Knowledge tracking method and system based on dual-view consistency evaluation and noise filtering

Abstract

The invention provides a novel knowledge tracking method and system based on double-view-angle consistency evaluation and noise filtering, and the method comprises the steps: removing noise information in a student interaction sequence from an exercise view angle and a student view angle; according to the method, noise information in an interaction sequence is removed by estimating the consistency degree of answering performance between students through a collaborative consistency measurement attention mechanism, and noise information is removed by capturing the consistency of the answering performance and historical learning performance through a time-aware multi-head self-attention mechanism provided by the latter. By introducing a denoising mechanism, the model can more accurately capture the knowledge state change of the student, so that more targeted learning feedback and guidance are provided for the student, and the development of personalized education is promoted. Besides, by utilizing the interactive sequence representation denoised from the two visual angles, a contrast learning mechanism of denoising information mutual supervision is designed to further enhance the denoising effectiveness of the interactive sequence representation, and integration of knowledge tracking models is performed to realize prediction of future performance of students.

Description

Technical Field

[0001] The present invention relates to the field of computer data processing technology, and in particular to a knowledge tracking method and system based on dual-view consistency evaluation and noise filtering. Background Art

[0002] Knowledge tracking is a key task in online intelligent education systems. Its core goal is to track the changes in students' knowledge status through their historical answer records, thereby providing them with personalized learning paths and learning resources to improve learning efficiency and effectiveness. Most current knowledge tracking models model the impact of various factors on students' knowledge status from different perspectives, thereby optimizing the network structure and improving the model's prediction performance. However, these studies often ignore data noise (i.e., abnormal interactions) in interaction sequences. For example, students’ performance in answering questions may be affected by external factors such as fatigue, guesswork, misoperation, and the quality of exercises, and may not always accurately reflect their true knowledge level. Due to the presence of noisy data, the model is prone to capture irrelevant information during the learning process, resulting in learning bias in sequence representation. This bias not only affects the accurate assessment of students’ knowledge status, but also further reduces the accuracy of the model’s prediction of students’ future performance.

[0003] Therefore, how to effectively process the noise signals in the interaction sequence, enhance the signal-to-noise ratio of the sequence representation, and improve the model robustness plays an important role in improving the evaluation accuracy of the knowledge tracking model. Summary of the invention

[0004] The main purpose of the present invention is to provide a knowledge tracking method and system based on dual-view consistency evaluation and noise filtering to solve the above technical problems.

[0005] To achieve the above object, the present invention provides a knowledge tracking method based on dual-view consistency evaluation and noise filtering, comprising the steps of: S1, build an interaction record query set based on learning interaction data , Indicates students Learning records, Indicates the number of students; S2, query set based on interaction records use The network is pre-trained to obtain the knowledge state evaluation model: ; S3, estimates the noise signal in the interaction sequence through the complex similarity information between students, and calculates the collaborative information to estimate the students About the exercises Response consistency information ; S4, based on historical interaction records, obtains a global evolutionary consistent projection representation and interaction representation to estimate the student's understanding of the exercise Evolutionary consistency probability of answer performance ; S5, denoising mutual supervision based on contrastive learning, including: S51, based on the denoising method in step S2, obtain the learning sequence interaction representation of the exercise perspective denoising , Based on the denoising method in step S3, the interactive representation of the learning sequence denoised by the student perspective is obtained , ; S52, yes and Conduct comparative learning for students The learning record at time step t represents the contrast loss calculated from a single batch, in the student Construct positive and negative pairs in the current batch, at time step The contrast loss is , ,in, is the temperature hyperparameter, is the cosine similarity function; S53, two interactive representations of the student's knowledge state obtained by denoising the sequence at time step t and Make the connection and input the connection result into the classifier to predict the student's correct answer to the exercise Probability , ,in, It is a knowledge tracking model.

[0006] Furthermore, the step S3 includes: S31, for students Interaction records to be denoised , based on its historical learning records , evaluate the student at time step Previous state of knowledge S32, from the query set Get about Cooperative Sequence Set , evaluate the model through the pre-trained knowledge state get Middle school students answering exercises Previous state of knowledge , ,in, Indicates that you are answering questions The knowledge state between and The similarity from Filter the most similar records, as The collaborative knowledge state information matrix ; S33, design collaborative consistency measurement attention mechanism to capture other students’ answers With students answer complex collaborative information to obtain similarity projection representation, , ,in, is corresponding to Other students answered Previous state of knowledge, Are a student answer Previous state of knowledge, is corresponding to Other students about The interactive embedding representation, represents the similarity projection representation, is the dimension of the embedding representation; S34, estimating student About the exercises Response consistency information , , is the weight matrix, is the bias term, Are a student about Embedded representation of interactions.

[0007] Furthermore, in step S31, the knowledge state is calculated by the following formula: : , , , , in, , , and is the weight matrix, and is the bias term, is the activation function, and is the weight matrix, is the bias term, is the activation function.

[0008] Furthermore, in step S32, the collaborative sequence Meet two conditions: ① The last item is students about Exercises Learning records; is a learning record generated by other students, .

[0009] Furthermore, the step S4 includes: S41, introduces sequence position encoding to capture learning sequence interaction representation location information; S42, for time step ( Encode to obtain the position encoding vector , Indicates the maximum length of the sequence, and encodes the position vector Interaction Perform additive fusion to obtain a new interactive representation . ; S43, using attention mechanism to map historical interaction representation Similarity projection representation , , ,in and represents the historical interaction representation, is the dimension of the embedding representation; S44, captured in different subspaces The similarity projection representation of the historical interaction sequence information is expressed as , , ,in, denote the projection weights of the query matrix, key matrix, and value matrix of the i-th attention head, respectively. is the weight of the attention head; S45, fusion based on historical interaction records to obtain global evolutionary consistent projection representation and interaction representation, estimate the student's knowledge of the exercises Evolutionary consistency probability of answer performance , ,in, is the weight matrix, .

[0010] Furthermore, the express , , , one of the Transformer modeling models.

[0011] The present invention also provides a knowledge tracking system based on dual-perspective consistency evaluation and noise filtering, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the knowledge tracking method based on dual-perspective consistency evaluation and noise filtering as described in any one of the above items are implemented.

[0012] The present invention also provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the steps of the knowledge tracking method based on dual-view consistency evaluation and noise filtering as described in any one of the above items are implemented: This application proposes a knowledge tracking method based on dual-perspective consistency evaluation and noise filtering, which removes noise information in student interaction sequences from the perspective of exercises and students respectively. The former removes noise information in the interaction sequence by estimating the consistency of the answer performance between students through the collaborative consistency measurement attention mechanism, and the latter proposes a time-aware multi-head self-attention mechanism to capture the consistency between the answer performance and the historical learning performance to remove noise information. By introducing a denoising mechanism, the model can more accurately capture the changes in students' knowledge status, thereby providing students with more targeted learning feedback and guidance, and promoting the development of personalized education. In addition, using the interaction sequence representation after denoising from two perspectives, a comparative learning mechanism with mutual supervision of denoising information is designed to further enhance the effectiveness of denoising the interaction sequence representation, and integrate the knowledge tracking model to achieve prediction of students' future performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The accompanying drawings are part of the present invention and are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but do not constitute an improper limitation of the present invention. Obviously, the drawings described below are only some embodiments. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the accompanying drawings: Figure 1 Schematic diagram of the flow of a knowledge tracking method based on dual-view consistency evaluation and noise filtering in one embodiment of the present invention.

[0014] Figure 2 The figure is a schematic diagram of the hardware structure for running the knowledge tracking method based on dual-view consistency evaluation and noise filtering in one embodiment of the present invention.

[0015] Figure 34 is a logical structure diagram of a knowledge tracking system based on dual-view consistency evaluation and noise filtering in one embodiment of the present invention.

[0016] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0017] The following is a clear and complete description of the technical problems solved by the embodiments of the present invention, the technical solutions adopted, and the technical effects achieved in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other equivalent or obviously modified embodiments obtained by ordinary technicians in this field without creative work fall within the scope of protection of the present invention. The embodiments of the present invention can be concretized in a variety of different ways as defined and covered in the claims.

[0018] It should be noted that in the following description, many specific details are given for ease of understanding, but it is obvious that the present invention can be implemented without these specific details.

[0019] It should be noted that, in the absence of clear limitation or conflict, the various embodiments of the present invention and the technical features therein can be combined with each other to form a technical solution.

[0020] Please refer to Figure 1-Figure 3 ,The present invention proposes a knowledge tracking method based on dual-view consistency evaluation and noise filtering, including steps S1-S5.

[0021] S1, build an interaction record query set based on learning interaction data , Indicates students Learning records, Indicates the number of students; S2, query set based on interaction records use The network is pre-trained to obtain the knowledge state evaluation model: .

[0022] Specifically, we can build an interaction record query set based on the learning interaction data of the online intelligent education platform. . Can be expressed , , One of the sequence modeling models such as Transformer.

[0023] S3, estimates the noise signal in the interaction sequence through the complex similarity information between students, and calculates the collaborative information to estimate the students About the exercises Response consistency information .

[0024] Specifically, the consistency of ability levels among students often leads to convergence of answer performance. Utilizing this collaborative information among students can effectively estimate the noise signal in the interaction sequence.

[0025] Furthermore, the step S3 includes: S31, for students Interaction records to be denoised , based on its historical learning records , evaluate the student at time step Previous state of knowledge S32, from the query set Get about Cooperative Sequence Set , evaluate the model through the pre-trained knowledge state get Middle school students answering exercises Previous state of knowledge , ,in, Indicates that you are answering questions The knowledge state between and The similarity from Filter the most similar records, as The collaborative knowledge state information matrix ; S33, design collaborative consistency measurement attention mechanism to capture other students’ answers With students answer complex collaborative information to obtain similarity projection representation, , ,in, is corresponding to Other students answered Previous state of knowledge, Are a student answer Previous state of knowledge, is corresponding to Other students about The interactive embedding representation, represents the similarity projection representation, is the dimension of the embedding representation; S34, estimating student About the exercises Response consistency information , , is the weight matrix, is the bias term, Are a student about Embedded representation of interactions.

[0026] Specifically, using Time step input and the student's knowledge state at the previous time step Perform nonlinear combination to obtain candidate knowledge states: , then use the update gate To balance the information introduced from the previous time step knowledge state and the candidate knowledge state, the knowledge state of the current time step is obtained .

[0027] Furthermore, in step S31, the knowledge state is calculated by the following formula: : , , , , in, , , and is the weight matrix, and is the bias term, is the activation function, and is the weight matrix, is the bias term, is the activation function.

[0028] Furthermore, in step S32, the collaborative sequence Meet two conditions: ① The last item is students about Exercises Learning records; is a learning record generated by other students, .

[0029] S4, based on historical interaction records, obtains a global evolutionary consistent projection representation and interaction representation to estimate the student's understanding of the exercise Evolutionary consistency probability of answer performance .

[0030] Specifically, as the time distance increases, the consistency between students’ answer performance and historical learning performance will decay. In order to capture the position information in the learning interaction sequence, sequence position encoding is introduced.

[0031] Furthermore, the step S4 includes: S41, introduces sequence position encoding to capture interactive representations of learning sequences location information; S42, for time step ( Encode to obtain the position encoding vector , Indicates the maximum length of the sequence, and encodes the position vector Interaction Perform additive fusion to obtain a new interactive representation . ; S43, using attention mechanism to map historical interaction representation Similarity projection representation , , ,in and represents the historical interaction representation, is the dimension of the embedding representation; S44, captured in different subspaces The similarity projection representation of the historical interaction sequence information is expressed as , , ,in, denote the projection weights of the query matrix, key matrix, and value matrix of the i-th attention head, respectively. is the weight of the attention head; S45, fusion based on historical interaction records to obtain global evolutionary consistent projection representation and interaction representation, estimate the student's knowledge of the exercises Evolutionary consistency probability of answer performance , ,in, is the weight matrix, .

[0032] S5, denoising mutual supervision based on contrastive learning, including: S51, based on the denoising method in step S2, obtain the learning sequence interaction representation of the exercise perspective denoising , Based on the denoising method in step S3, the interactive representation of the learning sequence denoised by the student perspective is obtained , ; S52, yes and Conduct comparative learning for students The learning record at time step t represents the contrast loss calculated from a single batch, in the student Construct positive and negative pairs in the current batch, at time step The contrast loss is , ,in, is the temperature hyperparameter, is the cosine similarity function; S53, two interactive representations of the student's knowledge state obtained by denoising the sequence at time step t and Make the connection and input the connection result into the classifier to predict the student's correct answer to the exercise Probability , ,in, It is a knowledge tracking model.

[0033] The present invention also provides a knowledge tracking system based on dual-perspective consistency evaluation and noise filtering, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the knowledge tracking method based on dual-perspective consistency evaluation and noise filtering as described in any one of the above items are implemented.

[0034] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the knowledge tracking method based on dual-view consistency evaluation and noise filtering as described in any one of the above items.

[0035] Figure 2 FIG. 1 is a schematic diagram of a hardware structure for running a knowledge tracking method based on dual-view consistency evaluation and noise filtering provided by an embodiment of the present invention. Figure 2As shown, the embodiment / computer 6 includes: a processor 60, a memory 61, and a computer program 62 stored in the memory 61 and executable on the processor 60, such as a program for executing a knowledge tracking method based on dual-view consistency evaluation and noise filtering. When the processor 60 executes the computer program 62, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor 60 executes the computer program 62, the functions of the modules / units in the above-mentioned device embodiments are implemented.

[0036] Exemplarily, the computer program 62 may be divided into one or more modules / units, which are stored in the memory 61 and executed by the processor 60 to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program 62 in the computer 6.

[0037] The computer 6 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The computer 6 may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will appreciate that Figure 2 It is only an example of computer 6 and does not constitute a limitation of computer 6. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the computer 6 may also include input and output devices, network access devices, buses, etc.

[0038] The processor 60 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0039] The memory 61 may be an internal storage unit of the computer 6, such as a hard disk or memory of the computer 6. The memory 61 may also be an external storage device of the terminal device / computer 6, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device. Further, the memory 61 may also include both an internal storage unit of the computer 6 and an external storage device. The memory 61 is used to store the computer program and other programs and data required by the terminal device. The memory 61 may also be used to temporarily store data that has been output or is to be output.

[0040] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0041] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0042] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0043] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminal equipment and methods can be implemented in other ways. For example, the device / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0044] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0045] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0046] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0047] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A knowledge tracking method based on dual-view consistency evaluation and noise filtering, characterized in that: Includes steps: S1, build an interaction record query set based on learning interaction data , Indicates students Learning records, Indicates the number of students; S2, query set based on interaction records use The network is pre-trained to obtain the knowledge state evaluation model: ; S3, estimates the noise signal in the interaction sequence through the complex similarity information between students, and calculates the collaborative information to estimate the students About the exercises Response consistency information ; S4, based on historical interaction records, obtains a global evolutionary consistent projection representation and interaction representation to estimate the student's understanding of the exercise Evolutionary consistency probability of answer performance ; S5, denoising mutual supervision based on contrastive learning, including: S51, based on the denoising method in step S2, obtain the learning sequence interaction representation of the exercise perspective denoising , Based on the denoising method in step S3, the interactive representation of the learning sequence denoised by the student perspective is obtained , ; S52, yes and Conduct comparative learning for students The learning record at time step t represents the contrast loss calculated from a single batch, in the student Construct positive and negative pairs in the current batch, at time step The contrast loss is , ,in, is the temperature hyperparameter, is the cosine similarity function; S53, two interactive representations of the student's knowledge state obtained by denoising the sequence at time step t and Make the connection and input the connection result into the classifier to predict the student's correct answer to the exercise Probability , ,in, It is a knowledge tracking model.

2. The knowledge tracking method based on dual-view consistency evaluation and noise filtering according to claim 1 is characterized in that: The step S3 comprises: S31, for students Interaction records to be denoised , based on its historical learning records , evaluate the student at time step Previous state of knowledge S32, from the query set Get about Cooperative Sequence Set , evaluate the model through the pre-trained knowledge state get Middle school students answering exercises Previous state of knowledge , ,in, Indicates that you are answering questions The knowledge state between and The similarity from Filter the most similar records, as The collaborative knowledge state information matrix ; S33, design collaborative consistency measurement attention mechanism to capture other students’ answers With students answer complex collaborative information to obtain similarity projection representation, , ,in, is corresponding to Other students answered Previous state of knowledge, Are a student answer Previous state of knowledge, is corresponding to Other students about The interactive embedding representation, represents the similarity projection representation, is the dimension of the embedding representation; S34, estimating student About the exercises Response consistency information , , is the weight matrix, is the bias term, Are a student about Embedded representation of interactions.

3. The knowledge tracking method based on dual-view consistency evaluation and noise filtering according to claim 2 is characterized in that: In step S31, the knowledge state is calculated by the following formula: : ， ， ， ， in, , , and is the weight matrix, and is the bias term, is the activation function, and is the weight matrix, is the bias term, is the activation function.

4. The knowledge tracking method based on dual-view consistency evaluation and noise filtering according to claim 2 is characterized in that: In step S32, the coordination sequence Meet two conditions: ① The last item is students about Exercises Learning records; is a learning record generated by other students, .

5. The knowledge tracking method based on dual-view consistency evaluation and noise filtering according to claim 1 is characterized in that: The step S4 comprises: S41, introduces sequence position encoding to capture learning sequence interaction representation location information; S42, for time step ( Encode to obtain the position encoding vector , Indicates the maximum length of the sequence, and encodes the position vector Interaction Perform additive fusion to obtain a new interactive representation . ; S43, using attention mechanism to map historical interaction representation Similarity projection representation , , ,in and represents the historical interaction representation, is the dimension of the embedding representation; S44, captured in different subspaces The similarity projection representation of the historical interaction sequence information is expressed as , , ,in, denote the projection weights of the query matrix, key matrix, and value matrix of the i-th attention head, respectively. is the weight of the attention head; S45, fusion based on historical interaction records to obtain global evolutionary consistent projection representation and interaction representation, estimate the student's knowledge of the exercises Evolutionary consistency probability of answer performance , ,in, is the weight matrix, is the bias term.

6. The knowledge tracking method based on dual-view consistency evaluation and noise filtering according to claim 1 is characterized in that: Said express , , , one of the Transformer modeling models.

7. A knowledge tracking system based on dual-view consistency evaluation and noise filtering, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the knowledge tracking method based on dual-view consistency evaluation and noise filtering are implemented as described in any one of claims 1 to 6.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the knowledge tracking method based on dual-view consistency evaluation and noise filtering are implemented as described in any one of claims 1 to 6.

Images