Online learning concentration degree identification method, system and device and storage medium

Abstract

The invention provides an online learning concentration degree identification method, system and device and a storage medium, and belongs to the field of computer processing, and the method comprises the steps: obtaining a learning image and an ECG signal of a learner; obtaining a plurality of learning features based on the learning image and the ECG signal, and quantifying each learning feature into different levels; setting a level corresponding relationship between the learning features of different levels and the concentration degree; constructing a concentration degree fuzzy fusion model, and calculating to obtain a comprehensive evaluation value of the learner; and obtaining a concentration degree identification result of the learner based on the corresponding relationship between the comprehensive evaluation value and the concentration degree grade. Compared with an existing method, the accuracy of concentration recognition is improved, monitoring of online learning is facilitated, and learning efficiency and quality are guaranteed.

Description

Technical Field

[0001] The present invention belongs to the field of computer processing, and specifically relates to a method, system, device and storage medium for identifying online learning concentration. Background Art

[0002] Concentration is one of the key factors affecting learning efficiency. With the popularization of online education and blended learning models, scientifically and accurately evaluating learners' online learning concentration can help improve teaching quality and learning outcomes.

[0003] With the rapid development of image processing theory and methods, as well as educational technology, facial cues such as eye gaze direction, head posture, and facial features have been widely used to assess concentration in online learning. These representation-based methods, which require no additional equipment beyond image acquisition, have attracted considerable attention from researchers, but their accuracy remains to be improved. Physiological signals, due to their spontaneity and objectivity, offer unique value in attention recognition. Electroencephalogram (EEG) has been shown to be effective for attention monitoring. However, EEG acquisition requires specialized equipment, which has practical limitations and makes it difficult to assess concentration in online learning. In contrast, electrocardiogram (ECG) signals can also provide information about an individual's attention state. Although ECG signals are inherently complex, the widespread availability of smart wristband ECG monitoring devices has provided a foundation for ECG-based attention assessment. ECG signals primarily reflect changes in cardiac activity. While heart rate fluctuations are somewhat correlated with concentration, they can be easily influenced by other physiological and psychological factors, such as anxiety and fatigue. Therefore, single-modality image data or physiological signal data are difficult to fully reflect the learner's concentration, resulting in low accuracy in the learner's concentration analysis. Summary of the Invention

[0004] In order to solve the problem of low accuracy in existing learning concentration recognition, the present invention provides an online learning concentration recognition method, system, device and storage medium.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A method for identifying online learning concentration includes the following steps:

[0007] Acquire a learning image and ECG signal of a learner; obtain a plurality of learning features based on the learning image and ECG signal, and quantify each learning feature into different levels; process the plurality of learning features through a sliding window to obtain a data matrix; wherein the plurality of learning features represent facial features and physiological features of the learner;

[0008] Setting a level correspondence between different levels of learning features and concentration, and converting the data matrix into a fuzzy matrix according to the level correspondence;

[0009] Inputting the fuzzy matrix into a fuzzy fusion model of concentration, calculating the weight relationship of multiple features using a composite proportional evaluation method, calculating the intuitive fuzzy value of the data segment based on the weight relationship, calculating the score of the data segment based on the intuitive fuzzy value scoring function, and calculating the comprehensive evaluation value of the learner based on the score of the data segment;

[0010] Based on the corresponding relationship between the comprehensive evaluation value and the concentration level, the learner's concentration level identification result is obtained according to the learner's comprehensive evaluation value.

[0011] Preferably, the weight relationship of multiple features is calculated by a composite ratio evaluation method, specifically comprising the following steps:

[0012] The average performance value N under the jth feature is calculated using the interval-valued generalized orthogonal fuzzy weighted average neutral aggregation IVq-ROFWANA operator. j , specifically:

[0013]

[0014] in, represents the intuitive fuzzy value corresponding to the jth feature in the i-th data segment, Indicates belonging to r ij The membership degree of Indicates that it does not belong to r ij The non-membership degree of Represents r ij The degree of uncertainty; Θ NA represents the IVq-ROEWANA operator; m represents the total number of data segments, n represents the total number of features, Represents the weight of each data segment when calculating the average performance value of the jth feature, and λ satisfies

[0015] Calculate the variation value φ of the performance value of all data segments for each feature j , specifically:

[0016]

[0017] Among them, s(·) represents the score function; by normalizing the variation value φ j , get the weight w of each feature j , specifically:

[0018]

[0019] According to the weight value of each feature, the weight relationship between multiple features is obtained.

[0020] Preferably, the calculation formula of the score function s(·) is specifically:

[0021]

[0022] Among them, r ij represents the intuitive fuzzy value corresponding to the jth feature in the i-th data segment, Indicates belonging to r ij The membership degree of Indicates that it does not belong to r ij The non-membership degree of Represents r ij degree of uncertainty.

[0023] Preferably, the learning image and ECG signal of the learner are obtained by obtaining the learner's learning video in each period, and using a segmentation tool to segment the video time sequence in the form of one picture per second to obtain the learner's learning image; the ECG signal is specifically extracted by a physiological signal device.

[0024] Preferably, the multiple learning features are specifically head posture features, fatigue features, facial expression features and heart rate change features.

[0025] Preferably, the facial expression features are extracted by a lightweight facial expression recognition FER model using the EfficientNet and MobileNet architectures; the head posture is estimated using the solvePnP algorithm to calculate the pitch angle of the learner's head and the yaw angle of the face; fatigue feature detection includes eye closure detection and yawn detection, and eye closure detection is performed by calculating the eyelid closure percentage PERCLOS, and yawn detection is performed by calculating the mouth closure degree; the heart rate change feature calculates the number of heart beats per minute by identifying the R wave peak.

[0026] The present invention also provides an online learning concentration recognition system, which specifically includes:

[0027] The data acquisition module is used to obtain a learner's learning image and ECG signal; obtain multiple learning features based on the learning image and ECG signal, and quantify each learning feature into different levels; process the multiple learning features through a sliding window to obtain a data matrix; wherein the multiple learning features represent the learner's facial features and physiological features.

[0028] The data processing module is used to set a level correspondence between different levels of learning features and concentration, and convert the data matrix into a fuzzy matrix according to the level correspondence.

[0029] The model processing module is used to input the fuzzy matrix into the concentration fuzzy fusion model, calculate the weight relationship of multiple features through a composite proportional evaluation method, calculate the intuitive fuzzy value of the data segment according to the weight relationship, calculate the score of the data segment based on the intuitive fuzzy value scoring function, and obtain the learner's comprehensive evaluation value according to the score of the data segment.

[0030] The concentration identification module is used to obtain a concentration level identification result of the learner according to the comprehensive evaluation value of the learner based on the correspondence between the comprehensive evaluation value and the concentration level.

[0031] The present invention also provides a computer device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps described in the method for identifying online learning concentration.

[0032] The present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is loaded by a processor, it can execute the steps described in the online learning concentration identification method.

[0033] The online learning concentration recognition method provided by the present invention has the following beneficial effects:

[0034] The present invention obtains multiple learning features by analyzing the learner's learning images and EGG signals, changes the limitations of single modality data in concentration recognition, introduces multiple features that affect concentration analysis, and is conducive to improving the accuracy of identifying the learner's online learning concentration. A hierarchical correspondence between learning features of different levels and concentration is established to reasonably reflect the learner's online learning concentration level. A fuzzy fusion model of concentration is constructed to calculate the learner's comprehensive evaluation value, and the comprehensive evaluation value is converted into a concentration level to obtain the final concentration evaluation result. It fully reflects the learner's concentration level and can quantitatively rank the concentration. It converts subjective attention into objective indicators to improve the efficiency and quality of online learning. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] To more clearly illustrate the embodiments of the present invention and its design, the following briefly introduces the drawings required for this embodiment. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be derived from these drawings without inventive effort.

[0036] Figure 1 This is a flow chart of a method for identifying online learning concentration according to the present invention.

[0037] Figure 2This is a processing flow chart of a method for identifying online learning concentration in an embodiment of the present invention.

[0038] Figure 3 This is a comparison chart of the EFocusOnline model and the expert evaluation confusion matrix in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to enable those skilled in the art to better understand the technical solution of the present invention and to be able to implement it, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solution of the present invention and are not intended to limit the scope of protection of the present invention.

[0040] Example

[0041] The present invention provides a method for identifying online learning concentration. Figure 2 As shown, the specific steps include:

[0042] S1. Obtain the learner's online learning videos, extract static images, and save them in a time-series format. Use the MTCNN model to detect faces and obtain the learner's head posture, fatigue characteristics, and facial expression features. Use the Biosignalsplux multi-channel physiological signal device to extract ECG signals and obtain heart rate (HRV) variation characteristics.

[0043] S2, a lightweight facial expression recognition (FER) model using EfficientNet and MobileNet architectures extracts learners’ facial features and classifies them into seven basic emotions, including: happiness, surprise, neutral, sadness, fear, anger, and disgust.

[0044] The solvePnP algorithm is used to estimate the head posture and calculate the pitch angle and yaw angle of the learner's head.

[0045] Fatigue feature detection includes eye closure detection and yawn detection. Eye closure detection is determined by calculating the percentage of eyelid closure (PERCLOS). This is achieved using 12 feature points within the left and right eye regions. The ratio of the distance between the upper and lower eyelids to the distance between the left and right eye corners is calculated, and the average of the two is the eye closure degree. Yawn detection is determined by calculating the degree of mouth closure using six feature points of the upper and lower lips. Mouth closure is quantified by the ratio of the distance between the upper and lower inner lips to the distance between the mouth corners.

[0046] The collected ECG signal is pre-processed (such as filtering, denoising, etc.), and then the R wave peak is identified, and the number of heart beats per minute is calculated.

[0047] For n feature metrics, we partition the data into multiple segments using a 10-second sliding window and an 8-second overlapping window. Within each 10-second window, we remove outliers and calculate the mean of each feature as the representative value for that window. This results in an m × n data matrix, where each row represents a data segment and each column represents a feature.

[0048] S3. Quantify online learning concentration into five different levels, with concentration levels from "I" to "V" corresponding to five different levels of concentration, from "very high" to "very low." Facial expressions, head posture, fatigue, and heart rate are converted into five different concentration levels, as shown in Table 1. The resulting features correspond to concentration levels.

[0049] Table 1 Classification of concentration levels

[0050]

[0051]

[0052] Each level is converted into a corresponding intuitive fuzzy value to quantify and express the learner's concentration state in online learning. The concentration level and the corresponding intuitive fuzzy value are shown in Table 2.

[0053] Table 2 Conversion of levels into intuitive fuzzy values

[0054]

[0055] The specific values of the n features of each data segment are converted into concentration according to Table 2, and the fuzzy matrix R with a dimension of m×n is obtained. ij ) m×n , r ij Represents the intuitive fuzzy value corresponding to the jth feature in the i-th data segment of the learner.

[0056] S4. Establish a fuzzy fusion model of concentration to calculate the learner's comprehensive evaluation value. The processing steps of the fuzzy fusion model of concentration are as follows:

[0057] S41. Feature weight calculation. The composite proportional evaluation (PSI-COPRAS) method is used to calculate the weight relationship of each feature. For the j-th feature, the interval-valued generalized orthogonal fuzzy weighted average neutral aggregation operator (IVq-ROFWANA operator) is first used to calculate the average performance value N under the j-th feature. j , as shown in Formula 1.

[0058]

[0059] in, represents the intuitive fuzzy value corresponding to the jth feature in the i-th data segment, This is called the hesitation degree, which means that for r ij The degree of uncertainty of neither belonging to nor not belonging to a certain category, m represents the total number of data segments, n represents the total number of features, Represents the weight of each data segment when calculating the average performance value of the jth feature, and λ satisfies Membership For r ij The degree of belonging to a certain category, non-membership For r ij The degree of not belonging to a certain category.

[0060] Calculate the variation value φ of the performance value of all data segments for each feature j , as shown in Formula 2.

[0061]

[0062] In Formula 2, s(·) represents the score function, and the calculation formula is shown in Figure 3.

[0063]

[0064] In formula 3, S(r ij ) represents the score corresponding to the jth feature in the i-th data segment, and the normalized variation value φ j , get the weight w of each feature j , as shown in Formula 4.

[0065]

[0066] S42. Aggregate the fragment data. Use the IVq-ROFFWA operator to aggregate each row of data in the matrix R, as shown in Formula 5. The result is the intuitive fuzzy value r of the corresponding row (data fragment). i .

[0067]

[0068] S43, calculate the score of each data segment. Use the score function of formula 3 to calculate the score of each r i , get the score S(r i ),(i=1,2,…,m).

[0069] S44: Calculate the comprehensive score: According to Formula 6, calculate the comprehensive evaluation value F of the learner's m data segments.

[0070]

[0071] S5. Convert the comprehensive evaluation value F into the corresponding concentration level according to Table 3. Obtain the learner's online learning concentration recognition result.

[0072] Table 3 Conversion of comprehensive evaluation values into concentration levels

[0073]

[0074]

[0075] Taking five scholars as examples, the process of identifying online learning concentration was further compared and analyzed.

[0076] (1) Feature extraction. Feature extraction is performed on the segmented image data and ECG signal data of the learner. After extraction, 596 rows and 4 columns of data are obtained. Each row represents a data segment, and each column represents a feature, as shown in Table 4.

[0077] Table 4 Some features extracted

[0078]

[0079] (2) Data conversion. Convert the specific value of each feature into an intuitive fuzzy value a i , as shown in Table 5.

[0080] Table 5 Conversion of characteristic values into fuzzy values

[0081]

[0082] (3) Determine the feature weight. Use Formula 4 to calculate the feature weight vector, and the result is: w = [0.00700332, 0.56946195, 0.09210248, 0.33143224].

[0083] (4) Use formula 5 to perform weighted fusion on the four intuitive fuzzy values of m data segments to obtain the comprehensive fuzzy value r of each data segment. i , as shown in Table 6.

[0084] Table 6 Fuzzy values after aggregation of data segments

[0085]

[0086] (5) Calculate the score. Use the score function of formula 3 to calculate the comprehensive fuzzy value r of each data segment. i The score S(r i ), as shown in Table 7.

[0087] Table 7 Comprehensive fuzzy value r of data segments i Score

[0088]

[0089] (6) Calculate the comprehensive score. According to Formula 6, calculate the comprehensive evaluation value of the learner's m data segments F = 0.5896140495090181.

[0090] (7) According to Table 4, the final comprehensive evaluation value F corresponds to the concentration level I.

[0091] This learner's online learning concentration was assessed as Level II, a result that closely matches the actual situation and demonstrates the effectiveness and feasibility of the fuzzy fusion model for assessing learner concentration. To further verify the reliability and universal applicability of this online concentration identification method, the online learning concentration of four other learners was assessed. The evaluation results for these five learners are shown in Table 8.

[0092] Table 8 Learners’ final online learning concentration scores and levels

[0093]

[0094] Table 8 shows that Learners 1, 2, and 4 have extremely high levels of focus, while Learners 3 and 5 have relatively high levels of focus. Based on the scores in Table 8, the focus ranking of the five learners is as follows: Learner 2 ranks first, followed by Learner 4 and Learner 1, while Learners 3 and 5 come in fourth and fifth, respectively. Therefore, compared to existing methods that can only determine whether a learner is focused, the focus fuzzy fusion model can not only determine the learner's level of focus but also quantify and rank it.

[0095] In order to verify the effectiveness of the online learning concentration evaluation model (EFocusOnline model) based on the fusion of facial clues and electrocardiogram (ECG) signals provided by the present invention in evaluating the concentration level, a confusion matrix is used to compare the evaluation results of the EFocusOnline model with the concentration levels evaluated by experts (the levels marked in the dataset). Figure 3 As shown, Figure 3 The five different confusion matrices in the figure show a comparison of the different levels of focus of five learners. Each row in the confusion matrix represents the expert assessment level, and each column represents the EFocusOnline model's assessment results. Specifically, the value in each cell indicates the number of cases where the experts actually assessed the level represented by that row, but the EFocusOnline model assessed the level represented by that column. The cells on the diagonal indicate the number of cases where the EFocusOnline model correctly assessed the focus level, while the cells off the diagonal indicate cases where the EFocusOnline model incorrectly assessed the focus level.

[0096] The present invention also provides an online learning concentration recognition system, which specifically includes:

[0097] The data acquisition module is used to obtain the learner's learning image and ECG signal; obtain multiple learning features based on the learning image and ECG signal, and quantify each learning feature into different levels; process the multiple learning features through a sliding window to obtain a data matrix; wherein the multiple learning features represent the learner's facial features and physiological features.

[0098] The data processing module is used to set the hierarchical correspondence between different levels of learning features and concentration, and convert the data matrix into a fuzzy matrix according to the hierarchical correspondence.

[0099] The model processing module is used to input the fuzzy matrix into the concentration fuzzy fusion model, calculate the weight relationship of multiple features through the composite proportional evaluation method, calculate the intuitive fuzzy value of the data segment based on the weight relationship, calculate the score of the data segment based on the intuitive fuzzy value scoring function, and obtain the learner's comprehensive evaluation value based on the score of the data segment.

[0100] The concentration recognition module is used to obtain the learner's concentration level recognition result according to the learner's comprehensive evaluation value based on the correspondence between the comprehensive evaluation value and the concentration level.

[0101] Each module in the above-mentioned online learning concentration recognition system can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a memory in the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0102] The present invention also provides a computer device comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of an embodiment of a method for identifying online learning concentration. The specific implementation method can be found in the method embodiment and will not be repeated here.

[0103] Furthermore, the present invention also provides a non-temporary computer-readable storage medium containing instructions, on which a computer program is stored. For example, a memory containing instructions, the instructions can be executed by a processor of a computer device to complete the above method. For example, a non-temporary computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device. When the computer program is executed by the processor, it can implement the steps in an embodiment of an online learning concentration identification method. The specific implementation method can be found in the method embodiment, which will not be repeated here.

[0104] Those skilled in the art will appreciate that embodiments of the present invention may provide methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0105] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0106] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0107] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0108] It should be pointed out that the specific implementation methods described above can enable those skilled in the art to understand the invention more comprehensively, but do not limit the invention in any way. Therefore, although the present specification and examples have described the invention in detail, those skilled in the art should understand that the invention can still be modified or replaced by equivalents; and all technical solutions and improvements that do not deviate from the spirit and scope of the invention are included in the scope of protection of the patent for the invention. Any figure mark in the claims should not be regarded as limiting the claims involved. Any simple change or equivalent replacement of the technical solution that can be obviously obtained by any person familiar with the art within the technical scope disclosed in the present invention falls within the scope of protection of the present invention.

Claims

1. A method for identifying online learning concentration, characterized in that: The following steps are involved: Acquire a learning image and ECG signal of a learner; obtain a plurality of learning features based on the learning image and ECG signal, and quantify each learning feature into different levels; process the plurality of learning features through a sliding window to obtain a data matrix; wherein the plurality of learning features represent facial features and physiological features of the learner; Setting a level correspondence between different levels of learning features and concentration, and converting the data matrix into a fuzzy matrix according to the level correspondence; Inputting the fuzzy matrix into a fuzzy fusion model of concentration, calculating the weight relationship of multiple features using a composite proportional evaluation method, calculating the comprehensive fuzzy value of the data segment based on the weight relationship, calculating the score of the data segment based on the comprehensive fuzzy value scoring function, and calculating the comprehensive evaluation value of the learner based on the score of the data segment; Based on the corresponding relationship between the comprehensive evaluation value and the concentration level, the learner's concentration level identification result is obtained according to the learner's comprehensive evaluation value.

2. The method for identifying online learning concentration according to claim 1, characterized in that: The weight relationship of multiple features is calculated by the composite ratio evaluation method, specifically comprising the following steps: The average performance value N under the jth feature is calculated using the interval-valued generalized orthogonal fuzzy weighted average neutral aggregation IVq-ROFWANA operator. j , specifically: in, represents the intuitive fuzzy value corresponding to the jth feature in the i-th data segment, Indicates belonging to r ij The membership degree, Indicates that it does not belong to r ij The non-membership degree of Represents r ij The degree of uncertainty; Θ NA represents the IVq-ROEWANA operator; m represents the total number of data segments, n represents the total number of features, Represents the weight of each data segment when calculating the average performance value of the jth feature, and λ satisfies Calculate the variation value φ of the performance value of all data segments for each feature j , specifically: Among them, s(·) represents the score function; by normalizing the variation value φ j , get the weight w of each feature j , specifically: According to the weight value of each feature, the weight relationship between multiple features is obtained.

3. The method for identifying online learning concentration according to claim 2, characterized in that: The calculation formula of the score function s(·) is specifically: Among them, r ij represents the intuitive fuzzy value corresponding to the jth feature in the i-th data segment, Indicates belonging to r ij The membership degree, Indicates that it does not belong to r ij The non-membership degree of Represents r ij degree of uncertainty.

4. The method for identifying online learning concentration according to claim 1, characterized in that: The learning image and ECG signal of the learner are obtained by obtaining the learner's learning video in each period, and using a segmentation tool to segment the video time sequence in the form of one picture per second to obtain the learner's learning image; the ECG signal is specifically extracted by a physiological signal device.

5. The method for identifying online learning concentration according to claim 1, characterized in that: The multiple learning features are specifically head posture features, fatigue features, facial expression features and heart rate change features.

6. The method for identifying online learning concentration according to claim 5, characterized in that: The facial expression features are extracted using a lightweight facial expression recognition (FER) model based on the EfficientNet and MobileNet architectures. The head posture is estimated using the solvePnP algorithm to calculate the pitch angle of the learner's head and the yaw angle of the face. Fatigue feature detection includes eye closure detection and yawn detection. Eye closure detection is performed by calculating the eyelid closure percentage PERCLOS, and yawn detection is performed by calculating the mouth closure degree. The heart rate change feature calculates the number of heart beats per minute by identifying the R wave peak.

7. An online learning concentration recognition system, characterized by: include: A data acquisition module, used to acquire the learner's learning images and ECG signals; Based on the learning image and ECG signal, a plurality of learning features are obtained, and each learning feature is quantified into different levels; the plurality of learning features are processed through a sliding window to obtain a data matrix; wherein the plurality of learning features represent facial features and physiological features of the learner; a data processing module, configured to set a hierarchical correspondence between different levels of learning features and concentration, and convert the data matrix into a fuzzy matrix according to the hierarchical correspondence; a model processing module, configured to input the fuzzy matrix into a concentration fuzzy fusion model, calculate a weight relationship among multiple features using a composite proportional evaluation method, calculate an intuitive fuzzy value of a data segment based on the weight relationship, calculate a score for the data segment based on the intuitive fuzzy value scoring function, and calculate a comprehensive evaluation value for the learner based on the score of the data segment; The concentration identification module is used to obtain a concentration level identification result of the learner according to the comprehensive evaluation value of the learner based on the correspondence between the comprehensive evaluation value and the concentration level.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory, wherein: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is loaded into a processor, it can execute the steps of the method according to any one of claims 1 to 6.

Images