Knowledge tracking method and system for cross-domain knowledge migration

Abstract

The invention discloses a knowledge tracking method and system for cross-domain knowledge migration, and the method comprises the steps: constructing a student learning behavior representation knowledge base, and obtaining a knowledge state vector based on question-level-perceived high-dimensional vector representation extracted from the student learning behavior representation knowledge base and joint question-answer interaction representation; constructing an autoregression information processing module to optimize the knowledge state vector to obtain an optimized knowledge state vector; and constructing a student answering prediction module based on a two-layer nonlinear full-connection network, inputting the optimized knowledge state vector into the student answering prediction module, obtaining a student answering prediction result, and realizing knowledge tracking. According to the method, by generating the representation knowledge base with domain adaptability, the model can migrate knowledge in a data-rich field to a data-scarce field, and the performance problem of an existing method in the data-scarce field is solved; through the adaptive knowledge state extraction network module, the sensitivity of specific hyper-parameter selection is reduced, and the resource computing power is greatly saved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent education technology, and in particular relates to a knowledge tracking method and system for cross-domain knowledge transfer. Background Art

[0002] As a key technology in intelligent education and personalized learning systems, knowledge tracking technology predicts students' mastery of new knowledge points by analyzing students' historical learning behaviors converted into digital data. This technology can dynamically monitor changes in students' knowledge status and provide a basis for planning personalized learning paths. Based on the prediction results, the knowledge tracking system can specifically recommend learning content that needs to be strengthened to students to avoid repeating the knowledge points that have been mastered. Teachers can also use this technology to understand students' learning status in real time, identify problems in a timely manner and adjust teaching strategies. In addition, by providing targeted learning support for students at different levels, knowledge tracking technology helps narrow the learning gap and promote the realization of educational equity.

[0003] Although existing knowledge tracking methods (such as DKT, AKT, simpleKT, etc.) have made significant progress in performance, they generally have two key problems: 1. Model performance depends largely on the quantity and quality of training data. However, due to the impact of privacy protection, it is not possible to obtain enough data in all disciplines. For example, in some fields, there are only one hundred student interaction sequences, while in some fields, there may be hundreds of thousands of interaction sequences. 2. The performance of the model is heavily dependent on the hyperparameter combination. For example, in one hyperparameter combination, the model performance is 0.56, while in another hyperparameter combination, the model effect can reach 0.75. These two problems restrict the possibility of using knowledge tracking methods to achieve large-scale personalized education in actual educational scenarios. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a knowledge tracking method and system for cross-domain knowledge transfer to solve the problems existing in the above prior art.

[0005] To achieve the above object, the present invention provides a knowledge tracking method for cross-domain knowledge transfer, comprising:

[0006] Constructing a student learning behavior representation knowledge base, wherein the student learning behavior representation knowledge base includes a question-level high-dimensional vector representation base and a joint question-answer interaction representation base;

[0007] Obtaining a knowledge state vector based on a high-dimensional vector representation of question-level perception extracted from the student learning behavior representation knowledge base and a joint question-answer interaction representation;

[0008] Constructing an autoregressive information processing module, and optimizing the knowledge state vector based on the autoregressive information processing module; obtaining an optimized knowledge state vector;

[0009] A student answer prediction module is constructed based on a two-layer nonlinear fully connected network, and the optimized knowledge state vector is input into the student answer prediction module to obtain the student answer prediction result and realize knowledge tracking.

[0010] Optionally, the process of building a problem-level high-dimensional vector representation library includes:

[0011] Feature mapping is performed on the question sequence to obtain a high-dimensional vector representation of the problem level perception, and a problem-level high-dimensional vector representation library is constructed based on the high-dimensional vector representation of the problem level perception. The process of obtaining the high-dimensional vector representation of the problem level perception is as follows:

[0012]

[0013] In the formula, is the level of questions that can be trained, is the average feature representation of all knowledge points corresponding to the question, Represents bitwise addition.

[0014] Optionally, the process of building a joint question-answer interaction representation library includes:

[0015] Based on the student answer sequence, the embedding layer of the answer behavior representation extractor is used to model the high-order interaction characteristics between the question knowledge points and the student answers, generate a joint question-answer interaction representation, and build a joint question-answer interaction representation library based on the joint question-answer interaction representation. The process of obtaining the joint question-answer interaction representation is as follows:

[0016]

[0017] In the formula, is the level of questions that can be trained, is the average feature representation of all knowledge points corresponding to the question, represents bitwise addition, Indicates the unique hot encoding of the knowledge points corresponding to the joint students' corresponding questions.

[0018] Optionally, extract the high-dimensional vector representation of problem level perception from the problem level high-dimensional vector representation library, and extract the joint question-answer interaction representation from the joint question-answer interaction representation library; use the high-dimensional vector representation of problem level perception and the obtained query vector and joint question-answer interaction representation as the query vector, key vector and value vector, use the scaled dot product attention mechanism to process the query vector and key vector to obtain the attention score, convert them into probability distribution through the Softmax function, perform weighted summation on the value vector according to the probability distribution, and obtain the knowledge state vector. The process is expressed as follows:

[0019]

[0020] h t+1 =Concat(Head1,Head2,…,Head n )

[0021] In the formula, Head represents the multi-head representation in the multi-head attention mechanism, Softmax represents the softmax function, represents the transpose of the key vector, represents the scaling factor of the scaled dot product attention, d is the dimension of the key vector, Q is the query vector, and K is the key vector.

[0022] Optionally, the autoregressive information processing module includes several autoregressive information processing units, each of which includes a knowledge state extraction module, a double normalization layer, a nonlinear transformation network layer and an adaptive residual connection; the knowledge state extraction module, the normalization layer, the residual connection, the nonlinear transformation network layer and the normalization layer are connected in sequence.

[0023] Optionally, the student answer prediction module is as follows:

[0024]

[0025] Among them, h l t+1 represents the optimized knowledge state vector, σ represents the Sigmoid function, b1, b2, b, W1, W2, W are learnable parameters, Represents the predicted result of the student’s answer, and the historical learning behavior includes T interactive processes, t∈T.

[0026] Optionally, it also includes: the student answer prediction module is optimized through a binary cross entropy loss function.

[0027] The present invention also provides a knowledge tracking system for cross-domain knowledge migration, comprising:

[0028] Student learning behavior representation and cross-domain transfer module, used to generate a student behavior representation knowledge base;

[0029] A knowledge state extraction module, for obtaining a knowledge state vector based on a high-dimensional vector representation of question-level perception extracted from the student learning behavior representation knowledge base and a joint question-answer interaction representation by using a multi-head self-attention mechanism and a scaled dot-product attention mechanism;

[0030] An autoregressive information processing module is used to construct a multi-layer cascade dynamic information interaction network, optimize the knowledge state vector, and obtain an optimized knowledge state vector;

[0031] The student answer prediction module is used to perform a high-dimensional to low-dimensional projection transformation on the optimized knowledge state vector, optimize parameters based on the binary cross entropy loss function, and finally output the prediction results.

[0032] Compared with the prior art, the present invention has the following advantages and technical effects:

[0033] The present invention constructs a multi-dimensional feature embedding space through the representation of student learning behavior and the cross-domain migration module, realizes cross-domain knowledge migration, and solves the performance problem of existing methods in data-scarce fields. By generating a representation knowledge base with domain adaptability, the model can migrate knowledge from data-rich fields to data-scarce fields; through the knowledge state extraction module and the autoregressive information processing module, it reduces the sensitivity to the selection of specific hyperparameters, greatly saving resources and computing power. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0035] Figure 1 It is a schematic diagram of the technical solution of an embodiment of the present invention. DETAILED DESCRIPTION

[0036] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0037] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0038] Embodiment 1

[0039] like Figure 1 As shown, this embodiment provides a knowledge tracking method for cross-domain knowledge migration, including:

[0040] Constructing a student learning behavior representation knowledge base, wherein the student learning behavior representation knowledge base includes a question-level high-dimensional vector representation base and a joint question-answer interaction representation base;

[0041] Specifically, for student S, the historical learning behavior includes T interaction processes, each of which is expressed as t ,{c t}, r t >, where t∈T. Specifically, q t represents the question answered by the student at time step t, c t represents the question q answered by the student at time step t t The corresponding knowledge point. If a question corresponds to multiple knowledge points, then it is the set of multiple knowledge points {c t}, r t Indicates that the student answers question q at time step t t The answer is r t =1 means the student answered correctly, r t = 0 means the student answered incorrectly. For the questions answered by the students, the problem behavior representation extractor E(·) is used to perform feature mapping on the question sequence to obtain a high-dimensional vector representation of the problem level perception:

[0042]

[0043] in, is the level of questions that can be trained, is the average feature representation of all knowledge points corresponding to the question, Represents bitwise addition.

[0044] For students’ answers, the answer behavior representation extractor is used to model the high-order interaction characteristics between question knowledge points and students’ answers through the embedding layer, thereby generating a joint question-answer interaction representation:

[0045]

[0046] in Indicates the unique hot encoding of the knowledge points corresponding to the joint students' corresponding questions.

[0047] For the transferability of students' learning behaviors, we use the cross-domain transfer representation extractor to construct a knowledge base of student learning behavior representations based on the high-dimensional vector representations of question-level perception obtained by the question behavior representation extractor and the answer behavior representation extractor and the joint question-answer interaction representation:

[0048]

[0049] in ​Represents a problem-level high-dimensional vector representation library, represents a joint question-answer interaction representation library, It represents the combination of the corresponding student problem and the characteristic representation of the corresponding subject area.

[0050] The student learning behavior representation and cross-domain transfer module effectively enhances the representation capability of cross-domain feature embedding through the cross-domain transfer representation extractor, question behavior representation extractor and answer behavior representation extractor, providing more accurate feature input for subsequent cross-domain knowledge state extraction.

[0051] Obtaining a knowledge state vector based on a high-dimensional vector representation of question-level perception extracted from the student learning behavior representation knowledge base and a joint question-answer interaction representation;

[0052] Further, a high-dimensional vector representation of problem-level perception is extracted from the problem-level high-dimensional vector representation library, and a joint question-answer interaction representation is extracted from the joint question-answer interaction representation library; the high-dimensional vector representation of problem-level perception and the obtained query vector and the joint question-answer interaction representation are used as the query vector, key vector and value vector, and the query vector and the key vector are processed by the scaled dot product attention mechanism to obtain the attention score, which is converted into a probability distribution through the Softmax function, and the value vector is weighted summed according to the probability distribution to obtain the knowledge state vector. The process is expressed as follows:

[0053]

[0054] h t+1 =Concat(Head1, Head2,..., Head n )

[0055] In the formula, Head represents the multi-head representation in the multi-head attention mechanism, Softmax represents the softmax function, represents the transpose of the key vector, represents the scaling factor of the scaled dot product attention, d is the dimension of the key vector, Q is the query vector, and K is the key vector.

[0056] Specifically, we will use the problem-level high-dimensional vector representation library The question-level-aware high-dimensional vector representation x extracted from t and joint question-answer interaction representation library Extracted y t As query vector (Q), key vector (K) and value vector (V), respectively, the vector x extracted from the question-level high-order vector representation library is used as the query vector and key vector, and the y extracted from the joint question-answer interaction representation library is used as the value vector. That is, the query vector (Q) and the key vector (K) are extracted from Indices, the value vector (V) is from The index is:

[0057] Q = x t+1 ; K = {x1,…,x t}; V = {y1,…,y t}

[0058] A multi-head attention mechanism integrating scaled dot-product attention is used to extract the knowledge state h from the student's historical answer sequence. t+1 :

[0059]

[0060] h t+1 =Concat(Head1, Head2,..., Head n )

[0061] Where Head represents the multi-head representation in the multi-head attention mechanism, Softmax represents the softmax function, represents the transpose of the key vector, represents the scaling factor of the scaled dot product attention, and d is the dimension of the key vector. This embodiment combines the two major advantages of the attention mechanism: one is the characteristics of distributed computing and hierarchical data extraction, and the other is to effectively process the order relationship and context information in the learner's behavior sequence through the scaled dot product mechanism. At the same time, the multi-head attention strategy can obtain deep correlation patterns between data from different dimensions.

[0062] Constructing an autoregressive information processing module, and optimizing the knowledge state vector based on the autoregressive information processing module; obtaining an optimized knowledge state vector;

[0063] Specifically, the module adopts an innovative iterative deep state optimization architecture to achieve layer-by-layer refinement of the knowledge state through the Auto-Regressive Information Processing Module. Each auto-regressive information processing unit contains a knowledge state extraction module, a double normalization layer, a nonlinear transformation network layer, and an adaptive residual connection to form a complete knowledge state optimization loop. This progressive deep architecture can not only capture subtle feature changes during the learning process, but also project the knowledge state into a more discriminative feature space through inter-layer feature reorganization and nonlinear mapping.

[0064] The knowledge state extraction module, normalization layer, residual connection, nonlinear transformation network layer and normalization layer in the autoregressive information processing unit are connected in sequence.

[0065] In addition, the architecture significantly reduces parameter redundancy while maintaining the model's expressiveness through feature reconstruction and dimensionality adaptation mechanisms. The multi-level feature extraction and optimization process enables the model to have stronger generalization capabilities, better adapt to learning tasks of different difficulties and types, and reduce sensitivity to specific hyperparameter selection.

[0066] A student answer prediction module is constructed based on a two-layer nonlinear fully connected network, and the optimized knowledge state vector is input into the student answer prediction module to obtain the student answer prediction result and realize knowledge tracking.

[0067] Furthermore, the student answer prediction module is as follows:

[0068]

[0069] Among them, h l t+1 represents the optimized knowledge state vector, σ represents the Sigmoid function, b1, b2, b, W1, W2 are learnable parameters, Represents the predicted result of the student’s answer, and the historical learning behavior includes T interactive processes, t∈T.

[0070] Furthermore, it also includes: the student answer prediction module is optimized through a binary cross entropy loss function.

[0071] Specifically, this module is based on a two-layer nonlinear fully connected network to realize the knowledge state h l t+1 The high-dimensional to low-dimensional projection of Optimize and dynamically adjust parameters to obtain the final prediction results of students' answers

[0072]

[0073] σ represents the Sigmoid function, b1, b2, b, W1, W2, W are learnable parameters, r t+1 The actual answer results for the students.

[0074] This embodiment also provides a knowledge tracking system for cross-domain knowledge migration, including:

[0075] Student learning behavior representation and cross-domain transfer module, used to generate a student behavior representation knowledge base;

[0076] A knowledge state extraction module, for obtaining a knowledge state vector based on a high-dimensional vector representation of question-level perception extracted from the student learning behavior representation knowledge base and a joint question-answer interaction representation by using a multi-head self-attention mechanism and a scaled dot-product attention mechanism;

[0077] An autoregressive information processing module is used to construct a multi-layer cascade dynamic information interaction network, optimize the knowledge state vector, and obtain an optimized knowledge state vector;

[0078] The student answer prediction module is used to perform a high-dimensional to low-dimensional projection transformation on the optimized knowledge state vector, optimize parameters based on the binary cross entropy loss function, and finally output the prediction results.

[0079] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A knowledge tracking method for cross-domain knowledge transfer, characterized in that: The following steps are involved: Constructing a student learning behavior representation knowledge base, wherein the student learning behavior representation knowledge base includes a question-level high-dimensional vector representation base and a joint question-answer interaction representation base; Obtaining a knowledge state vector based on a high-dimensional vector representation of question-level perception extracted from the student learning behavior representation knowledge base and a joint question-answer interaction representation; Constructing an autoregressive information processing module, and optimizing the knowledge state vector based on the autoregressive information processing module; obtaining an optimized knowledge state vector; A student answer prediction module is constructed based on a two-layer nonlinear fully connected network, and the optimized knowledge state vector is input into the student answer prediction module to obtain the student answer prediction result and realize knowledge tracking.

2. The knowledge tracking method for cross-domain knowledge transfer according to claim 1 is characterized in that: The process of building a problem-level high-dimensional vector representation library includes: Feature mapping is performed on the question sequence to obtain a high-dimensional vector representation of the problem level perception. A problem-level high-dimensional vector representation library is constructed based on the high-dimensional vector representation of the problem level perception. The process of obtaining the high-dimensional vector representation of the problem level perception is as follows: In the formula, is the level of questions that can be trained, is the average feature representation of all knowledge points corresponding to the question, Represents bitwise addition.

3. The knowledge tracking method for cross-domain knowledge transfer according to claim 1 is characterized in that: The process of building a joint question-answer interaction representation library includes: Based on the student answer sequence, the embedding layer of the answer behavior representation extractor is used to model the high-order interaction characteristics between the question knowledge points and the student answers, generate a joint question-answer interaction representation, and build a joint question-answer interaction representation library based on the joint question-answer interaction representation. The process of obtaining the joint question-answer interaction representation is as follows: In the formula, is the level of questions that can be trained, is the average feature representation of all knowledge points corresponding to the question, represents bitwise addition, Indicates the unique hot encoding of the knowledge points corresponding to the joint students' corresponding questions.

4. The knowledge tracking method for cross-domain knowledge transfer according to claim 1 is characterized in that: Extract the high-dimensional vector representation of problem-level perception from the problem-level high-dimensional vector representation library, and extract the joint question-answer interaction representation from the joint question-answer interaction representation library; use the high-dimensional vector representation of problem-level perception and the obtained query vector and joint question-answer interaction representation as the query vector, key vector and value vector, use the scaled dot product attention mechanism to process the query vector and key vector to obtain the attention score, convert them into probability distribution through the Softmax function, and perform weighted summation on the value vector according to the probability distribution to obtain the knowledge state vector. The process is expressed as follows: h t+1 =Concat(Head1,Head2,…,Head n ) In the formula, Head represents the multi-head representation in the multi-head attention mechanism, Softmax represents the softmax function, represents the transpose of the key vector, represents the scaling factor of the scaled dot product attention, d is the dimension of the key vector, Q is the query vector, and K is the key vector.

5. The knowledge tracking method for cross-domain knowledge transfer according to claim 1 is characterized in that: The autoregressive information processing module includes a plurality of autoregressive information processing units, each of which includes a knowledge state extraction module, a double normalization layer, a nonlinear transformation network layer and an adaptive residual connection; the knowledge state extraction module, the normalization layer, the residual connection, the nonlinear transformation network layer and the normalization layer are connected in sequence.

6. The knowledge tracking method for cross-domain knowledge transfer according to claim 1, characterized in that: The student answer prediction module is as follows: Among them, h l t+1 represents the optimized knowledge state vector, σ represents the Sigmoid function, b1, b2, b, W1, W2, W are learnable parameters, Represents the predicted result of the student’s answer, and the historical learning behavior includes T interactive processes, t∈T.

7. The knowledge tracking method for cross-domain knowledge transfer according to claim 1 is characterized in that: Also includes: The student answer prediction module is optimized through a binary cross entropy loss function.

8. A knowledge tracking system for cross-domain knowledge transfer, characterized in that: include: Student learning behavior representation and cross-domain transfer module, used to generate a student behavior representation knowledge base; A knowledge state extraction module, for obtaining a knowledge state vector based on a high-dimensional vector representation of question-level perception extracted from the student learning behavior representation knowledge base and a joint question-answer interaction representation by using a multi-head self-attention mechanism and a scaled dot-product attention mechanism; An autoregressive information processing module is used to construct a multi-layer cascade dynamic information interaction network, optimize the knowledge state vector, and obtain an optimized knowledge state vector; The student answer prediction module is used to perform a high-dimensional to low-dimensional projection transformation on the optimized knowledge state vector, optimize parameters based on the binary cross entropy loss function, and finally output the prediction results.

Images