 predicts that student side is not being done according to Bayesian model
The first accuracy on the topic crossed, the recommending module 170 according to the first accuracy obtain that module 160 obtains it is first correct
Rate is ranked up the topic not done, and the topic for choosing the first forward setting quantity that sorts pushes student side.
It in one alternate embodiment, further include the first screening module when recommending to review topic in the stage of review, according to
Second master degree screens knowledge point, and first screening module includes:
Knowledge point weight obtaining unit, according to the knowledge point frequency of occurrences of graduation examination over the years (such as being examined in middle school) and
The knowledge point weight of significance level acquisition knowledge point；
Recommendation obtaining unit obtains the recommendation of knowledge point according to knowledge point weight and the second master degree of knowledge point；
Sequencing unit is ranked up knowledge point according to the descending sequence of recommendation；
First selection unit chooses the knowledge point for the second forward setting quantity that sorts, when recommendation previews topic, comprising:
Wherein, extraction module 150 extracts the topic relevant information of the knowledge point for the second setting quantity chosen, and first is correct
The topic relevant information for the knowledge point that rate acquisition module 160 is extracted by extraction module is according to Bayesian model come pre-
First accuracy of the student on the topic not done comprising the knowledge point is surveyed, recommending module 170 is set the second of selection
In the knowledge point of fixed number amount, the topic that do not did of each knowledge point is ranked up according to the first accuracy, it is forward to choose sequence
The topic of each knowledge point of the first setting quantity student side is pushed.
In another alternative embodiment, previewing the stage, further including the second screening module, according to the first master degree or/
Knowledge point is screened with the second master degree, second screening module includes:
Knowledge tree construction unit constructs knowledge tree according to the knowledge hierarchy of school's teaching material, and the knowledge tree includes having father
Multiple knowledge points of subrelation；
Father node obtaining unit obtains the father's section corresponding no more than the new knowledge point of setting difficulty of degree-of-difficulty factor in knowledge tree
Point；
Second selection unit judges whether the first master degree of the knowledge point of father node or/and the second master degree are greater than and sets
Definite value filters out the father node that the first master degree or/and the second master degree are greater than the set value,
Wherein, history answer data for the father node that extraction module 150 is chosen according to the second selection unit and its corresponding
Second master degree of knowledge point, extracts the topic relevant information of the topic of the knowledge point comprising father node, and the first accuracy obtains
Module 160 predicts student comprising above-mentioned father node and new knowledge point by the topic relevant information according to Bayesian model
The topic not done on the first accuracy, recommending module 170 is to including that above-mentioned father node and new knowledge point did not did topic
Mesh is ranked up according to the first accuracy, and the topic for choosing the first forward setting quantity that sorts pushes student side.
It further include third screening module when recommending its weakness knowledge point topic to student side in third alternative embodiment,
Knowledge point of second master degree less than setting value is filtered out as weakness knowledge point, extraction module 150 according to student side to comprising
The history answer data of the topic of weakness knowledge point and the second master degree to the weakness knowledge point, extract the topic
Topic relevant information；First accuracy acquisition module 160 is by the topic relevant information according to Bayesian model come prediction science
Raw the first accuracy on the topic not done comprising the weakness knowledge point；Recommending module 170 is to including weakness knowledge
The topic that do not did of point is ranked up according to the first accuracy, chooses the topic for the first forward setting quantity that sorts to student side
It is pushed.
In addition, the present invention also provides a kind of topic recommended methods.Referring to shown in Fig. 3, be topic recommended method of the present invention compared with
The flow chart of good embodiment.This method can be executed by a device, which can be by software and or hardware realization.
In the present embodiment, topic recommended method includes:
Step S1 acquires the history answer data of student side, predicts student side to by deep knowledge tracing model
The first master degree gained knowledge a little；
Step S2 obtains student side to the forgetting degree of knowledge point according to the learning time of knowledge point and forgetting curve；
First master degree and forgetting degree of the student side to knowledge point are combined and predict present student side to Xue Zhi by step S3
The second master degree of point is known, for example, Chinese mugwort this great forgetting curve of guest is ground by German psychologist Chinese mugwort guest great this (H.Ebbinghaus)
Study carefully proposition, the rule that human brain forgets new things.Fast speed is initially forgotten, if it is more preferable to immediately begin to review effect.
The process of forgetting is not that uniformly, the speed of forgetting can gradually slowly.According to the time of knowledge point study and forgetting curve skill
Art can obtain student to the forgetting degree of knowledge point.When student inscribes on line, DKT can be according to each topic record that does to knowing
Know a point master degree to be updated, but according to forgetting curve theory, knowledge 74% can pass into silence after study one day, and 26% is remembered
Firmly；After a week, it 77% passes into silence, 23% is remembered；79% passes into silence after one month, and 21% is remembered.Therefore, according to the last time
Doing time that topic and DKT value are updated combines forgetting curve, it is assumed that the first master degree of knowledge point is 0.6, if after inscribing
Recommend topic to student within 6th day, the second master degree of knowledge point is 0.6x0.23=0.138；
It is related to extract topic according to history answer data and knowledge the second master degree of point of student side by step S4
Information, the topic relevant information include item difficulty, topic information entropy and topic identification；
Step S5 predicts student side on the topic not done by the topic relevant information according to Bayesian model
The first accuracy；
Step S6 is ranked up the topic not done according to the first accuracy, chooses the first forward setting number that sorts
The topic of amount pushes student side.
In above-mentioned topic recommended method, DKT model carries out real-time update, that is to say, that topic record input mould is done in new
Type calculates the first new master degree again.
In step s3, preferably the first master degree is multiplied with forgetting degree as the second master degree.
In step s 4, student side history answer is recorded into cuit reaction theory (IRT, Item Response
Theory) model obtains item difficulty and topic identification, specifically, comprising:
(1) or (2) constructs IRT model according to the following formula
Wherein, P (θ) is the prediction probability that topic is answered questions by student side, and a is discrimination parameter, and D is constant 1.7, and b is topic
Mesh difficult parameters, c is conjecture parameter or pseudo- conjecture parameter, θ are the learning ability of student side；
Multiple topics have been done by multiple student sides to be trained IRT model, obtain each parameter value of model.
Preferably, the step of IRT model is trained include:
Using the first master degree of the knowledge point of student side as the study energy of the student side for the topic for including the knowledge point
Power θ；
Initial assignment is carried out to parameter a, b and c in IRT model, the learning ability of above-mentioned student side is substituted into IRT model
It obtains prediction probability P (θ)；
Using answer questions the quantity including the knowledge point topic divided by include all topic quantity in the knowledge point as learning
The root-mean-square error of destination probability and prediction probability is less than setting by the destination probability for the topic including the knowledge point caused trouble
Value is trained IRT model as iterated conditional, obtains optimum model parameter a, b and c.
Further, it is preferable that the step of described parameter a, b and c in IRT model carries out initial assignment include:
Initial item difficulty parameter b is obtained by knowledge professional library or (the second accuracy of 1-) is used as initial item difficulty
Parameter b, second accuracy are to answer questions the student side number of topic divided by the student side sum for doing the topic；
Using the difference of the first master degree and second accuracy as initial discrimination parameter a；
Using the minimum value of the destination probability of all student sides as conjecture parameter c.
In step s 4, the topic information entropy indicates that a topic gives user's bring information content, the topic information
When entropy is larger, information that may be unrelated is relatively more, is easy the understanding of fuzzy student, and the information is indicated from distribution or data flow
In time, sample or feature.
The method for extracting the topic information entropy includes:
The number occurred by letter in topic and Chinese character obtains the topic information entropy of the topic according to the following formula (3)
Wherein, Hs is the topic information entropy of a topic, piFor the number that i-th of letter or text occur in topic, IeFor
The information content of i-th of letter or text, n is the sum of letter and text in the topic, such as English has 26 letters,
Each letter is added to occur in article at this time averagely, the information content of each letter are as follows:And
Chinese character there are commonly 2500, if if each Chinese character frequency of occurrence in article is average, the information content of each Chinese character are as follows:
In step s 6, Error processing is carried out to the first accuracy, to the topic not done according to the after Error processing
One accuracy is ranked up, for example, setting error as the intermediate value of the first accuracy location, is subtracted to the first accuracy
Intermediate value and the Error processing to take absolute value, for another example, the first accuracy is within the scope of 0-1, intermediate value 0.5, uses | and first is correct
Rate -0.5 | value the topic not done is ranked up, that is to say, that select close to intermediate value (0.5) the first accuracy it is corresponding
Topic is recommended.
In one particular embodiment of the present invention, from reviewing and previewing two angles to student side recommendation topic, in which:
In the stage of review, when recommending to review topic, comprising:
Include the steps that screening knowledge point between step S3 and S4, the step includes:
Obtain the knowledge point weight of the knowledge point, the knowledge point weight is according to graduation examination over the years (such as middle school
In examine) the knowledge point frequency of occurrences and significance level determine knowledge point different degree, for example, by expert knowledge library (or have through
The verification teacher tested) pass through the frequency of this case of knowledge appearance and the power of shared score value (or shared score value percentage) estimation knowledge point
Weight, such as can (such as the number that knowledge point occurs in setting year is divided by institute by the frequency that set period of time knowledge point occurs
State year) and the product of score value percentage is averagely accounted for as knowledge point weight；
Knowledge point weight and the second master degree of knowledge point are combined by following formula (4), obtain the recommendation of knowledge point
Wherein, TjFor the recommendation of j-th of knowledge point, WjFor the knowledge point weight of j-th of knowledge point,Know for j-th
Know the second master degree of point；
Knowledge point is ranked up according to recommendation descending sequence, chooses the second forward setting quantity that sorts
Knowledge point, the value of recommendation it is bigger illustrate that master degree is lower, knowledge point weight is bigger, it is preferable that selection recommendation be greater than one
A setting value and knowledge point weight are not less than the knowledge point of another setting value, that is to say, that selection master degree, knowledge point power
The biggish knowledge point of weight.
In step s 4, the topic relevant information of the knowledge point of the second setting quantity of selection is extracted.
In step s 5, predicted student comprising above-mentioned knowledge according to Bayesian model by the topic relevant information
The first accuracy on the topic of point not done.
In step s 6, selection second setting quantity knowledge point in, to each knowledge point do not did topic by
It is ranked up according to the first accuracy, chooses the topic of each knowledge point for the first forward setting quantity that sorts to student side
It is pushed.
Preferably, in step sl, topic record is done in the examination for acquiring student side, updates student side to Xue Zhi by DKT
The first master degree for knowing point, in the stage of review, we are updated the topic record that does of examination to DKT model, in conjunction with knowledge point
Weight and next term/in examine the knowledge point that can also touch and recommended, the knowledge point recommended using recommendation selection is right
Face college entrance examination, in the scholar of great examination such as examine and greatly strengthened while the knowledge point not high to master degree is practiced
The confidence that the knowledge point is grasped.
In the stage that previews, when recommendation previews topic, comprising:
Include the steps that screening knowledge point between step S3 and S4, the step includes:
Knowledge tree is constructed according to the knowledge hierarchy of school's teaching material, the knowledge tree includes multiple knowledge with set membership
Point, knowledge tree are a hierarchical instruction figures, express according to teaching material logic establish architectonic knowledge point between because
Fruit relationship or subordinate relation, such as the chapters and sections of first textbook first teach what teaches what all after is according to this knowledge hierarchy；
Degree-of-difficulty factor in knowledge tree is obtained to know no more than the corresponding father node of new knowledge point of setting difficulty (such as 0.5)
The degree-of-difficulty factor for knowing point can be obtained by expert knowledge library；
Whether the first master degree or/and the second master degree for judging the knowledge point of father node are greater than the set value, and filter out
The father node that one master degree or/and the second master degree are greater than the set value, for example, the second master degree and the first master degree maximum value are
100%, the setting value is median 50%, if father node has multiple, it can be determined that the knowledge point of all father nodes
First master degree or/and the second master degree also may determine that the first master degree or/and second palm of the average value of all father nodes
Degree of holding.
In step s 4, according to student side to the history answer data of the topic of the knowledge point of the father node comprising filtering out
And the second master degree of the knowledge point to the father node, extract the topic relevant information of the topic.
In step s 5, predict that student saves comprising above-mentioned father according to Bayesian model by the topic relevant information
The first accuracy on the topic of point and new knowledge point not done.
In step s 6, the topic that do not did for including above-mentioned father node and new knowledge point is arranged according to the first accuracy
Sequence, the topic for choosing the first forward setting quantity that sorts pushes student side, for example, knowledge point D (difficulty 0.5) is new
The new knowledge point in term, the father knowledge point compared to it has and the knowledge point A of only last term, when the second master degree of the knowledge point A
When greater than 50%, knowledge point A grasps relatively good, and the topic comprising knowledge point A and knowledge point D is picked out to student side.
In the above-described embodiments, when father node has it is multiple when, the first master degree or the second master degree can be recommended highest
The topic of father node and new knowledge point.
Topic recommended method described in above-described embodiment, the pass that knowledge point can be taken into account and do not gained knowledge between a little
System, is conducive to student and previews to new knowledge point, allows student to reinforce the grasp situation to knowledge point and to learn future
The knowledge point of habit is previewed.
In one embodiment of the invention, recommend its weakness knowledge point topic to student side, specifically, comprising:
In step s3, knowledge point of second master degree less than setting value is filtered out as weakness knowledge point；
In step s 4, according to student side to the history answer data of the topic comprising weakness knowledge point and to described weak
Second master degree of point knowledge point, extracts the topic relevant information of the topic；
In step s 5, predicted student comprising the weakness according to Bayesian model by the topic relevant information
The first accuracy on the topic of knowledge point not done；
In step s 6, the topic that do not did comprising weakness knowledge point is ranked up according to the first accuracy, the row of selection
The topic of the first forward setting quantity of sequence pushes student side, it is preferable that the first accuracy is within the scope of 0-1, screening
Topic comprising the light current knowledge point of first accuracy not less than 0.5 is further as recommendation topic out, it is preferable that screening
The topic comprising the light current knowledge point that the first accuracy is 0.5 out is as recommendation topic.
Topic recommended method described in above-described embodiment, can predict each student to the grasp situation of knowledge point, and point
Do not recommend the topic under different knowledge points, carries out reinforcement practice for the weakness of each student, specifically: chasing after with deep knowledge
Track model (DKT) does the calculating that topic record does master degree to student, and the relatively low knowledge point of master degree is weak tendency knowledge point.Then
Be with the first accuracy again | 0.5 | value topic is screened.Because the difficulty of topic can be objective, such as from most of
The relatively low topic of the first accuracy is completed in life under relatively high master degree, is defined as the relatively high topic of difficulty；But topic
Difficulty is also subjective, because it is with the quality of teacher's teaching, it is related that the degree of students or student do the time inscribed, example
If student does topic when just learning knowledge point, it can think that comparison is difficult naturally, but if student is when the end of term reviews
Time does same road topic, naturally can comparatively than presensation it is easy.Therefore, we with DKT and do topic difficulty modeling, calculate
Out under student's current ability, do topic the first accuracy be | 0.5 | value allow student to practice, Lai Jiaqiang student is to weak tendency knowledge point
It improves, is not recommended as topic using the first accuracy of weakness knowledge point is higher, that improves student side does topic enthusiasm, when
The big topic of the right difficulty that also can choose the setting quantity of the first accuracy sequence rearward is recommended.
In addition, the embodiment of the present invention also proposes a kind of computer readable storage medium, the computer readable storage medium
In include topic recommended program, the topic recommended program realizes following steps when being executed by processor:
Step S1 acquires the history answer data of student side, predicts student side to by deep knowledge tracing model
The first master degree gained knowledge a little；
Step S2 obtains student side to the forgetting degree of knowledge point according to the learning time of knowledge point and forgetting curve；
First master degree and forgetting degree of the student side to knowledge point are combined and predict present student side to Xue Zhi by step S3
Know the second master degree of point；
It is related to extract topic according to history answer data and knowledge the second master degree of point of student side by step S4
Information, the topic relevant information include item difficulty, topic information entropy and topic identification；
Step S5 predicts student side on the topic not done by the topic relevant information according to Bayesian model
The first accuracy；
Step S6 is ranked up the topic not done according to the first accuracy, chooses the first forward setting number that sorts
The topic of amount pushes student side.
The specific embodiment of the computer readable storage medium of the present invention and above-mentioned topic recommended method, electronic device
Specific embodiment is roughly the same, and details are not described herein.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, device, article or the method that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, device, article or method institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, device of element, article or method.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.Pass through above embodiment party
The description of formula, it is required general that those skilled in the art can be understood that above-described embodiment method can add by software
The mode of hardware platform is realized, naturally it is also possible to which by hardware, but in many cases, the former is more preferably embodiment.It is based on
Such understanding, substantially the part that contributes to existing technology can be with software product in other words for technical solution of the present invention
Form embody, which is stored in a storage medium (such as ROM/RAM, magnetic disk, light as described above
Disk) in, including some instructions are used so that a client terminals (can be mobile phone, computer, server or network visitor
Family end etc.) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of topic recommended method characterized by comprising
Step S1, acquires the history answer data of student side, predicts student side to Xue Zhi by deep knowledge tracing model
Know the first master degree of point；
Step S2 obtains student side to the forgetting degree of knowledge point according to the learning time of knowledge point and forgetting curve；
First master degree and forgetting degree of the student side to knowledge point are combined and predict present student side to knowledge point by step S3
The second master degree；
Step S4 extracts topic relevant information according to history answer data and knowledge the second master degree of point of student side,
The topic relevant information includes item difficulty, topic information entropy and topic identification；
Step S5 predicts of student side on the topic not done according to Bayesian model by the topic relevant information
One accuracy；
Step S6 is ranked up the topic not done according to the first accuracy, chooses the first forward setting quantity that sorts
Topic pushes student side.
2. topic recommended method according to claim 1, which is characterized in that after step S3, which comprises
Obtain the knowledge point weight of the knowledge point；
Knowledge point weight and the second master degree of knowledge point are combined by following formula, obtain the recommendation of knowledge point:
Wherein, TjFor the recommendation of j-th of knowledge point, WjFor the knowledge point weight of j-th of knowledge point,For j-th of knowledge point
The second master degree；
Knowledge point is ranked up according to the recommendation descending sequence, chooses the second forward setting quantity that sorts
Knowledge point；
Extract the topic relevant information of the knowledge point for the second setting quantity chosen；
Predicted student in the topic that do not did comprising above-mentioned knowledge point according to Bayesian model by the topic relevant information
The first accuracy on mesh；
In the knowledge point of the second setting quantity of selection, the topic that do not did of each knowledge point is carried out according to the first accuracy
Sequence, the topic for choosing each knowledge point for the first forward setting quantity that sorts push student side.
3. topic recommended method according to claim 1, which is characterized in that after step S3, the method also includes:
Knowledge tree is constructed according to the knowledge hierarchy of school's teaching material, the knowledge tree includes multiple knowledge points with set membership；
Obtain the corresponding father node of new knowledge point that degree-of-difficulty factor in knowledge tree is not more than setting difficulty；
Whether the first master degree or/and the second master degree for judging the knowledge point of father node are greater than the set value, and filter out first palm
The father node that degree of holding or/and the second master degree are greater than the set value；
According to student side to the history answer data of the topic of the knowledge point of the father node comprising filtering out and to father section
Second master degree of the knowledge point of point, extracts the topic relevant information of the topic；
Predicted student comprising above-mentioned father node and new knowledge point according to Bayesian model by the topic relevant information
The first accuracy on the topic not done；
The topic that do not did comprising father node and new knowledge point is ranked up according to the first accuracy, chooses and sorts forward the
The topic of one setting quantity pushes student side.
4. topic recommended method according to claim 1, which is characterized in that step S4 includes:
Student side history answer is recorded into cuit reaction theory model, obtains item difficulty and topic identification, wherein root
Item response theory model is constructed according to following formula:
Wherein, P (θ) is the prediction probability that topic is answered questions by student side, and a is discrimination parameter, and D is constant 1.7, and b is that topic is difficult
Parameter is spent, c is conjecture parameter or pseudo- conjecture parameter, θ are the learning ability of student side；
Multiple topics have been done by multiple student sides to be trained item response theory model, obtain each parameter of model
Value.
5. topic recommended method according to claim 4, which is characterized in that described to be instructed to item response theory model
Experienced step includes:
Using the first master degree of the knowledge point of student side as the learning ability θ of the student side for the topic for including the knowledge point；
Initial assignment is carried out to parameter a, b and c in item response theory model, the learning ability of above-mentioned student side is substituted into item
Mesh reaction theory model obtains prediction probability P (θ)；
Using answer questions the quantity including the knowledge point topic divided by include all topic quantity in the knowledge point as student side
The topic including the knowledge point destination probability, by the root-mean-square error of destination probability and prediction probability less than setting value make
Item response theory model is trained for iterated conditional, obtains optimum model parameter a, b and c.
6. topic recommended method according to claim 1, which is characterized in that after step S3, the method also includes:
Knowledge point of second master degree less than setting value is filtered out as weakness knowledge point；
According to student side to the history answer data of the topic comprising weakness knowledge point and to the second of the weakness knowledge point
Master degree extracts the topic relevant information of the topic；
Predicted student in not doing comprising the weakness knowledge point according to Bayesian model by the topic relevant information
Topic on the first accuracy；
The topic that do not did comprising weakness knowledge point is ranked up according to the first accuracy, chooses the first forward setting of sorting
The topic of quantity pushes student side.
7. topic recommended method according to claim 1, which is characterized in that in step s 4, extract the topic information
The step of entropy includes:
Obtain the topic information entropy of the topic according to the following formula by the number that letter in topic and Chinese character occur:
Wherein, Hs is the topic information entropy of a topic, piFor the number that i-th of letter or text occur in topic, IeIt is described
The information content of i-th of letter or text, n are the sum of letter and text in the topic.
8. topic recommended method according to claim 1, which is characterized in that in step s 6, carried out to the first accuracy
Error processing is ranked up the topic not done according to the first accuracy after Error processing.
9. a kind of electronic device, which is characterized in that including memory and processor, topic is stored in the memory and recommends journey
Sequence, the topic recommended program realize following steps when being executed by the processor:
Step S1, acquires the history answer data of student side, predicts student side to Xue Zhi by deep knowledge tracing model
Know the first master degree of point；
Step S2 obtains student side to the forgetting degree of knowledge point according to the learning time of knowledge point and forgetting curve；
First master degree and forgetting degree of the student side to knowledge point are combined and predict present student side to knowledge point by step S3
The second master degree；
Step S4 extracts topic relevant information according to history answer data and knowledge the second master degree of point of student side,
The topic relevant information includes item difficulty, topic information entropy and topic identification；
Step S5 predicts of student side on the topic not done according to Bayesian model by the topic relevant information
One accuracy；
Step S6 is ranked up the topic not done according to the first accuracy, chooses the first forward setting quantity that sorts
Topic pushes student side.
10. a kind of computer readable storage medium, which is characterized in that include that topic pushes away in the computer readable storage medium
Program is recommended, when the topic recommended program is executed by processor, is realized as described in any one of claims 1 to 8 claim
The step of topic recommended method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910468820.6A CN110377814A (en) | 2019-05-31 | 2019-05-31 | Topic recommended method, device and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910468820.6A CN110377814A (en) | 2019-05-31 | 2019-05-31 | Topic recommended method, device and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110377814A true CN110377814A (en) | 2019-10-25 |
Family
ID=68249585
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910468820.6A Pending CN110377814A (en) | 2019-05-31 | 2019-05-31 | Topic recommended method, device and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110377814A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110851723A (en) * | 2019-11-14 | 2020-02-28 | 上海钦文信息科技有限公司 | English exercise recommendation method based on large-scale knowledge point labeling result |
| CN110929020A (en) * | 2019-12-18 | 2020-03-27 | 四川文轩教育科技有限公司 | Knowledge point mastery degree analysis method based on test achievement |
| CN110991830A (en) * | 2019-11-19 | 2020-04-10 | 浙江创课网络科技有限公司 | Test paper analysis report system and method |
| CN110991645A (en) * | 2019-11-18 | 2020-04-10 | 广东宜学通教育科技有限公司 | Self-adaptive learning method, system and storage medium based on knowledge model |
| CN110992741A (en) * | 2019-11-15 | 2020-04-10 | 深圳算子科技有限公司 | Learning auxiliary method and system based on classroom emotion and behavior analysis |
| CN111081106A (en) * | 2019-12-13 | 2020-04-28 | 北京爱论答科技有限公司 | Job pushing method, system, equipment and storage medium |
| CN111861815A (en) * | 2020-06-19 | 2020-10-30 | 北京国音红杉树教育科技有限公司 | Method and device for evaluating memory level of user in word listening learning |
| CN111858906A (en) * | 2020-07-21 | 2020-10-30 | 广州视源电子科技股份有限公司 | Exercise recommendation method and device, electronic equipment and computer-readable storage medium |
| CN111898803A (en) * | 2020-07-09 | 2020-11-06 | 西北大学 | Exercise prediction method, system, equipment and storage medium |
| CN111914176A (en) * | 2020-08-07 | 2020-11-10 | 腾讯科技（深圳）有限公司 | Method and device for recommending subjects |
| CN111949882A (en) * | 2020-08-18 | 2020-11-17 | 西安邮电大学 | Intelligent diagnosis method for structural defects of domain knowledge points |
| CN112256743A (en) * | 2020-10-22 | 2021-01-22 | 北京猿力未来科技有限公司 | Adaptive question setting method, equipment and storage medium |
| CN112667801A (en) * | 2021-03-16 | 2021-04-16 | 北京世纪好未来教育科技有限公司 | Test question recommendation method and device |
| CN114021029A (en) * | 2021-11-05 | 2022-02-08 | 海信集团控股股份有限公司 | Test question recommendation method and device |
| WO2022068435A1 (en) * | 2020-09-30 | 2022-04-07 | 上海松鼠课堂人工智能科技有限公司 | Method for measuring knowledge point mastering state |
| CN117743699A (en) * | 2024-02-20 | 2024-03-22 | 山东省计算中心（国家超级计算济南中心） | Problem recommendation method and system based on DKT and Topson sampling algorithm |
| CN117891532A (en) * | 2024-03-15 | 2024-04-16 | 东北大学 | Terminal energy efficiency optimization unloading method based on attention multi-index sorting |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080124696A1 (en) * | 2006-10-26 | 2008-05-29 | Houser Ronald L | Empirical development of learning content using educational measurement scales |
| US20150170536A1 (en) * | 2013-12-18 | 2015-06-18 | William Marsh Rice University | Time-Varying Learning and Content Analytics Via Sparse Factor Analysis |
| JP2016126029A (en) * | 2014-12-26 | 2016-07-11 | 公益財団法人 日本英語検定協会 | Computing for estimating capability value of multiple examinees on basis of item response theory |
| CN107067849A (en) * | 2017-04-21 | 2017-08-18 | 杭州博世数据网络有限公司 | Parent's end system based on cloud teaching platform |
| CN107203583A (en) * | 2017-03-27 | 2017-09-26 | 杭州博世数据网络有限公司 | It is a kind of that method is inscribed based on the smart group that big data is analyzed |
| CN108229718A (en) * | 2016-12-22 | 2018-06-29 | 上海谦问万答吧云计算科技有限公司 | A kind of information forecasting method and device |
-
2019
- 2019-05-31 CN CN201910468820.6A patent/CN110377814A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080124696A1 (en) * | 2006-10-26 | 2008-05-29 | Houser Ronald L | Empirical development of learning content using educational measurement scales |
| US20150170536A1 (en) * | 2013-12-18 | 2015-06-18 | William Marsh Rice University | Time-Varying Learning and Content Analytics Via Sparse Factor Analysis |
| JP2016126029A (en) * | 2014-12-26 | 2016-07-11 | 公益財団法人 日本英語検定協会 | Computing for estimating capability value of multiple examinees on basis of item response theory |
| CN108229718A (en) * | 2016-12-22 | 2018-06-29 | 上海谦问万答吧云计算科技有限公司 | A kind of information forecasting method and device |
| CN107203583A (en) * | 2017-03-27 | 2017-09-26 | 杭州博世数据网络有限公司 | It is a kind of that method is inscribed based on the smart group that big data is analyzed |
| CN107067849A (en) * | 2017-04-21 | 2017-08-18 | 杭州博世数据网络有限公司 | Parent's end system based on cloud teaching platform |
Cited By (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110851723A (en) * | 2019-11-14 | 2020-02-28 | 上海钦文信息科技有限公司 | English exercise recommendation method based on large-scale knowledge point labeling result |
| CN110851723B (en) * | 2019-11-14 | 2022-09-23 | 学利网络科技(上海)有限公司 | English exercise recommendation method based on large-scale knowledge point labeling result |
| CN110992741A (en) * | 2019-11-15 | 2020-04-10 | 深圳算子科技有限公司 | Learning auxiliary method and system based on classroom emotion and behavior analysis |
| CN110992741B (en) * | 2019-11-15 | 2022-05-17 | 深圳算子科技有限公司 | Learning auxiliary method and system based on classroom emotion and behavior analysis |
| CN110991645A (en) * | 2019-11-18 | 2020-04-10 | 广东宜学通教育科技有限公司 | Self-adaptive learning method, system and storage medium based on knowledge model |
| CN110991645B (en) * | 2019-11-18 | 2024-03-29 | 广东宜学通教育科技有限公司 | Self-adaptive learning method, system and storage medium based on knowledge model |
| CN110991830A (en) * | 2019-11-19 | 2020-04-10 | 浙江创课网络科技有限公司 | Test paper analysis report system and method |
| CN110991830B (en) * | 2019-11-19 | 2023-07-07 | 河北仓澜教育科技集团有限公司 | Test paper analysis reporting system and method |
| CN111081106A (en) * | 2019-12-13 | 2020-04-28 | 北京爱论答科技有限公司 | Job pushing method, system, equipment and storage medium |
| CN110929020A (en) * | 2019-12-18 | 2020-03-27 | 四川文轩教育科技有限公司 | Knowledge point mastery degree analysis method based on test achievement |
| CN111861815A (en) * | 2020-06-19 | 2020-10-30 | 北京国音红杉树教育科技有限公司 | Method and device for evaluating memory level of user in word listening learning |
| CN111861815B (en) * | 2020-06-19 | 2024-02-02 | 北京国音红杉树教育科技有限公司 | Method and device for evaluating memory level of user in word listening learning |
| CN111898803B (en) * | 2020-07-09 | 2023-10-24 | 西北大学 | Problem prediction method, system, equipment and storage medium |
| CN111898803A (en) * | 2020-07-09 | 2020-11-06 | 西北大学 | Exercise prediction method, system, equipment and storage medium |
| CN111858906B (en) * | 2020-07-21 | 2023-11-03 | 广州视源电子科技股份有限公司 | Problem recommendation method and device, electronic equipment and computer readable storage medium |
| CN111858906A (en) * | 2020-07-21 | 2020-10-30 | 广州视源电子科技股份有限公司 | Exercise recommendation method and device, electronic equipment and computer-readable storage medium |
| CN111914176B (en) * | 2020-08-07 | 2023-10-27 | 腾讯科技（深圳）有限公司 | Question recommendation method and device |
| CN111914176A (en) * | 2020-08-07 | 2020-11-10 | 腾讯科技（深圳）有限公司 | Method and device for recommending subjects |
| CN111949882B (en) * | 2020-08-18 | 2023-09-08 | 西安邮电大学 | Intelligent diagnosis method for domain knowledge point structure defects |
| CN111949882A (en) * | 2020-08-18 | 2020-11-17 | 西安邮电大学 | Intelligent diagnosis method for structural defects of domain knowledge points |
| WO2022068435A1 (en) * | 2020-09-30 | 2022-04-07 | 上海松鼠课堂人工智能科技有限公司 | Method for measuring knowledge point mastering state |
| CN112256743A (en) * | 2020-10-22 | 2021-01-22 | 北京猿力未来科技有限公司 | Adaptive question setting method, equipment and storage medium |
| CN112256743B (en) * | 2020-10-22 | 2024-06-04 | 北京猿力未来科技有限公司 | Self-adaptive question setting method, device and storage medium |
| CN112667801A (en) * | 2021-03-16 | 2021-04-16 | 北京世纪好未来教育科技有限公司 | Test question recommendation method and device |
| CN114021029A (en) * | 2021-11-05 | 2022-02-08 | 海信集团控股股份有限公司 | Test question recommendation method and device |
| CN117743699A (en) * | 2024-02-20 | 2024-03-22 | 山东省计算中心（国家超级计算济南中心） | Problem recommendation method and system based on DKT and Topson sampling algorithm |
| CN117743699B (en) * | 2024-02-20 | 2024-05-14 | 山东省计算中心（国家超级计算济南中心） | Problem recommendation method and system based on DKT and Topson sampling algorithm |
| CN117891532A (en) * | 2024-03-15 | 2024-04-16 | 东北大学 | Terminal energy efficiency optimization unloading method based on attention multi-index sorting |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110377814A (en) | Topic recommended method, device and medium | |
| Van Buuren | Flexible imputation of missing data | |
| Rodriguez et al. | A computational social science perspective on qualitative data exploration: Using topic models for the descriptive analysis of social media data | |
| CN112632385B (en) | Course recommendation method, course recommendation device, computer equipment and medium | |
| CN108021616B (en) | Community question-answer expert recommendation method based on recurrent neural network | |
| Kitchenham et al. | Evidence-based software engineering and systematic reviews | |
| CN110399541A (en) | Topic recommended method, device and storage medium based on deep learning | |
| He | Improving user experience with case-based reasoning systems using text mining and Web 2.0 | |
| CN108921221A (en) | Generation method, device, equipment and the storage medium of user characteristics | |
| CN108292205A (en) | System and method for refining concept automatically according to mathematical problem and carrying out dynamic construction and test to mathematical problem according to multiple mathematical concepts | |
| CN113656687B (en) | Teacher portrait construction method based on teaching and research data | |
| CN111709575A (en) | Academic achievement prediction method based on C-LSTM | |
| CN111090809A (en) | Topic recommendation method and device, computer equipment and storage medium | |
| CN113886567A (en) | Teaching method and system based on knowledge graph | |
| CN106796618A (en) | Time series forecasting device and time sequence forecasting method | |
| CN110276382A (en) | Listener clustering method, apparatus and medium based on spectral clustering | |
| Maity et al. | Analysis and prediction of question topic popularity in community Q&A sites: a case study of Quora | |
| He et al. | Predictive analytics machinery for STEM student success studies | |
| Marshall | Contribution of open-ended questions in student evaluation of teaching | |
| Tam et al. | Rough set theory for distilling construction safety measures | |
| Charitopoulos et al. | Educational data mining and data analysis for optimal learning content management: Applied in moodle for undergraduate engineering studies | |
| CN105512976A (en) | Data collection system | |
| Baker | Research methods using R: Advanced data analysis in the Behavioural and biological sciences | |
| Kharis et al. | Design of student success prediction application in online learning using Fuzzy-KNN | |
| CN112734142B (en) | Resource learning path planning method and device based on deep learning |
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: 20191025 |
|
| RJ01 | Rejection of invention patent application after publication |