Cooperative learning method and system based on multi-user eye movement data analysis

Abstract

The invention discloses a collaborative learning method and collaborative learning system based on multi-user eye movement data analysis. The collaborative learning method comprises the steps of obtaining eye movement data of collaborative learning of all users in a current stage; constructing an attention synchronization matrix according to the eye movement data; constructing an interaction intensity matrix according to the eye movement data; according to the attention synchronization matrix and the interaction intensity matrix, an overall cooperation index of all the users is obtained through calculation; and according to the attention synchronization matrix, the interaction intensity matrix and the overall cooperation index, based on a preset cooperation learning model, obtaining a cooperation learning scheme of a next stage. According to the method, the attention synchronization matrix and the interaction intensity matrix can be constructed by analyzing the multi-user eye movement data in real time, the collaborative learning effect is accurately evaluated, and personalized optimization suggestions and dynamic adjustment are provided. Therefore, the efficiency and quality of collaborative learning are remarkably improved, and highly personalized learning experience is provided for each learner.

Description

Technical Field

[0001] The invention relates to a collaborative learning method and a collaborative learning system based on multi-user eye movement data analysis, and belongs to the technical field of collaborative learning. Background Art

[0002] Existing collaborative learning platforms mainly rely on traditional methods such as text communication, audio and video interaction for group learning, and lack in-depth analysis and optimization of interactions between group members. Although some systems use eye tracking technology to evaluate the performance of individual learners, they fail to fully utilize this data to optimize the group learning process. Existing technologies cannot analyze the eye movement patterns of multiple learners in real time, so it is impossible to adjust the learning content or group interaction strategies in time to improve the overall learning effect. In addition, existing systems also lack quantitative evaluation of attention distribution and knowledge sharing efficiency among group members, which limits the in-depth understanding and improvement of the collaborative learning process. Summary of the invention

[0003] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a collaborative learning method and a collaborative learning system based on multi-user eye movement data analysis, which can accurately evaluate the collaborative learning effect by analyzing multi-user eye movement data in real time, construct an attention synchronization matrix and an interaction intensity matrix, and provide personalized optimization suggestions and dynamic adjustments. This not only significantly improves the efficiency and quality of collaborative learning, but also provides each learner with a highly personalized learning experience.

[0004] To achieve the above object, the present invention is implemented by adopting the following technical solutions:

[0005] On the one hand, the present invention discloses a collaborative learning method based on multi-user eye movement data analysis, comprising:

[0006] Obtain the eye movement data of all users’ collaborative learning in the current stage;

[0007] constructing an attention synchronization matrix according to the eye movement data, wherein the attention synchronization matrix includes the degree of attention synchronization between any two users;

[0008] constructing an interaction intensity matrix according to the eye movement data, wherein the interaction intensity matrix includes interaction intensity between any two users;

[0009] Calculate the overall collaboration index of all users according to the attention synchronization matrix and the interaction intensity matrix;

[0010] According to the attention synchronization matrix, the interaction intensity matrix and the overall collaboration index, based on a preset collaborative learning model, a collaborative learning plan for the next stage is obtained.

[0011] Furthermore, the eye movement data of any user's collaborative learning includes user ID, gaze point, gaze duration, eye movement trajectory and pupil data.

[0012] Furthermore, the attention synchronization matrix is ​​an N×N attention synchronization matrix, and the expression is as follows:

[0013] ;

[0014] In the formula, represents the attention synchronization matrix; represents the attention synchronization between user i and user j, represents the attention synchronization between user j and user i, and The value of is the same, , where N represents the number of users.

[0015] Furthermore, the attention synchronization The calculation steps are as follows:

[0016] For any user i and user j, analyze and count the frequency of overlapping gaze points and the corresponding gaze duration in the eye movement data of the two users;

[0017] Each time the gaze points overlap is regarded as an independent gaze point overlap event, and the number of gaze point overlap events is obtained according to the frequency of gaze point overlap; the duration of the corresponding gaze point overlap event is obtained according to the corresponding gaze duration;

[0018] According to the number and duration of the gaze point overlap events, the attention synchronization between user i and user j is obtained. , the expression is as follows:

[0019] ;

[0020] In the formula, represents the duration of the kth gaze point overlap event, K represents the number of times the gaze points of user i and user j overlap; Indicates the total duration of the current stage.

[0021] Furthermore, the interaction intensity matrix is ​​an N×N interaction intensity matrix, and the expression is as follows:

[0022] ;

[0023] In the formula, represents the interaction intensity matrix; represents the interaction intensity between user i and user j, represents the interaction intensity between user j and user i, and The value of is the same, , where N represents the number of users.

[0024] Furthermore, the calculation steps of the interaction strength are as follows:

[0025] For any user i and user j, by analyzing the gaze points and gaze duration, the frequency and average duration of the two users' gazes on each other are statistically obtained, including the frequency and average duration of user i's gaze on user j, and the frequency and average duration of user j's gaze on user i;

[0026] According to the frequency and average duration of the two users looking at each other, the interaction intensity between user i and user j is obtained , the expression is as follows:

[0027] ;

[0028] In the formula, represents the frequency of user i looking at user j; represents the average time that user i looks at user j; represents the frequency with which user j looks at user i; represents the average time that user j looks at user i.

[0029] Furthermore, the expression of the overall collaboration index is as follows:

[0030] ;

[0031] In the formula, represents the overall collaboration index; represents the first weight coefficient; represents the attention synchronization between user i and user j; Indicates the number of users; represents the second weight coefficient; represents the interaction intensity between user i and user j.

[0032] Furthermore, the preset collaborative learning model includes:

[0033] A collaboration bottleneck identification unit, used to obtain user pairs with collaboration bottlenecks according to the attention synchronization matrix and the interaction intensity matrix;

[0034] A personalized collaboration unit, used for generating a next-stage attention synchronization plan and / or an interaction intensity plan for the user pair with a collaboration bottleneck;

[0035] The overall collaboration unit is used to generate the next stage system learning tasks and user interface layout according to the overall collaboration index.

[0036] Furthermore, the collaborative learning model further includes:

[0037] A data analysis unit, used to obtain a collaboration status and a personalized report based on the attention synchronization matrix, the interaction intensity matrix and the overall collaboration index;

[0038] The data storage unit is used to store the data of each unit.

[0039] On the other hand, the present invention discloses a collaborative learning system based on multi-user eye movement data analysis, which is applicable to the above-mentioned collaborative learning method based on multi-user eye movement data analysis, including:

[0040] The data acquisition module is used to obtain the eye movement data of all users' collaborative learning in the current stage;

[0041] An attention synchronization module, used to construct an attention synchronization matrix according to the eye movement data, wherein the attention synchronization matrix includes the degree of attention synchronization between any two users;

[0042] An interaction strength module, used to construct an interaction strength matrix according to the eye movement data, wherein the interaction strength matrix includes the interaction strength between any two users;

[0043] An overall collaboration module, used to calculate the overall collaboration index of all users according to the attention synchronization matrix and the interaction intensity matrix;

[0044] The collaborative learning module is used to obtain a collaborative learning plan for the next stage based on a preset collaborative learning model according to the attention synchronization matrix, the interaction intensity matrix and the overall collaborative index.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] The collaborative learning method and collaborative learning system based on multi-user eye movement data analysis of the present invention, firstly, can accurately evaluate the collaborative learning effect by real-time analysis of multi-user eye movement data, and provide personalized optimization suggestions and dynamic adjustments. This not only significantly improves the efficiency and quality of collaborative learning, but also provides a highly personalized learning experience for each learner. Secondly, it is possible to calculate the overall collaboration index between learners based on eye movement data, and optimize the overall learning plan based on the overall collaboration index, thereby realizing intelligent and personalized collaborative learning recommendations, so that the overall learning plan can well meet the needs of different learners. Finally, the phased collaborative learning plan can promote learners' self-regulation ability through real-time feedback, and help learners improve their learning effects.

[0047] In addition, this method is particularly effective in remote collaboration scenarios, solving the problems of attention synchronization and interactive effects in distance learning, and providing a new solution for improving the quality of online education. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a flow chart of the collaborative learning method based on multi-user eye movement data analysis provided in Example 1. DETAILED DESCRIPTION

[0049] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0050] Example 1

[0051] This embodiment 1 provides a collaborative learning method based on multi-user eye movement data analysis, including:

[0052] Obtain the eye movement data of all users’ collaborative learning in the current stage;

[0053] Based on the eye movement data, an attention synchronization matrix is ​​constructed. The attention synchronization matrix contains the degree of attention synchronization between any two users.

[0054] Based on the eye movement data, an interaction intensity matrix is ​​constructed. The interaction intensity matrix contains the interaction intensity between any two users.

[0055] Based on the attention synchronization matrix and the interaction intensity matrix, the overall collaboration index of all users is calculated;

[0056] According to the attention synchronization matrix, interaction intensity matrix and overall collaboration index, based on the preset collaborative learning model, the collaborative learning plan for the next stage is obtained.

[0057] The technical concept of the present invention is: first, it can accurately evaluate the collaborative learning effect by real-time analysis of multi-user eye movement data, and provide personalized optimization suggestions and dynamic adjustments. This not only significantly improves the efficiency and quality of collaborative learning, but also provides each learner with a highly personalized learning experience. Secondly, it can calculate the overall collaboration index between learners based on eye movement data, and optimize the overall learning plan based on the overall collaboration index, so as to achieve intelligent and personalized collaborative learning recommendations, so that the overall learning plan can well meet the needs of different learners. Finally, the phased collaborative learning plan realizes the real-time feedback function, promotes the learners' self-regulation ability, and helps learners improve their learning effects. Long-term data analysis provides valuable support for educational research. In addition, this system is particularly effective in remote collaboration scenarios, solves the problems of attention synchronization and interactive effects in remote learning, and provides a new solution for improving the quality of online education.

[0058] like Figure 1 As shown, the specific steps are as follows:

[0059] Step 1: Obtain the eye movement data of all users’ collaborative learning in the current stage;

[0060] The specific steps for obtaining user eye movement data are: collecting user eye movement data in real time through the eye tracking device installed on each user's device, and transmitting the data to the central server.

[0061] The eye movement data of any user's collaborative learning includes user ID, gaze point, gaze duration, eye movement trajectory and pupil data.

[0062] Specifically, the eye movement data includes all the eye movement information of the user during the learning process, and each data item is separated by a semicolon. For example, (userId_1, gaze point coordinates: x1, y1; gaze duration: t1; eye movement trajectory: path1; pupil size: d1......) represents the eye movement data of the user with ID userId_1 at a certain moment.

[0063] By collecting and transmitting eye movement information in real time, the system can obtain the eye movement data of all users participating in collaborative learning.

[0064] In order to accurately evaluate the effect of collaborative learning and perform personalized optimization, this embodiment measures the quality of collaborative learning by analyzing the synchronization of attention and interaction intensity among users, as follows:

[0065] Step 2: Based on the eye movement data, construct an attention synchronization matrix, which contains the degree of attention synchronization between any two users.

[0066] For all users, consider their pairwise combinations, for example, the combination of user i and user j, and count the frequency and duration of the overlap of the two people's gaze points during the learning process to obtain the attention synchronization between users i and j. In this way, the attention synchronization between all pairs of users is obtained, and then the attention synchronization matrix between all users is obtained as an N×N matrix, which is expressed as follows:

[0067] ;

[0068] In the formula, represents the attention synchronization matrix; represents the attention synchronization between user i and user j, represents the attention synchronization between user j and user i, and The value of is the same, , where N represents the number of users.

[0069] Among them, attention synchronization The calculation steps are as follows:

[0070] For any user i and user j, analyze and count the frequency of overlapping gaze points and the corresponding gaze duration in the eye movement data of the two users;

[0071] Each time the gaze points overlap is regarded as an independent gaze point overlap event, and the number of gaze point overlap events is obtained according to the frequency of gaze point overlap; the duration of the corresponding gaze point overlap event is obtained according to the corresponding gaze duration;

[0072] According to the number and duration of the gaze point overlap events, the attention synchronization between user i and user j is obtained. , the expression is as follows:

[0073] ;

[0074] In the formula, represents the duration of the kth gaze point overlap event, K represents the number of times the gaze points of user i and user j overlap; Indicates the total duration of the current stage.

[0075] The core idea of ​​this method is to calculate the ratio of the total time of the two people's gaze points overlapping to the total learning time. The higher the ratio, the higher the degree of synchronization of the two people's attention.

[0076] Assume that the total duration T of the current stage is 20 seconds, and there are 3 users in the learning group. The gaze points of user 1 and user 2 overlap three times during the learning process. The first overlap lasts for 2 seconds, the second overlap lasts for 5 seconds, and the third overlap lasts for 3 seconds. Then, , , . N is 3, which means there are 3 coincidence events in total.

[0077] ;

[0078] and The values ​​of are the same, both are 0.5. In order to get a complete matrix, we also need , and the 0 element on the diagonal. If the total time that user 1 and user 3's gaze points overlap is 8 seconds, then , the total time that the gaze points of user 2 and user 3 overlap is 6 seconds, then .

[0079] Then the matrix S is:

[0080] .

[0081] Through the attention synchronization matrix, the attention synchronization between all users is obtained, which can quickly evaluate the collaboration status between group members.

[0082] Step 3: Based on the eye movement data, construct an interaction intensity matrix, which contains the interaction intensity between any two users.

[0083] In this embodiment, the interaction intensity is measured by analyzing the mutual gaze behavior between users. The frequency and duration of user i's gaze on user j, and the frequency and duration of user j's gaze on user i are comprehensively considered to obtain the interaction intensity between user i and user j, that is, the interaction intensity between user j and user i.

[0084] The interaction intensity matrix is ​​an N×N interaction intensity matrix, and the expression is as follows:

[0085] ;

[0086] In the formula, represents the interaction intensity matrix; represents the interaction intensity between user i and user j, represents the interaction intensity between user j and user i, and The value of is the same, , where N represents the number of users.

[0087] The calculation steps of interaction intensity are as follows:

[0088] For any user i and user j, by analyzing the gaze points and gaze duration, the frequency and average duration of the two users' gazes on each other are statistically obtained, including the frequency and average duration of user i's gaze on user j, and the frequency and average duration of user j's gaze on user i;

[0089] According to the frequency and average duration of the two users looking at each other, the interaction intensity between user i and user j is obtained , the expression is as follows:

[0090] ;

[0091] In the formula, represents the frequency of user i looking at user j; represents the average time that user i looks at user j; represents the frequency with which user j looks at user i; represents the average time that user j looks at user i.

[0092] Step 4: Calculate the overall collaboration index of all users based on the attention synchronization matrix and the interaction intensity matrix.

[0093] The expression of the overall collaboration index is as follows:

[0094] ;

[0095] In the formula, represents the overall collaboration index; represents the first weight coefficient, which is used to adjust the importance of attention synchronization in the overall evaluation; represents the attention synchronization between user i and user j; Indicates the number of users; represents the second weight coefficient, which is used to adjust the importance of interaction strength in the overall evaluation; represents the interaction intensity between user i and user j.

[0096] It should be noted that the solution of this embodiment can be replaced by not constructing the attention synchronization matrix and the interaction intensity matrix, but directly calculating the collaboration index between each pair of users based on the multi-user eye movement data, and summing up these indices to obtain the overall collaboration index.

[0097] Step 5: According to the attention synchronization matrix, interaction intensity matrix and overall collaboration index, based on the preset collaborative learning model, obtain the collaborative learning plan for the next stage.

[0098] Pre-set collaborative learning models, including:

[0099] (1) A collaborative bottleneck identification unit is used to obtain user pairs with collaborative bottlenecks based on the attention synchronization matrix and the interaction intensity matrix.

[0100] According to the attention synchronization matrix and the interaction intensity matrix, the attention synchronization degree and interaction intensity of different user pairs are obtained.

[0101] In response to the fact that the attention synchronization degree of any user pair is less than a preset attention synchronization threshold and the interaction intensity is less than a preset interaction threshold, it is determined that the user pair has a collaboration bottleneck.

[0102] (2) A personalized collaboration unit, which is used to generate the next-stage attention synchronization plan and / or interaction intensity plan for user pairs with collaboration bottlenecks.

[0103] Based on the identified collaboration bottlenecks, the system generates personalized collaboration suggestions for each user. For example:

[0104] For user pairs with collaboration bottlenecks, if the attention synchronization is too low, it is recommended to increase the learning materials of common focus as the attention synchronization plan for the next stage; if the interaction intensity is too low, it is recommended to increase the opportunities for direct communication as the interaction intensity plan for the next stage.

[0105] (3) Overall collaboration unit, which is used to generate the next stage system learning tasks and user interface layout according to the overall collaboration index.

[0106] According to the group's overall collaboration index and individual collaboration performance, the system dynamically adjusts the presentation of learning tasks and the layout of the user interface. For example:

[0107] In response to the overall collaboration index being less than a preset overall collaboration threshold, a learning task with strong interactivity is added.

[0108] For users with poor collaboration performance, highlight other members' concerns in the interface.

[0109] (4) A data analysis unit, which is used to obtain collaboration status and personalized reports based on the attention synchronization matrix, interaction intensity matrix, and overall collaboration index.

[0110] The current collaboration status of users is displayed in a visual way, including indicators such as attention synchronization and interaction intensity, as well as corresponding improvement suggestions.

[0111] Collaboration status can be presented in different visualizations. For example, a heat map can be used to show the degree of synchronization between group members, or a network diagram can be used to show interaction patterns. The system can also automatically select the most suitable visualization method based on user preferences or device type.

[0112] Based on long-term data analysis, a personalized learning report is regularly generated for each user, including the changing trend of collaboration ability, advantages and areas for improvement, etc. The system can introduce machine learning algorithms such as cluster analysis or time series prediction to identify collaboration patterns and predict future performance.

[0113] (5) Data storage unit, used to store data of each unit,

[0114] Continuously record and store users' collaborative learning data, including eye movement data, collaboration index, etc., and identify users' collaborative learning patterns and progress through long-term analysis.

[0115] This method can also have the following two expansion solutions:

[0116] Solution 1: Multimodal collaborative learning status evaluation based on eye movement, EEG and keyboard input.

[0117] At the same time, the user's eye movement data, EEG data and keyboard input data are collected. The eye movement data includes the average fixation duration, number of glances, and frequency of fixation point transfer. EEG features include the power spectrum density of each frequency band, the coherence of brain waves, etc. Keyboard input features include input speed, error rate, code modification frequency, etc.

[0118] The long short-term memory network LSTM is used to take the eye movement, EEG and keyboard input feature sequences as the input of LSTM. The LSTM network learns the temporal dependency between different modal features and outputs the collaborative learning state vector of the user at each time point. The collaborative learning state vector output by LSTM is input into a support vector machine (SVM) classifier to classify the user's collaborative learning state, such as concentration, distraction, confusion, active participation, etc.

[0119] Solution 2: Collaborative pattern recognition and learning effect prediction based on deep learning.

[0120] The user's eye movement data, interaction data, and learning performance data are cleaned and standardized. The interaction data includes the number of speeches and the number of messages sent. The user's collaborative behavior data is reduced in dimension and features are extracted using an autoencoder. The extracted features are input into a K-Means clustering algorithm to divide users into different collaborative mode groups, such as leading, following, independent, etc.

[0121] A multi-layer perceptron (MLP) network is used to predict the user's learning effect. The user's collaboration mode, eye movement features, interaction features, and other learning-related features are used as inputs to the MLP to predict the user's final learning results or other learning effect indicators.

[0122] Model training uses existing collaborative learning data to train and optimize Autoencoder, K-Means and MLP.

[0123] The innovations of this method are mainly reflected in the following aspects:

[0124] 1. Real-time analysis of multi-user eye movement data. Not only does it collect the eye movement data of a single user, but it also analyzes the eye movement relationship between multiple users in real time, which provides a new dimension for evaluating the effectiveness of collaborative learning.

[0125] 2. Quantification of attention synchronization and interaction intensity. By constructing the attention synchronization matrix and the interaction intensity matrix, the collaborative state between group members can be objectively quantified, providing a data basis for optimizing collaborative learning.

[0126] 3. Dynamic learning content and interface adjustment. Based on the results of real-time analysis, the system can automatically adjust the presentation of learning content and the layout of the user interface to promote better collaboration. This dynamic adjustment capability makes the learning process more personalized and efficient.

[0127] 4. Personalized collaboration suggestions. The system can not only identify collaboration bottlenecks, but also generate targeted collaboration suggestions for each user to help users improve their collaboration skills.

[0128] 5. Long-term learning tracking and analysis. By continuously recording and analyzing users’ collaborative learning data, users’ learning patterns and progress can be identified, providing a basis for the evaluation and improvement of long-term learning effects.

[0129] 6. Real-time visual feedback. The system shows users their collaboration status in an intuitive and visual way, allowing users to understand their performance and make adjustments in a timely manner.

[0130] Through the above technical solution, the present invention provides an innovative collaborative learning optimization method based on multi-user eye movement data analysis. The method can analyze the eye movement data of multiple users in real time, evaluate the collaborative learning effect, and provide personalized optimization suggestions and dynamic adjustments. This can not only improve the efficiency and quality of collaborative learning, but also provide valuable data support for educational researchers and promote the development of collaborative learning theory and practice.

[0131] Example 2

[0132] This embodiment 2 provides a collaborative learning system based on multi-user eye movement data analysis, which is applicable to the collaborative learning method based on multi-user eye movement data analysis in embodiment 1, including:

[0133] The data acquisition module is used to obtain the eye movement data of all users' collaborative learning in the current stage;

[0134] The attention synchronization module is used to construct an attention synchronization matrix based on eye movement data. The attention synchronization matrix contains the degree of attention synchronization between any two users.

[0135] An interaction strength module is used to construct an interaction strength matrix based on eye movement data. The interaction strength matrix contains the interaction strength between any two users.

[0136] The overall collaboration module is used to calculate the overall collaboration index of all users based on the attention synchronization matrix and the interaction intensity matrix;

[0137] The collaborative learning module is used to obtain the collaborative learning plan for the next stage based on the preset collaborative learning model according to the attention synchronization matrix, the interaction intensity matrix and the overall collaborative index.

[0138] Example 3

[0139] This embodiment 3 provides a collaborative learning method based on multi-user eye movement data analysis in a classroom group discussion scenario. In a group discussion session in a university classroom, each student is equipped with an eye tracking device. The operation is performed according to the following steps:

[0140] (1) Data collection:

[0141] Eye movement data is collected in real time through each student's eye tracking device, including gaze point, gaze duration, eye movement trajectory, etc. The data is transmitted to the central server in real time via wireless network.

[0142] (2) Data analysis:

[0143] Data analysis was performed every 5 seconds to construct the attention synchronization matrix and interaction intensity matrix.

[0144] Calculate the overall collaboration index of the group and set α=0.6 and β=0.4.

[0145] (3) Learning Optimization:

[0146] If two students are found to have significantly lower-than-average attention synchronization (e.g., less than 50%), the system will highlight the shared focus on their screens.

[0147] If a student's interaction intensity is generally low, the system will encourage that student to participate more in the discussion through slight visual cues.

[0148] (4) Feedback and follow-up:

[0149] Each student's interface has a small dashboard that displays the current collaboration status.

[0150] After the discussion, the system generates a group report, including the overall trend of collaboration index changes, the contribution of each member, etc.

[0151] Example 4

[0152] This embodiment 4 provides a collaborative learning method based on multi-user eye movement data analysis in a remote online collaborative learning scenario. In a multinational online collaborative project, participants use webcams equipped with eye tracking functions. The operation is as follows:

[0153] (1) Data collection:

[0154] The webcam collects participants’ eye movement data in real time through a specific algorithm.

[0155] The data is transmitted to the cloud server through an encrypted channel.

[0156] (2) Data analysis:

[0157] Analyze data in real time and update the attention synchronization matrix and interaction intensity matrix every 10 seconds.

[0158] Taking into account the particularity of remote collaboration, the system adjusted the weights when calculating the overall collaboration index, setting α=0.5 and β=0.5.

[0159] (3) Learning Optimization:

[0160] Automatically adjust the layout of the shared screen to place the video stream of participants who collaborate better in a more prominent position.

[0161] When it is detected that a participant has not received attention from others for a long time, the system will prompt the host to give the participant a chance to speak.

[0162] (4) Feedback and follow-up:

[0163] The system displays a collaborative network diagram on each participant's interface, visually showing the group's interaction status.

[0164] After the project is completed, a detailed collaboration analysis report is generated, including changes in the collaboration index for each meeting, individual participation, best practices, etc.

[0165] Example 5

[0166] This embodiment 5 provides a collaborative learning method based on multi-user eye movement data analysis in a collaborative programming scenario. In a software development team's pair programming, each programmer's monitor is equipped with an eye tracking device. The operation is performed according to the following steps:

[0167] (1) Data collection:

[0168] The eye tracking device collects the programmer's eye movement data in real time, paying special attention to the gaze in the code area. It also records information such as keyboard input and code changes.

[0169] (2) Data analysis:

[0170] The data were analyzed every 3 seconds to construct an attention synchronization matrix.

[0171] In this scenario, special attention should be paid to the synchronization of the two programmers' gazes on the same code segment.

[0172] The interaction intensity matrix is ​​based on the frequency of conversations between programmers and the degree of alternation of code modifications.

[0173] (3) Learning Optimization:

[0174] When it detects that the attention of two programmers is out of sync for a long time, the code segments that may need to be discussed will be highlighted on the screen.

[0175] If a programmer's involvement is significantly reduced, a role swap, such as from driver to navigator, is suggested.

[0176] (4) Feedback and follow-up:

[0177] A small toolbar is integrated into the IDE to display the current collaboration status in real time.

[0178] After each programming session, a report is generated, including information such as the relationship analysis between code quality and collaboration index, and the best collaboration period.

[0179] In summary, this method helps participants better synchronize their attention and improve collaboration efficiency through real-time analysis and feedback. It can provide personalized suggestions based on each user's collaborative performance to promote the improvement of personal capabilities. It can provide personalized suggestions based on each user's collaborative performance to promote the improvement of personal capabilities.

[0180] Large-scale eye movement data provides researchers with rich research materials, which helps to gain a deeper understanding of the cognitive process of collaborative learning.

[0181] With the increasing popularity of distance education, this system provides a new solution for improving the quality of online collaborative learning.

[0182] Although it is mainly aimed at educational scenarios, the core concepts and technologies of this system can also be extended to other fields that require collaboration, such as team management, product design, etc.

[0183] The improvement of the present invention lies in the introduction of multi-user eye movement data analysis, the construction of attention synchronization matrix and interaction intensity matrix, and the dynamic optimization of collaborative learning process. These improvements solve the technical problems of traditional collaborative learning evaluation, such as strong subjectivity, lack of real-time feedback, and difficulty in personalized optimization. Through real-time data collection and analysis, it is possible to objectively evaluate the quality of collaboration, identify problems in a timely manner, and provide personalized optimization suggestions, thereby significantly improving the efficiency and quality of collaborative learning.

[0184] The collaborative learning method and collaborative learning system based on multi-user eye movement data analysis of the present invention can accurately evaluate the collaborative learning effect by real-time analysis of multi-user eye movement data, and provide personalized optimization suggestions and dynamic adjustments. This not only significantly improves the efficiency and quality of collaborative learning, but also provides a highly personalized learning experience for each learner. The system can calculate the attention synchronization and interaction intensity between learners based on eye movement data, and optimize the overall learning plan based on the attention synchronization and interaction intensity, thereby realizing intelligent and personalized collaborative learning recommendations, so that the overall learning plan can well meet the needs of different learners.

[0185] At the same time, the system's real-time feedback function promotes learners' self-regulation ability and helps learners improve their learning outcomes. Long-term data analysis provides valuable support for educational research. In addition, this system is particularly effective in remote collaboration scenarios, solving the problems of attention synchronization and interactive effects in remote learning, and providing a new solution for improving the quality of online education.

[0186] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0187] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, 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 generate 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 flowchart and / or block diagram. 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.

[0188] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate 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 A function specified in one or more boxes.

[0189] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0190] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A collaborative learning method based on multi-user eye movement data analysis, characterized by: include: Obtain the eye movement data of all users’ collaborative learning in the current stage; constructing an attention synchronization matrix according to the eye movement data, wherein the attention synchronization matrix includes the degree of attention synchronization between any two users; constructing an interaction intensity matrix according to the eye movement data, wherein the interaction intensity matrix includes interaction intensity between any two users; Calculate the overall collaboration index of all users according to the attention synchronization matrix and the interaction intensity matrix; According to the attention synchronization matrix, the interaction intensity matrix and the overall collaboration index, based on a preset collaborative learning model, a collaborative learning plan for the next stage is obtained.

2. The collaborative learning method based on multi-user eye movement data analysis according to claim 1 is characterized in that: The eye movement data of any user's collaborative learning includes user ID, gaze point, gaze duration, eye movement trajectory and pupil data.

3. The collaborative learning method based on multi-user eye movement data analysis according to claim 2 is characterized in that: The attention synchronization matrix is ​​an N×N attention synchronization matrix, and the expression is as follows: ； In the formula, represents the attention synchronization matrix; represents the attention synchronization between user i and user j, represents the attention synchronization between user j and user i, and The value of is the same, , where N represents the number of users.

4. The collaborative learning method based on multi-user eye movement data analysis according to claim 3 is characterized in that: The degree of attention synchronization The calculation steps are as follows: For any user i and user j, analyze and count the frequency of overlapping gaze points and the corresponding gaze duration in the eye movement data of the two users; Each time the gaze points overlap is regarded as an independent gaze point overlap event, and the number of gaze point overlap events is obtained according to the frequency of gaze point overlap; the duration of the corresponding gaze point overlap event is obtained according to the corresponding gaze duration; According to the number and duration of the gaze point overlap events, the attention synchronization between user i and user j is obtained. , the expression is as follows: ； In the formula, represents the duration of the kth gaze point overlap event, K represents the number of times the gaze points of user i and user j overlap; Indicates the total duration of the current stage.

5. The collaborative learning method based on multi-user eye movement data analysis according to claim 1 is characterized in that: The interaction intensity matrix is ​​an N×N interaction intensity matrix, and the expression is as follows: ； In the formula, represents the interaction intensity matrix; represents the interaction intensity between user i and user j, represents the interaction intensity between user j and user i, and The value of is the same, , where N represents the number of users.

6. The collaborative learning method based on multi-user eye movement data analysis according to claim 5 is characterized in that: The calculation steps of the interaction strength are as follows: For any user i and user j, by analyzing the gaze points and gaze duration, the frequency and average duration of the two users' gazes on each other are statistically obtained, including the frequency and average duration of user i's gaze on user j, and the frequency and average duration of user j's gaze on user i; According to the frequency and average duration of the two users looking at each other, the interaction intensity between user i and user j is obtained , the expression is as follows: ； In the formula, represents the frequency of user i looking at user j; represents the average time that user i looks at user j; represents the frequency with which user j looks at user i; represents the average time that user j looks at user i.

7. The collaborative learning method based on multi-user eye movement data analysis according to claim 1 is characterized in that: The expression of the overall collaboration index is as follows: ； In the formula, represents the overall collaboration index; represents the first weight coefficient; represents the attention synchronization between user i and user j; Indicates the number of users; represents the second weight coefficient; represents the interaction intensity between user i and user j.

8. The collaborative learning method based on multi-user eye movement data analysis according to claim 1 is characterized in that: The collaborative learning model includes: A collaboration bottleneck identification unit, used to obtain user pairs with collaboration bottlenecks according to the attention synchronization matrix and the interaction intensity matrix; A personalized collaboration unit, used for generating a next-stage attention synchronization plan and / or an interaction intensity plan for the user pair with a collaboration bottleneck; The overall collaboration unit is used to generate the next stage system learning tasks and user interface layout according to the overall collaboration index.

9. The collaborative learning method based on multi-user eye movement data analysis according to claim 1, characterized in that: The collaborative learning model further includes: A data analysis unit, used to obtain a collaboration status and a personalized report based on the attention synchronization matrix, the interaction intensity matrix and the overall collaboration index; The data storage unit is used to store the data of each unit.

10. A collaborative learning system based on multi-user eye movement data analysis, applicable to the collaborative learning method based on multi-user eye movement data analysis according to any one of claims 1 to 9, characterized in that: include: The data acquisition module is used to obtain the eye movement data of all users' collaborative learning in the current stage; An attention synchronization module, used to construct an attention synchronization matrix according to the eye movement data, wherein the attention synchronization matrix includes the degree of attention synchronization between any two users; An interaction strength module, used to construct an interaction strength matrix according to the eye movement data, wherein the interaction strength matrix includes the interaction strength between any two users; An overall collaboration module, used to calculate the overall collaboration index of all users according to the attention synchronization matrix and the interaction intensity matrix; The collaborative learning module is used to obtain a collaborative learning plan for the next stage based on a preset collaborative learning model according to the attention synchronization matrix, the interaction intensity matrix and the overall collaborative index.

Images