A method for dynamically tracking the knowledge changes of students

Abstract

The present invention discloses a method for dynamically tracking the knowledge changes of students, which relates to the field of knowledge tracking in intelligent education. The method includes the following steps: constructing a knowledge state model of students based on the independent incremental characteristics of students' knowledge growth; constructing a meta-knowledge dictionary based on the knowledge points of learning projects, and constructing a Bayesian framework for students to answer questions according to the meta-knowledge dictionary and the knowledge state model of students, so as to obtain a loss function model for students to answer questions; according to the answer data of students and the loss function model for students to answer questions, using the alternating direction multiplier method to obtain the loss function result for students to answer questions, and according to the loss function result for students to answer questions, using the coordinate descent strategy to update the meta-knowledge dictionary and the knowledge state of students, so as to dynamically track the changes of students' knowledge state. The present invention can effectively diagnose the current knowledge state of students, and does not require a large amount of offline training, and can adapt to the dynamically changing online learning environment.

Description

Technical Field

[0001] The present invention relates to the field of knowledge tracking in intelligent education, and particularly to a method for dynamically tracking the knowledge changes of students. Background Art

[0002] Accurately evaluating the learning status of students is crucial for recommending resources suitable for their learning needs. Knowledge tracking technology is the key to achieving this goal, which predicts students' future performance by analyzing their learning history. The method based on the traditional Bayesian knowledge tracking model simplifies the knowledge state of learners into a binary form, while the method based on the deep knowledge tracking model regards the knowledge state of students as continuous and uses a recurrent neural network to predict the performance of learners.

[0003] Currently, although the methods based on the deep knowledge tracking model have made remarkable progress in knowledge tracking, they mainly focus on predicting the future learning performance of learners and fail to effectively diagnose the current knowledge state of students. In addition, these methods often require a large amount of offline training and are not suitable for the dynamically changing online learning environment. Summary of the Invention

[0004] In view of the above deficiencies in the prior art, the present invention provides a method for dynamically tracking the knowledge changes of students.

[0005] In order to achieve the above invention purpose, the technical solution adopted by the present invention is as follows:

[0006] A method for dynamically tracking the knowledge changes of students, comprising the following steps:

[0007] S1. Construct a knowledge state model of a student based on the independent increment characteristic of the student's knowledge growth;

[0008] S2. Construct a meta-knowledge dictionary based on the knowledge points of the learning project, and construct a Bayesian framework for the student to answer questions according to the meta-knowledge dictionary and the knowledge state model of the student in step S1, so as to obtain a loss function model for the student to answer questions;

[0009] S3. According to the student's answer data and the loss function model for the student to answer questions in step S2, use the alternating direction method of multipliers to obtain the result of the loss function for the student to answer questions, and update the meta-knowledge dictionary and the knowledge state of the student by using the coordinate descent strategy according to the result of the loss function for the student to answer questions, so as to dynamically track the change of the student's knowledge state.

[0010] Further, in step S1, constructing a knowledge state model of a student based on the independent increment characteristic of the student's knowledge growth, which is expressed as:

[0011] ,

[0012] ,

[0013] ;

[0014] wherein: is the knowledge state of the student at time t, is the knowledge state of the student at time s, is a normal distribution with a mean of 0 and a variance of t - s, is the knowledge state of the student, is the incremental distribution satisfied by the knowledge state of the student, is a linear transformation matrix, is the variance of the normal distribution that X follows, where X is the increment of the student's knowledge growth, is the transpose symbol, is the student's knowledge state from transferring to the probability density function, is the exponential function symbol, is the student at time the knowledge state vector, is the student at time the knowledge state vector, is the i th moment in the time series, is the i +1 th moment in the time series.

[0015] Furthermore, in step S2, a meta - knowledge dictionary is constructed based on the knowledge points of the learning project. The specific process is as follows: The knowledge points of the learning project are divided into the smallest knowledge - point units. The smallest knowledge - point unit is indivisible and includes only one of the states of mastered or not mastered, and its linear combination is used to represent the knowledge entity. The smallest knowledge - point units are numbered to construct a corresponding matrix to build the meta - knowledge dictionary.

[0016] Furthermore, in step S2, according to the meta - knowledge dictionary and the student's knowledge state model in step S1, a Bayesian framework for the student to answer questions is constructed, expressed as:

[0017] ,

[0018] ,

[0019] ,

[0020] ;

[0021] wherein: is the given dictionary matrix and the knowledge state of students , the situation y of students answering questions follows a normal distribution with a mean of and a variance of . is a normal distribution with a mean of and a variance of . is the th column of the dictionary matrix i . is a normal distribution with a mean of 0 and a variance of . is the knowledge state vector of the i th row of the student. is a Laplace distribution with a mean of 0 and a scale parameter of b. is a normal distribution with a mean of 0 and a variance of t - s. is the knowledge state vector of the student at time . is the knowledge state vector of the student at time . is the i th moment in the time series. is the i +1 th moment in the time series.

[0022] Furthermore, in step S2, a loss function model for students answering questions is obtained, expressed as:

[0023]

[0024] where: is the loss function for students answering questions. is to take the minimum value. represents the knowledge dictionary and the knowledge state for the reconstruction error of the student answer data matrix . is the Frobenius norm squared loss function. is the student answer data matrix. is the dictionary matrix. is the knowledge state of the student. is the number of students. is the tracking term. is the knowledge state vector of the i +1 th row of the student. is the knowledge state vector of the i th row of the student. is the th i row of the dictionary matrix.

[0025] Further, in step S3, according to the student's answer data and the loss function model of the student answering questions in step S2, the alternating direction multiplier method is used to obtain the loss function result of the student answering questions, including the following steps:

[0026] A1. When fixing the student's knowledge state, use the alternating direction multiplier method to update the meta-knowledge dictionary, expressed as:

[0027]

[0028] Where: is the loss function of the dictionary matrix is the reconstruction error of the knowledge dictionary and the knowledge state with respect to the student answer data matrix ; is the Frobenius norm square loss function, is the student answer data matrix, is the dictionary matrix, is the student's knowledge state, is the number of students, is the th row of the dictionary matrix ; i row;

[0029] A2. When fixing the meta-knowledge dictionary, use the alternating direction multiplier method to update the student's knowledge state, expressed as:

[0030]

[0031] Where: is the loss function of the student's knowledge state ; is the i +1th row knowledge state vector of the student, is the i th row knowledge state vector of the student;

[0032] A3. Calculate the first term of the loss function model of the student answering questions, expressed as:

[0033]

[0034] Where: is the trace of the matrix, is the transpose symbol;

[0035] A4. Obtain the loss function result of the student's answer to the question based on the updated meta-knowledge dictionary in step A1, the updated knowledge state of the student in step A2, and the first term of the loss function model for the student's answer to the question.

[0036] Further, in step S3, according to the loss function result of the student's answer to the question, update the meta-knowledge dictionary and the student's knowledge state using the coordinate descent strategy, including the following steps:

[0037] B1. Determine the meta-knowledge dictionary and the student's knowledge state based on the loss function result of the student's answer to the question, expressed as:

[0038] ,

[0039] ,

[0040] ,

[0041] ,

[0042] ;

[0043] Where: is the knowledge state of the student at time t, is a function for returning the index position of the minimum element in the array, is the reconstruction error between the student's answer data matrix and the dictionary representation , is the student's answer data matrix at time t, is the dictionary matrix at time t - 1, is the knowledge state of the student, is the Frobenius norm squared loss function, is the sparsity regularization parameter, is the temporal smoothness regularization parameter, is the knowledge state of the student at time t - 1, is the dictionary matrix at time t, is the trace of the matrix, is the dictionary matrix, , , is the student's answer data matrix, is the transpose symbol, is the balancing weight, is the objective function symbol;

[0044] B2. Update the knowledge dictionary and the student's knowledge state in step B1 using the coordinate descent strategy, expressed as:

[0045] ,

[0046] ;

[0047] wherein: is the value of the i-th column of the student's knowledge state at time t, is the value of the i-th column of the student's knowledge state at time t-1, is the outer product matrix of the knowledge dictionary at time t-1, is the matrix product of the knowledge dictionary and the student's answer data matrix at time t-1, is the value of the j-th column of the dictionary matrix at time t, is the value of the j-th column of the dictionary matrix at time t-1, is the value of the j-th row and j-th column at time t, is the value of the j-th column at time t, is the value of the j-th column of the transpose at time t.

[0048] The present invention has the following beneficial effects:

[0049] (1) The present invention constructs a knowledge state model of students based on the independent increment characteristic of students' knowledge growth, then constructs a meta-knowledge dictionary based on the knowledge points of learning projects, constructs a Bayesian framework for students to answer questions according to the meta-knowledge dictionary and the knowledge state model of students to obtain a loss function model for students to answer questions, and finally, according to the students' answer data and the loss function model for students to answer questions, uses the alternating direction multiplier method to obtain the result of the loss function for students to answer questions, and updates the meta-knowledge dictionary and the knowledge state of students by adopting a coordinate descent strategy according to the result of the loss function for students to answer questions. The whole process can effectively diagnose the current knowledge state of students, and does not require a large amount of offline training, and can adapt to the dynamically changing online learning environment;

[0050] (2) The present invention constructs a meta-knowledge dictionary based on the knowledge points of learning projects, and a meta-knowledge dictionary theory is proposed in this process for a more detailed division of professional knowledge points. This theory assumes that there are indivisible minimum meta-knowledge point units, each unit has the same difficulty, and knowledge entities are regarded as linear combinations of these meta-knowledge points. The more meta-knowledge points it contains, the greater the difficulty of the knowledge entity. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a schematic flowchart of a method for dynamically tracking students' knowledge changes. DETAILED DESCRIPTION OF THE INVENTION

[0052] The following describes the specific embodiments of the present invention to facilitate those skilled in the art of this technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art of this technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions and creations using the concept of the present invention are within the scope of protection.

[0053] As Figure 1 shown, a method for dynamically tracking the knowledge change of students includes steps S1 - S3, specifically as follows:

[0054] S1. Construct a knowledge state model of students based on the independent increment characteristic of students' knowledge growth.

[0055] In an alternative embodiment of the present invention, the present invention constructs a knowledge state model of students based on the independent increment characteristic of students' knowledge growth, expressed as:

[0056] ,

[0057] ,

[0058] ;

[0059] Wherein: is the knowledge state of the student at time t, is the knowledge state of the student at time s, is a normal distribution with a mean of 0 and a variance of t - s, is the knowledge state of the student, is the increment distribution satisfied by the knowledge state of the student, is a linear transformation matrix, is the variance of the normal distribution that X follows, where X is the increment of the student's knowledge growth, is the transpose symbol, is the probability density function of the student's knowledge state transferring from to , is the exponential function symbol, is the knowledge state vector of the student at time , is the knowledge state vector of the student at time , is the i th moment in the time series, is the i +1 th moment in the time series.

[0060] S2. Construct a meta-knowledge dictionary based on the knowledge points of the learning project. According to the meta-knowledge dictionary and the student's knowledge state model in step S1, construct a Bayesian framework for the student to answer questions to obtain the loss function model for the student to answer questions.

[0061] In an alternative embodiment of the present invention, the present invention constructs a meta-knowledge dictionary based on the knowledge points of the learning project. The specific process is as follows: Divide the knowledge points of the learning project into the smallest knowledge point units. The smallest knowledge point units are indivisible and only include one of the states of mastered or not mastered, and their linear combinations are used to represent knowledge entities. Number the smallest knowledge point units to construct a corresponding matrix to construct the meta-knowledge dictionary.

[0062] The present invention constructs a Bayesian framework for the student to answer questions according to the meta-knowledge dictionary and the student's knowledge state model in step S1, which is expressed as:

[0063] ,

[0064] ,

[0065] ,

[0066] ;

[0067] Where: is the given dictionary matrix and the student's knowledge state , the situation y of the student answering the question follows a normal distribution with a mean of and a variance of , is a normal distribution with a mean of and a variance of , is the th column of the dictionary matrix i , is a normal distribution with a mean of 0 and a variance of , is the i th row knowledge state vector of the student, is a Laplace distribution with a mean of 0 and a scale parameter of b, is a normal distribution with a mean of 0 and a variance of t - s, is the knowledge state vector of the student at time , is the knowledge state vector of the student at time , is the i th moment in the time series, is the iAt time +1.

[0068] Specifically, according to the Bayesian framework of students' answering questions, the present invention obtains the posterior parameters of students' answering questions, expressed as:

[0069]

[0070] Where: is the posterior distribution, representing the joint distribution of the dictionary matrix and the hyperparameters given the student answer data matrix and the student's knowledge state , is the likelihood function of the student answer, representing the probability of the student answer data matrix and the student's knowledge state given the dictionary matrix , is the prior distribution of the dictionary matrix , is the prior distribution of the student's knowledge state .

[0071] The present invention calculates the likelihood function of the student answer, expressed as:

[0072] ;

[0073] Where: is the product of the likelihood functions of each row of the student answer in the student answer data matrix , is the variance of the observation noise of the normal distribution corresponding to the observation model in A1, is the scale parameter of the prior of the knowledge state in A1, is the number of students, is the i -th row knowledge state vector of the student.

[0074] The present invention calculates the prior distribution of the dictionary matrix , expressed as:

[0075]

[0076] Where: is the variance of the normal distribution corresponding to the prior of the dictionary matrix in A1, is the symbol of the quadrature formula, k is the number of knowledge points, is the -th knowledge point in the dictionary matrix.

[0077] The present invention calculates the student's knowledge state The prior distribution of is expressed as:

[0078] .

[0079] The present invention updates the posterior parameter of the student's answer to:

[0080] .

[0081] The present invention simplifies the above formula. Assuming that the time interval Δt = m is the same, the formula is then simplified to:

[0082] .

[0083] The present invention takes the logarithm of both sides of the above formula and expresses it as:

[0084] .

[0085] The present invention obtains the loss function model of the student's answer from the above formula.

[0086] The present invention obtains the loss function model of the student's answer, which is expressed as:

[0087]

[0088] Where: is the loss function of the student's answer, is to take the minimum value, represents the knowledge dictionary and the knowledge state for the reconstruction error of the student answer data matrix , is the Frobenius norm square loss function, is the student answer data matrix, is the dictionary matrix, is the student's knowledge state, is the number of students, is the tracking term, is the i +1 row knowledge state vector of the student, is the i row knowledge state vector of the student, is the of the dictionary matrix i column.

[0089] S3. According to the student's answer data and the loss function model for the student to answer questions in step S2, the alternating direction multiplier method is used to obtain the loss function result for the student to answer questions. And according to the loss function result for the student to answer questions, the coordinate descent strategy is adopted to update the meta-knowledge dictionary and the student's knowledge state, so as to dynamically track the change of the student's knowledge state.

[0090] In an optional embodiment of the present invention, according to the student's answer data and the loss function model for the student to answer questions in step S2, the alternating direction multiplier method is used to obtain the loss function result for the student to answer questions, including the following steps:

[0091] A1. When fixing the student's knowledge state, the alternating direction multiplier method is used to update the meta-knowledge dictionary, which is expressed as:

[0092]

[0093] Where: is the loss function of the dictionary matrix represents the reconstruction error of the knowledge dictionary and the knowledge state for the student answer data matrix is the Frobenius norm square loss function, is the student answer data matrix, is the dictionary matrix, is the student's knowledge state, is the number of students, is the th column of the dictionary matrix i ;

[0094]

[0095] A2. When fixing the meta-knowledge dictionary, the alternating direction multiplier method is used to update the student's knowledge state, which is expressed as:

[0096] Where: is the loss function of the student's knowledge state i is the i +1 row knowledge state vector of the student, is the

[0097] th row knowledge state vector of the student;

[0098] A3. Calculate the first item of the loss function model for the student to answer questions, which is expressed as:

[0099] ​​​Wherein: is the trace of the matrix, is the transpose symbol;

[0100] A4. Obtain the loss function result of the student's answer to the question according to the updated meta-knowledge dictionary in step A1, the updated knowledge state of the student in step A2, and the first item of the loss function model of the student's answer to the question.

[0101] According to the loss function result of the student's answer to the question, the present invention updates the meta-knowledge dictionary and the knowledge state of the student by adopting a coordinate descent strategy, including the following steps:

[0102] B1. Determine the meta-knowledge dictionary and the knowledge state of the student according to the loss function result of the student's answer to the question, expressed as:

[0103] ,

[0104] ,

[0105] ,

[0106] ,

[0107] ;

[0108] Wherein: is the knowledge state of the student at time t, is a function for returning the index position of the minimum element in the array, is the reconstruction error between the student's answer data matrix and the dictionary representation , is the student's answer data matrix at time t, is the dictionary matrix at time t-1, is the knowledge state of the student, is the Frobenius norm square loss function, is the sparsity regularization parameter, is the temporal smoothness regularization parameter, is the knowledge state of the student at time t-1, is the dictionary matrix at time t, is the trace of the matrix, is the dictionary matrix, , , is the student's answer data matrix, is the transpose symbol, is the balancing weight, is the objective function symbol;

[0109] B2. Update the knowledge dictionary and the student's knowledge state in step B1 using the coordinate descent strategy, expressed as:

[0110] ,

[0111] ;

[0112] Where: is the value of the i-th column of the student's knowledge state at time t, is the value of the i-th column of the student's knowledge state at time t - 1, is the outer product matrix of the knowledge dictionary at time t - 1, is the matrix product of the knowledge dictionary and the student's answer data matrix at time t - 1, is the value of the j-th column of the dictionary matrix at time t, is the value of the j-th column of the dictionary matrix at time t - 1, is the value of the j-th row and j-th column of at time t, is the value of the j-th column of at time t, is the value of the j-th column of the transpose of at time t.

[0113] Specifically, in order for the knowledge increment to be represented in terms of expert concepts, the present invention assumes that expert concepts can be linearly combined by a subset of meta-knowledge. The present invention defines the relationship matrix between items and expert concepts, and then obtains the knowledge dictionary and the student's knowledge state in terms of expert concepts, expressed as:

[0114] ,

[0115] ,

[0116] ;

[0117] Where: is the knowledge dictionary in terms of expert concepts, is the relationship matrix between items and expert concepts, is the student's knowledge state in terms of expert concepts.

[0118] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or means for implementing the functions specified in one block or multiple blocks.

[0119] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing devices to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or means for implementing the functions specified in one block or multiple blocks.

[0120] These computer program instructions can also be loaded onto a computer or other programmable data processing devices, such that a series of operation steps are executed on the computer or other programmable devices to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable devices provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or means for implementing the functions specified in one block or multiple blocks.

[0121] Specific embodiments are applied in the present invention to elaborate on the principles and implementation manners of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present invention.

[0122] Those of ordinary skill in the art will realize that the embodiments described herein are for helping readers understand the principles of the present invention, and it should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those of ordinary skill in the art can make various other specific deformations and combinations that do not depart from the essence of the present invention based on the technical revelations disclosed in the present invention, and these deformations and combinations are still within the protection scope of the present invention.

Claims

1. A method for dynamically tracking student knowledge changes, characterized in that: The following steps are involved: S1. Construct a student's knowledge status model based on the independent incremental characteristics of student knowledge growth; S2. Build a meta-knowledge dictionary based on the knowledge points of the learning project, and build a Bayesian framework for students to answer questions based on the meta-knowledge dictionary and the student's knowledge state model in step S1 to obtain a loss function model for students to answer questions; S3, according to the student's answer data and the loss function model of the student's answer question in step S2, the alternating direction multiplication method is used to obtain the loss function result of the student's answer question, and according to the loss function result of the student's answer question, the coordinate descent strategy is used to update the meta-knowledge dictionary and the student's knowledge state to dynamically track the change of the student's knowledge state; According to the loss function results of the students’ answers, the coordinate descent strategy is used to update the meta-knowledge dictionary and the students’ knowledge status, including the following steps: B1. According to the loss function result of the student's answer, determine the meta-knowledge dictionary and the student's knowledge state, expressed as: ， ， ， ， in: is the student's knowledge state at time t, is a function that returns the index position of the smallest element in an array. Answer the data matrix for students and dictionary representation The reconstruction error between is the student answer data matrix at time t, is the dictionary matrix at time t-1, The knowledge status of students, is the Frobenius norm square loss function, is the sparsity regularization parameter, is the norm-one-norm loss function, is the temporal smoothness regularization parameter, is the student’s knowledge state at time t-1, is the dictionary matrix at time t, is the trace of the matrix, is the dictionary matrix, is the Gram matrix of the student’s knowledge state at time t, is the Gram matrix of the student’s knowledge state at time t-1, is the product of the student’s knowledge state at time t and the student’s answer data matrix, is the product of the student’s knowledge state at time t-1 and the student’s answer data matrix, For the student answer data matrix, is the transpose symbol, To balance the weight, is the symbol of the objective function, is the transposed matrix of the student’s knowledge state at time t; B2. Use the coordinate descent strategy to update the knowledge dictionary and the student's knowledge status in step B1, expressed as: ， in: is the value of the i-th column of the student’s knowledge state at time t, is the value of the i-th column of the student’s knowledge status at time t-1, is the outer product matrix of the knowledge dictionary at time t-1, is the matrix product of the knowledge dictionary and the student answer data matrix at time t-1, is the value of the jth column of the dictionary matrix at time t, is the value of the jth column of the dictionary matrix at time t-1, for The value of the jth row and jth column at time t, for The value of the jth column at time t, for The value of the j-th column of the transpose at time t.

2. The method for dynamically tracking student knowledge changes according to claim 1, characterized in that: In step S1, the student's knowledge state model is constructed based on the independent incremental characteristics of the student's knowledge growth, which is expressed as: ， ， in: is the student's knowledge state at time t, is the student's knowledge state at time s, is a normal distribution with mean 0 and variance ts, The knowledge status of students, is the incremental distribution satisfied by the student’s knowledge state, is the linear transformation matrix, is the normal distribution variance of X, X is the increment of students’ knowledge growth, is the transpose symbol, For students' knowledge status Transfer to The probability density function of is the exponential function symbol, For students at all times The knowledge state vector, For students at all times The knowledge state vector, For the time series i time, For the time series i +1 moment.

3. The method for dynamically tracking student knowledge changes according to claim 1, characterized in that: In step S2, a meta-knowledge dictionary is constructed based on the knowledge points of the learning project. The specific process is: the knowledge points of the learning project are divided into minimum knowledge point units, the minimum knowledge point units are indivisible and only include one state of mastering or not mastering, and their linear combination is used to represent knowledge entities, the minimum knowledge point units are numbered to construct a corresponding matrix to construct a meta-knowledge dictionary.

4. The method for dynamically tracking student knowledge changes according to claim 1, characterized in that: In step S2, based on the meta-knowledge dictionary and the student's knowledge state model in step S1, a Bayesian framework for students to answer questions is constructed, which is expressed as: ， ， ， in: For a given dictionary matrix and students’ knowledge status , the students' answers to the questions y follow the mean , the variance is The normal distribution of The mean is , the variance is The normal distribution of is the dictionary matrix No. i List, The mean is 0 and the variance is The normal distribution of For students i row knowledge state vector, is a Laplace distribution with mean 0 and scale parameter b, is a normal distribution with mean 0 and variance ts, For students at all times The knowledge state vector, For students at all times The knowledge state vector, For the time series i time, For the time series i +1 moment.

5. The method for dynamically tracking student knowledge changes according to claim 1, characterized in that: In step S2, the loss function model of the student answering the question is obtained, which is expressed as: in: is the loss function for students answering questions, To obtain the minimum value, Representation Knowledge Dictionary and the state of knowledge Student answer data matrix The reconstruction error, is the Frobenius norm square loss function, For the student answer data matrix, is the dictionary matrix, The knowledge status of students, is the number of students, For tracking items, For students i +1 row knowledge state vector, For students i row knowledge state vector, is the dictionary matrix No. i OK.

6. The method for dynamically tracking student knowledge changes according to claim 1, characterized in that: In step S3, according to the student's answer data and the loss function model of the student's answer question in step S2, the alternating direction multiplication method is used to obtain the loss function result of the student's answer question, including the following steps: A1. Using the alternating direction multiplier method, the meta-knowledge dictionary is updated with the student's knowledge state fixed, expressed as: in: is the dictionary matrix The loss function is Representation Knowledge Dictionary and the state of knowledge Student answer data matrix The reconstruction error, is the Frobenius norm square loss function, For the student answer data matrix, is the dictionary matrix, The knowledge status of students, is the number of students, is the dictionary matrix No. i OK; A2. Using the alternating direction multiplier method, the student's knowledge state is updated under the condition of a fixed meta-knowledge dictionary, which is expressed as: in: The state of knowledge for students The loss function is For students i +1 row knowledge state vector, For students i row knowledge state vector; A3. Calculate the first term of the loss function model for students’ answers, expressed as: in: is the trace of the matrix, is the transpose symbol; A4. According to the updated meta-knowledge dictionary in step A1, the updated student's knowledge status in step A2, and the first item of the loss function model of the student's answer to the question in step A3, obtain the loss function result of the student's answer to the question.

Images