An interactive learning method and related device for pottery courses
Through interactive learning methods and data fusion algorithms, identify the learning status and weak knowledge points of autistic children, adjust the ceramic courses, solve the problems of unclear course division and incomplete learning status, improve the quality of learning and interest, and promote psychology and the transformation of social functions.
Patent Information
- Application Number
- CN202410219947.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-02-28
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-02-28
AI Technical Summary
The existing ceramic art course education has poor learning effect on autistic children, the course classification is unclear, the learning status is not integrated into the course, and the weak knowledge points are inaccurately matched, resulting in insufficient learning interest and unimproved quality.
Using interactive learning methods, grid division is performed by obtaining the task system of ceramic courses, real-time image acquisition and data fusion algorithms identify learning status, detect learning progress and master score rates, generate and consolidate learning paths, and adjust knowledge bases in combination with tree structures course.
It has improved the interest and concentration of autistic children in learning ceramic courses, scientifically and objectively reflects psychological and learning changes, and promoted the transformation of ceramic activities to realistic psychological and social functions.
Smart Images

Figure CN118096067B_ABST
Abstract
Description
Technical Field
[0001] The present invention mainly relates to the technical field of course learning, and in particular to an interactive learning method and related devices for a pottery course. Background Art
[0002] At present, music and painting teaching are usually used for auxiliary treatment of emotion perception and behavioral interaction for autistic children, while pottery teaching is rarely used for auxiliary treatment. However, pottery teaching has a more positive effect on the release of subconscious emotions of autistic children, and domestic ceramic art has a deeper cultural heritage and artistic expression. However, the current study of pottery courses still has problems such as unclear course division, judgment of learning status not integrated into the course, inaccurate matching of weak knowledge points, and no consideration of consolidation of learning based on weak knowledge points. The division of pottery courses is relatively general, and pottery courses are not divided according to relevant systems and themes. It will increase the difficulty of learning and make autistic children lack interest in learning. The judgment of learning status is not integrated into the course learning during the learning process, which will make it impossible to clearly understand the learning status of autistic children in the pottery course. At the same time, in the study of pottery courses, the matching accuracy of weak knowledge points is insufficient, which will make autistic children unable to clearly understand their own weak links in pottery course learning. Moreover, due to the problem of not considering consolidating learning according to weak knowledge points, the study of pottery courses will be insufficient, and the learning quality will not be effectively improved, which is not conducive to improving the overall learning of pottery courses for autistic children and the effect of pottery course education implementation. Summary of the invention
[0003] The purpose of the present invention is to overcome the shortcomings of the prior art. The present invention provides an interactive learning method and related devices for a pottery course, which not only ensures the effect of implementing the pottery course education for autistic children, but also helps to promote the transformation of autistic children's pottery activities into realistic psychological functions and social functions.
[0004] In order to solve the above technical problems, the present invention provides an interactive learning method for a pottery course, the method comprising:
[0005] Obtain a task system for a pottery course, divide the pottery course into grids based on the task system for the pottery course, and obtain a number of pottery course grids;
[0006] Continuously collecting real-time images of the learning object in the process of learning the ceramic art course grid based on the image acquisition device, performing learning state recognition processing based on the real-time image using a data fusion algorithm, and sending a reminder message to the learning object if it is recognized that the learning object is in an abnormal learning state;
[0007] Detecting the learning progress of each ceramics course grid, and when it is detected that the learning progress of the corresponding ceramics course grid reaches a preset standard value, generating a course test by combining the learning interaction data between the learning object and the corresponding ceramics course grid, and calculating the learning mastery score rate of the learning object based on the test answer data generated by the learning object according to the course test;
[0008] If the learning mastery score rate reaches the preset threshold, the learning of the ceramic art course grid is continued; if the learning mastery score rate does not reach the preset threshold, a number of weak knowledge points are obtained based on the learning mastery score rate using a knowledge base with a tree structure, and a consolidation learning path is generated based on the weak knowledge points, and the consolidation learning path is sent to the corresponding learning object;
[0009] When it is detected that the learning progress of all pottery course grids has reached a preset total value, the learning mastery score rate of all pottery course grids is evaluated, and a course improvement plan is determined based on the evaluation result combined with the learning state recognition result of the learning object and the consolidation learning path, and the course improvement plan is visualized.
[0010] Optionally, the step of obtaining a task system for a pottery course and dividing the pottery course into grids based on the task system for the pottery course to obtain a plurality of pottery course grids includes:
[0011] Based on the preset platform, the task system of ceramics courses and the total ceramics course resource data set are obtained;
[0012] Performing data format conversion on the total ceramics course resource data set to obtain a standard total ceramics course resource data set;
[0013] Performing data stream classification on the standard total pottery course resource dataset to obtain a pottery course teaching video dataset and a pottery course teaching text dataset;
[0014] Performing word embedding encoding processing on the ceramic art course teaching text dataset to obtain a text semantic feature dataset;
[0015] Performing feature extraction processing on the ceramic art course teaching video dataset to obtain a video feature dataset;
[0016] The text semantic feature dataset and the video feature dataset are integrated and processed with the same subject classification to obtain a classified pottery course resource dataset. The classified pottery course resource dataset is combined with the task system of the pottery course to divide the course grid according to the preset difficulty level to obtain a plurality of pottery course grids.
[0017] Optionally, the performing learning state recognition processing based on the real-time image using a data fusion algorithm includes:
[0018] Performing face detection processing on the learning object based on the real-time image, if the face of the learning object is not detected, determining the learning state as an absent learning state, if the face of the learning object is detected, locating the face area of the learning object;
[0019] Performing facial feature point positioning and eye feature point positioning on the face region based on a neural domain feature point positioning algorithm to obtain facial feature points and eye feature points;
[0020] Calculating fatigue state parameters and distraction state parameters based on the facial feature points and eye feature points using a data fusion algorithm combined with a proportional coefficient and a correction parameter;
[0021] The fatigue state parameter is compared with a first preset threshold value. If the fatigue state parameter is greater than or equal to the first preset threshold value, the learning state is judged to be a fatigue state. The distraction state parameter is compared with a second preset threshold value. If the distraction state parameter is greater than or equal to the second preset threshold value, the learning state is judged to be a distraction state.
[0022] Optionally, generating a course test by combining the learning interaction data between the learning object and the corresponding ceramics course grid, and calculating the learning mastery score rate of the learning object based on the test answer data generated by the learning object according to the course test, includes:
[0023] Constructing a correspondence between the learning interaction data and the course test type, and generating a first target test question based on the correspondence;
[0024] Generate a second target test question based on the knowledge points of the corresponding ceramics course grid, and construct a course test based on the first target test question and the second target test question;
[0025] The test answer data generated by the learning object according to the course test is obtained, and the learning mastery score rate of the learning object is calculated based on the test answer data using the test score and difficulty coefficient of each test question in the course test.
[0026] Optionally, the method of obtaining a plurality of weak knowledge points based on the learning mastery score rate using a knowledge base having a tree structure includes:
[0027] Calculating association weights for relevant knowledge points in each ceramics course grid, constructing a knowledge tree based on the association weights, and constructing a knowledge base with a tree structure based on the knowledge tree;
[0028] Constructing a mastery matrix of relevant knowledge points based on the learning mastery score rate;
[0029] Based on the mastery matrix, a knowledge base with a tree structure is used to obtain several weak knowledge points of the learning object.
[0030] Optionally, generating a consolidation learning path based on the weak knowledge points includes:
[0031] Count the proportion of each weak knowledge point;
[0032] For the weak knowledge points, using a knowledge base with a tree structure to obtain corresponding pre-knowledge points and post-knowledge points, and forming a learning path relationship between the weak knowledge points based on the pre-knowledge points and post-knowledge points;
[0033] A consolidation learning path is constructed based on the learning path relationship and the proportion of each weak knowledge point.
[0034] Optionally, the learning mastery score rates of all ceramic art course grids are evaluated, and a course improvement plan is determined based on the evaluation results combined with the learning state identification results of the learning objects and the consolidation learning path, including:
[0035] The learning mastery score rate of each ceramics course grid is calculated according to the preset weight, and the learning effect of the learning object is evaluated based on the calculation result using the preset hierarchical category to obtain the evaluation result;
[0036] Based on the evaluation results combined with the learning state recognition results and learning interaction data of the learning object, the learning object's knowledge points of interest are obtained, and a course improvement plan is determined based on the knowledge points of interest and the consolidation learning path.
[0037] In addition, the present invention also provides an interactive learning device for a pottery course, the device comprising:
[0038] Course grid division module: used to obtain the task system of the pottery course, divide the pottery course into grids based on the task system of the pottery course, and obtain several pottery course grids;
[0039] Learning state recognition module: used for continuously collecting real-time images of the learning object in the process of learning the ceramic course grid based on the image acquisition device, and performing learning state recognition processing based on the real-time image using the data fusion algorithm. If the learning object is identified to be in an abnormal learning state, a reminder message is issued to the learning object;
[0040] Learning mastery score rate calculation module: used to detect the learning progress of each ceramic course grid. When it is detected that the learning progress of the corresponding ceramic course grid reaches a preset standard value, a course test is generated by combining the learning interaction data of the learning object and the corresponding ceramic course grid, and the learning mastery score rate of the learning object is calculated based on the test answer data generated by the learning object according to the course test;
[0041] A consolidation learning path generation module: used for continuing the learning of the ceramic art course grid if the learning mastery score rate reaches a preset threshold; if the learning mastery score rate does not reach the preset threshold, obtaining a number of weak knowledge points based on the learning mastery score rate using a knowledge base with a tree structure, and generating a consolidation learning path based on the weak knowledge points, and sending the consolidation learning path to the corresponding learning object;
[0042] Learning evaluation and course improvement module: used to evaluate the learning mastery score rate of all pottery course grids when it is detected that the learning progress of all pottery course grids has reached the preset total value, determine the course improvement plan based on the evaluation results combined with the learning status recognition results of the learning objects and the consolidation learning path, and visualize the course improvement plan.
[0043] In addition, the present invention also provides an electronic device, which includes a processor and a memory, wherein the memory is used to store instructions, and the processor is used to call the instructions in the memory, so that the electronic device executes the above-mentioned interactive learning method of pottery courses.
[0044] In addition, the present invention also provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes the above-mentioned interactive learning method for pottery courses.
[0045] In the embodiment of the present invention, the pottery course is divided into several pottery course grids through the task system of the pottery course, so that the pottery course can be studied from shallow to deep according to the subject classification and difficulty level, so that autistic children can better understand the course content and improve their learning interest. At the same time, the learning state is identified through the data fusion algorithm, so as to further improve the learning concentration. The learning behavior of the autistic children can be known through the learning state identification, and then it can be judged which knowledge point they are more focused on and interested in. After the study of the pottery course at each stage, the learning mastery score rate is used to obtain several weak knowledge points using a knowledge base with a tree structure, and a consolidation learning path is generated based on the weak knowledge points, which improves the learning quality and can scientifically and objectively reflect the psychological and learning change process of autistic children, which not only ensures the effect of the implementation of the pottery course education for autistic children, but also is beneficial to promote the transformation of autistic children's pottery activities into realistic psychological functions and social functions. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0047] Figure 1 is a flow chart of an interactive learning method for a pottery course in an embodiment of the present invention;
[0048] Figure 2 is a schematic diagram of the structure of an interactive learning device for a pottery course in an embodiment of the present invention;
[0049] Figure 3 It is a schematic diagram of the structure of an electronic device in an embodiment of the present invention. DETAILED DESCRIPTION
[0050] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0051] Embodiment 1
[0052] See also Figure 1 , Figure 1 It is a flowchart of an interactive learning method for a pottery course in an embodiment of the present invention.
[0053] like Figure 1 An interactive learning method for a pottery course is shown, the method comprising:
[0054] S11: obtaining a task system of a pottery course, dividing the pottery course into grids based on the task system of the pottery course, and obtaining a plurality of pottery course grids;
[0055] In the specific implementation process of the present invention, the task system of the ceramics course is obtained, and the ceramics course is grid-divided based on the task system of the ceramics course to obtain a number of ceramics course grids, including: obtaining the task system of the ceramics course and the total ceramics course resource data set based on a preset platform; performing data format conversion on the total ceramics course resource data set to obtain a standard total ceramics course resource data set; performing data stream classification on the standard total ceramics course resource data set to obtain a ceramics course teaching video data set and a ceramics course teaching text data set; performing word embedding encoding processing on the ceramics course teaching text data set to obtain a text semantic feature data set; performing feature extraction processing on the ceramics course teaching video data set to obtain a video feature data set; performing same-theme classification and integration processing on the text semantic feature data set and the video feature data set to obtain a classified ceramics course resource data set; dividing the classified ceramics course resource data set into a course grid according to a preset difficulty level in combination with the task system of the ceramics course to obtain a number of ceramics course grids.
[0056] Specifically, based on a preset platform, a task system of a ceramics course and a total ceramics course resource data set are obtained. The task system of the ceramics course includes online understanding of ceramics culture and crafts, online preview videos and floor plan design learning, etc.; data format conversion is performed on the total ceramics course resource data set. Before the data format conversion, duplicate data removal processing is also required for the total ceramics course resource data set. A hash algorithm can be used to compare and remove duplicates using resource identifiers to ensure that the total ceramics course resource data set does not contain duplicate course resources. The total ceramics course resource data set after duplicate data removal is converted into a unified data format, the fields and attributes of the data set are defined, and the data types and data constraints of the data set are selected. Based on the defined fields and attributes and the selected data types and data constraints, the total ceramics course resource data set after duplicate data removal is traversed and converted to obtain a standardized total ceramics course resource data set; data stream classification is performed on the standardized total ceramics course resource data set, and the standardized total ceramics course resource data set is modally divided by a deep neural network model, and classified in combination with the attributes of the data set. , obtain a ceramics course teaching video dataset and a ceramics course teaching text dataset; perform word embedding encoding processing on the ceramics course teaching text dataset, obtain a bidirectional autoregressive vocabulary, perform vocabulary expansion on the bidirectional autoregressive vocabulary, obtain an expanded bidirectional autoregressive vocabulary, pre-train the expanded bidirectional autoregressive vocabulary, obtain a vocabulary embedding matrix, and perform word embedding on the ceramics course teaching text dataset through the vocabulary embedding matrix to obtain a text semantic feature dataset; perform feature extraction processing on the ceramics course teaching video dataset, extract frames of the ceramics course teaching video dataset, obtain several ceramics course teaching video frames, perform multi-level pooling processing on each ceramics course teaching video frame, determine the corresponding feature matrix based on each ceramics course teaching video frame, perform multi-level pooling processing on the corresponding feature matrix, that is, reduce the corresponding feature matrix step by step until it is reduced to a feature matrix containing only one point, obtain several ceramics course teaching video frames after multi-level pooling processing, and perform splicing processing on several ceramics course teaching video frames after multi-level pooling processing to obtain a video feature dataset;The text semantic feature data set and the video feature data set are integrated for the same subject classification, a likelihood function is constructed, the text semantic feature data set and the video feature data set are input into the likelihood function, a subject feature vector is obtained, the similarity of each two feature data sets is calculated according to the subject feature vector, and two data sets whose similarity reaches a preset threshold are classified into the same subject classification until the text semantic feature data set and the video feature data set are classified to obtain a classified pottery course resource data set, the classified pottery course resource data set is divided into a course grid according to a preset difficulty level in combination with the task system of the pottery course, and a number of pottery course grids are obtained, and the learning of each pottery course grid is progressive, and the pottery course is divided into a number of pottery course grids through the task system of the pottery course, so that the pottery course can be studied from shallow to deep according to the subject classification and difficulty level, so that children with autism can better understand the course content and improve their interest in learning. ;
[0057] S12: continuously collecting real-time images of the learning object in the process of learning the ceramic art course grid based on the image acquisition device, performing learning state recognition processing based on the real-time images using a data fusion algorithm, and sending a reminder message to the learning object if it is recognized that the learning object is in an abnormal learning state;
[0058] In the specific implementation process of the present invention, the learning state recognition processing based on the real-time image using the data fusion algorithm includes: performing face detection processing on the learning object based on the real-time image, if the face of the learning object is not detected, then judging the learning state to be an absent learning state, if the face of the learning object is detected, then locating the face area of the learning object; locating facial feature points and eye feature points of the face area based on the neural domain feature point positioning algorithm to obtain facial feature points and eye feature points; calculating fatigue state parameters and distraction state parameters based on the facial feature points and eye feature points using the data fusion algorithm combined with the proportional coefficient and the correction parameter; comparing the fatigue state parameter with a first preset threshold value, if the fatigue state parameter is greater than or equal to the first preset threshold value, then judging the learning state to be a fatigue state, comparing the distraction state parameter with a second preset threshold value, if the distraction state parameter is greater than or equal to the second preset threshold value, then judging the learning state to be a distraction state.
[0059] Specifically, the learning object is an autistic child. The learning object learns each pottery course grid on the learning platform. The real-time image of the learning object in the process of learning the pottery course grid is continuously collected by an image acquisition device. The face detection processing is performed on the learning object based on the real-time image. The face detection is performed using an opencv algorithm. If the face of the learning object is not detected, the learning state is judged to be an absent learning state. If the face of the learning object is detected, a two-dimensional face shape model and a local texture model are established. The posture parameters and shape parameters are determined based on the two-dimensional face shape model and the local texture model. The face area image is accurately located according to the posture parameters and the shape parameters to obtain the face area of the learning object; the facial feature points and eye feature points of the face area are located based on the neural domain feature point positioning algorithm, and the face area image is similarly transformed, that is, the face area image is rotated, scaled and translated, and the features in the face area image after the similarity transformation are detected by the neural domain feature points. points, respectively set facial feature constraints and eye feature constraints, frame the facial image feature points according to the facial feature constraints to obtain facial feature points, frame the eye feature points according to the eye feature constraints to obtain eye feature points; based on the facial feature points and eye feature points, use the data fusion algorithm in combination with the proportional coefficient and the correction parameter to calculate the fatigue state parameter and the distraction state parameter, calculate the mouth opening degree, eye opening degree, blinking frequency, yaw angle of the head in the vertical direction and horizontal direction, and horizontal deflection angle of the eye gaze direction through the facial feature points and the eye feature points in combination with the correction parameter and the proportional coefficient, and construct the fatigue state contribution value and the distraction state contribution value through the mouth opening degree, eye opening degree, blinking frequency, yaw angle of the head in the vertical direction and horizontal direction, and horizontal deflection angle of the eye gaze direction through the data fusion algorithm. The data fusion algorithm is to fuse different data to obtain more useful information, and calculate the fatigue state parameter through the fatigue state contribution value. The calculation expression of the fatigue state parameter is:
[0060] T=N1*t N +D T *t T ,
[0061] Among them, T is the fatigue state parameter, N1 is the normal state contribution value, t N is the normal state time, D T is the fatigue state contribution value, t Tis the fatigue state time, and the distraction state parameter is also calculated according to the distraction state contribution value; the fatigue state parameter is compared with the first preset threshold value, if the fatigue state parameter is greater than or equal to the first preset threshold value, the learning state is judged to be a fatigue state, if the fatigue state parameter is less than the first preset threshold value, the learning state is judged to be a normal learning state, the distraction state parameter is compared with the second preset threshold value, if the distraction state parameter is greater than or equal to the second preset threshold value, the learning state is judged to be a distraction state, if the distraction state parameter is less than the second preset threshold value, the learning state is judged to be a non-distracted state, when it is judged that the learning object is in a distracted state or a fatigue state, the learning object can be reminded through voice prompts, and when it is identified that the learning state of the learning object is an abnormal learning state, the current recognition time and the teaching resource learning position in the corresponding pottery course grid, that is, the video viewing position and the text browsing position, etc., will be marked, and the learning state recognition is performed through the data fusion algorithm, so as to further improve the learning concentration, and the learning behavior of autistic children can be known through learning state recognition, and then it can be judged which knowledge point they are more focused on and interested in.
[0062] S13: Detecting the learning progress of each ceramics course grid. When it is detected that the learning progress of the corresponding ceramics course grid reaches a preset standard value, generating a course test by combining the learning interaction data between the learning object and the corresponding ceramics course grid, and calculating the learning mastery score rate of the learning object based on the test answer data generated by the learning object according to the course test;
[0063] In the specific implementation process of the present invention, the learning interaction data of the learning object and the corresponding pottery course grid is combined to generate a course test, and the learning mastery score rate of the learning object is calculated based on the test answer data generated by the learning object according to the course test, including: constructing a correspondence between the learning interaction data and the course test type, and generating a first target test question based on the correspondence; generating a second target test question based on the knowledge points of the corresponding pottery course grid, and constructing a course test based on the first target test question and the second target test question; obtaining the test answer data generated by the learning object according to the course test, and calculating the learning mastery score rate of the learning object based on the test answer data using the test score and difficulty coefficient of each question in the course test.
[0064] Specifically, the learning progress of each ceramic course grid is detected, and the learning progress of each ceramic course grid is detected through the video viewing progress and the text reading progress. When it is detected that the learning progress of the corresponding ceramic course grid reaches the preset standard value, a course test is generated by combining the learning interaction data between the learning object and the corresponding ceramic course grid. The learning interaction data is the interaction behavior data between the learning object and the corresponding ceramic course grid, such as the operation behavior data on the video text and the operation behavior data on the teaching text, etc., and a corresponding relationship between the learning interaction data and the course test type is constructed. The number of operations on the video teaching resources of the ceramic course grid or the reading time of the teaching text in the learning interaction data can be used to roughly understand the learning object's interested knowledge points or difficult to understand knowledge points. According to the interested knowledge points or difficult to understand knowledge points, a correspondence between the course test type can be established. A corresponding relationship is generated, and a number of corresponding test questions can be searched in the test question database based on the corresponding relationship to form a first target test question; a second target test question is generated based on the knowledge point of the corresponding ceramic course grid, and the knowledge point of the corresponding ceramic course grid also has a number of corresponding test questions in the test question database, and the number of test questions are combined to form a second target test question, and a course test is constructed based on the first target test question and the second target test question; the course test is sent to the learning object, and the learning object answers according to the course test, and test answer data is generated, and the test answer data is obtained, and the learning mastery score rate of the learning object is calculated based on the test answer data using the test score and difficulty coefficient of each test question in the course test, the test score and difficulty coefficient of each test question are calculated, and then the calculation results of all the test questions are comprehensively counted to obtain the learning mastery score rate.
[0065] S14: If the learning mastery score rate reaches a preset threshold, the learning of the ceramic art course grid is continued; if the learning mastery score rate does not reach the preset threshold, a plurality of weak knowledge points are obtained based on the learning mastery score rate using a knowledge base with a tree structure, and a consolidation learning path is generated based on the weak knowledge points, and the consolidation learning path is sent to the corresponding learning object;
[0066] In the specific implementation process of the present invention, the method of obtaining a number of weak knowledge points based on the learning mastery score rate using a knowledge base with a tree structure includes: calculating association weights for relevant knowledge points in each pottery course grid, constructing a knowledge tree based on the association weights, and constructing a knowledge base with a tree structure based on the knowledge tree; constructing a mastery matrix of relevant knowledge points based on the learning mastery score rate; and obtaining a number of weak knowledge points of a learning object based on the mastery matrix using a knowledge base with a tree structure.
[0067] Furthermore, the generation of a consolidation learning path based on the weak knowledge points includes: counting the proportion of each weak knowledge point; obtaining corresponding preceding knowledge points and succeeding knowledge points for the weak knowledge points using a knowledge base with a tree structure, and constructing a learning path relationship between the weak knowledge points based on the preceding knowledge points and succeeding knowledge points; and constructing a consolidation learning path based on the learning path relationship and the proportion of each weak knowledge point.
[0068] Specifically, the learning mastery score is compared with a preset threshold value. If the learning mastery score rate reaches the preset threshold value, the learning of the ceramic course grid is continued. If the learning mastery score rate does not reach the preset threshold value, the relevant knowledge points in each ceramic course grid are traversed in turn, and the frequency of occurrence of two adjacent relevant knowledge points in the corresponding ceramic course grid is calculated. The frequency of occurrence is used as an initial weight value, and then the initial weight value is adjusted by a preset ratio to obtain an associated weight. The relevant knowledge points are added to the knowledge tree in turn according to the associated relationship, starting from the root node of the knowledge tree. The first related knowledge point added is connected to the root node, and its connection strength is set to zero. The associated weight is used as the connection strength between the remaining related knowledge points. After all knowledge points are added to the corresponding knowledge tree, a knowledge base with a tree structure is established according to the constructed knowledge tree; a mastery matrix of the relevant knowledge points is constructed based on the learning mastery score rate, and the learning ability of the corresponding knowledge point can be obtained according to the learning mastery score rate. The relevant knowledge points are constructed based on the learning ability of the corresponding knowledge point combined with the complexity of the knowledge point. A mastery matrix; based on the mastery matrix, a knowledge base with a tree structure is used to obtain several weak knowledge points of the learning object. Based on the mastery matrix, the overall mastery rating of the learning object can be determined. The knowledge points designed in this course test are sorted according to the overall mastery rating, and the knowledge points with the last digit in the sorting are used as the initial weak points. The knowledge base with a tree structure is used to obtain the association relationship between the initial weak points, and several knowledge weak points are finally determined through the association relationship to prevent omissions in the search for knowledge weak points due to accidental factors; the proportion of each weak knowledge point is counted, and the proportion of each weak knowledge point in the corresponding ceramic course grid is counted, that is, the number of occurrences of the corresponding weak knowledge point in the corresponding ceramic course grid is counted; the knowledge base with a tree structure is used to obtain the corresponding pre-knowledge point and post-knowledge point for the weak knowledge point. The knowledge base with a tree structure stores the association relationship of each knowledge point, so the corresponding pre-knowledge point and post-knowledge point can be quickly found, and the learning path relationship between the weak knowledge points is constructed based on the pre-knowledge point and post-knowledge point;Based on the learning path relationship and the proportion of each weak knowledge point, a consolidation learning path is constructed. Each weak knowledge point is connected in series through the learning path relationship. At the same time, the proportion of teaching resources and learning time for consolidation learning can be determined according to the proportion of weak knowledge points. The learning object re-learns according to the consolidation learning path. Since autistic children have certain obstacles in social interaction and communication, they usually lack communication with teachers about their own weak links in the course, resulting in insufficient learning of the pottery course. Therefore, a number of weak knowledge points are obtained by using a knowledge base with a tree structure through learning mastery score rate, and a consolidation learning path is generated based on the weak knowledge points, which can effectively avoid the problem of insufficient learning. In addition, the learning detection of each pottery course grid can not only improve the learning quality, but also scientifically and objectively reflect the psychological and learning changes of autistic children. ;
[0069] S15: When it is detected that the learning progress of all pottery course grids reaches a preset total value, the learning mastery score rate of all pottery course grids is evaluated, and a course improvement plan is determined based on the evaluation result combined with the learning state recognition result of the learning object and the consolidation learning path, and the course improvement plan is visualized.
[0070] In the specific implementation process of the present invention, the learning mastery score rates of all pottery course grids are evaluated, and a course improvement plan is determined based on the evaluation results combined with the learning state recognition results and consolidation learning paths of the learning objects, including: calculating the learning mastery score rate of each pottery course grid according to a preset weight, and evaluating the learning effect of the learning object based on the calculation results using preset hierarchical categories to obtain evaluation results; obtaining the learning object's knowledge points of interest based on the evaluation results combined with the learning state recognition results and learning interaction data of the learning object, and determining the course improvement plan based on the knowledge points of interest and the consolidation learning paths.
[0071] Specifically, when it is detected that the learning progress of all ceramics course grids has reached a preset total value, the learning mastery score rate of each ceramics course grid is calculated according to the preset weight, the learning mastery score rate of each ceramics course grid is multiplied by the preset weight, the multiplication results are accumulated to obtain the calculation result, and based on the calculation result, the learning effect of the learning object is evaluated using the preset hierarchical category to obtain the evaluation result; based on the evaluation result, the learning state recognition result and the learning interaction data of the learning object are combined to obtain the learning object's interested knowledge points, and the course improvement plan is determined based on the interested knowledge points and the consolidation learning path, and the learning state recognition result and the learning interaction data of the learning object are combined to obtain the learning object's interested knowledge points. Knowledge points of interest, the learning status identification results of the learning objects can tell at which knowledge points the learning objects are more likely to be distracted or produce learning fatigue, and the learning interaction data can tell at which knowledge points the learning objects have more operation behaviors in the resource data. The initial curriculum plan can be improved through the knowledge points of interest and the consolidation of learning paths, so that the curriculum improvement plan is more in line with the interests of autistic children, and the curriculum improvement plan is visualized. The curriculum improvement plan can be visualized in the form of charts to improve the learning interest of autistic children, which not only ensures the effect of the implementation of pottery course education for autistic children, but also is beneficial to promote the transformation of autistic children's pottery activities into realistic psychological functions and social functions.
[0072] In the embodiment of the present invention, the pottery course is divided into several pottery course grids through the task system of the pottery course, so that the pottery course can be studied from shallow to deep according to the subject classification and difficulty level, so that autistic children can better understand the course content and improve their learning interest. At the same time, the learning state is identified through the data fusion algorithm, so as to further improve the learning concentration. The learning behavior of the autistic children can be known through the learning state identification, and then it can be judged which knowledge point they are more focused on and interested in. After the study of the pottery course at each stage, the learning mastery score rate is used to obtain several weak knowledge points using a knowledge base with a tree structure, and a consolidation learning path is generated based on the weak knowledge points, which improves the learning quality and can scientifically and objectively reflect the psychological and learning change process of autistic children, which not only ensures the effect of the implementation of the pottery course education for autistic children, but also is beneficial to promote the transformation of autistic children's pottery activities into realistic psychological functions and social functions.
[0073] Embodiment 2
[0074] See also Figure 2 , Figure 2 Schematic diagram of the structure of an interactive learning device for pottery courses in an embodiment of the present invention.
[0075] like Figure 2 As shown, an interactive learning device for a pottery course, the device comprising:
[0076] The course grid division module 21 is used to obtain the task system of the pottery course, divide the pottery course into grids based on the task system of the pottery course, and obtain a plurality of pottery course grids;
[0077] In the specific implementation process of the present invention, the task system of the ceramics course is obtained, and the ceramics course is grid-divided based on the task system of the ceramics course to obtain a number of ceramics course grids, including: obtaining the task system of the ceramics course and the total ceramics course resource data set based on a preset platform; performing data format conversion on the total ceramics course resource data set to obtain a standard total ceramics course resource data set; performing data stream classification on the standard total ceramics course resource data set to obtain a ceramics course teaching video data set and a ceramics course teaching text data set; performing word embedding encoding processing on the ceramics course teaching text data set to obtain a text semantic feature data set; performing feature extraction processing on the ceramics course teaching video data set to obtain a video feature data set; performing same-theme classification and integration processing on the text semantic feature data set and the video feature data set to obtain a classified ceramics course resource data set; dividing the classified ceramics course resource data set into a course grid according to a preset difficulty level in combination with the task system of the ceramics course to obtain a number of ceramics course grids.
[0078] Specifically, based on a preset platform, a task system of a ceramics course and a total ceramics course resource data set are obtained. The task system of the ceramics course includes online understanding of ceramics culture and crafts, online preview videos and floor plan design learning, etc.; data format conversion is performed on the total ceramics course resource data set. Before the data format conversion, duplicate data removal processing is also required for the total ceramics course resource data set. A hash algorithm can be used to compare and remove duplicates using resource identifiers to ensure that the total ceramics course resource data set does not contain duplicate course resources. The total ceramics course resource data set after duplicate data removal is converted into a unified data format, the fields and attributes of the data set are defined, and the data types and data constraints of the data set are selected. Based on the defined fields and attributes and the selected data types and data constraints, the total ceramics course resource data set after duplicate data removal is traversed and converted to obtain a standardized total ceramics course resource data set; data stream classification is performed on the standardized total ceramics course resource data set, and the standardized total ceramics course resource data set is modally divided by a deep neural network model, and classified in combination with the attributes of the data set. , obtain a ceramics course teaching video dataset and a ceramics course teaching text dataset; perform word embedding encoding processing on the ceramics course teaching text dataset, obtain a bidirectional autoregressive vocabulary, perform vocabulary expansion on the bidirectional autoregressive vocabulary, obtain an expanded bidirectional autoregressive vocabulary, pre-train the expanded bidirectional autoregressive vocabulary, obtain a vocabulary embedding matrix, and perform word embedding on the ceramics course teaching text dataset through the vocabulary embedding matrix to obtain a text semantic feature dataset; perform feature extraction processing on the ceramics course teaching video dataset, extract frames of the ceramics course teaching video dataset, obtain several ceramics course teaching video frames, perform multi-level pooling processing on each ceramics course teaching video frame, determine the corresponding feature matrix based on each ceramics course teaching video frame, perform multi-level pooling processing on the corresponding feature matrix, that is, reduce the corresponding feature matrix step by step until it is reduced to a feature matrix containing only one point, obtain several ceramics course teaching video frames after multi-level pooling processing, and perform splicing processing on several ceramics course teaching video frames after multi-level pooling processing to obtain a video feature dataset;The text semantic feature data set and the video feature data set are integrated for the same subject classification, a likelihood function is constructed, the text semantic feature data set and the video feature data set are input into the likelihood function, a subject feature vector is obtained, the similarity of each two feature data sets is calculated according to the subject feature vector, and two data sets whose similarity reaches a preset threshold are classified into the same subject classification until the text semantic feature data set and the video feature data set are classified to obtain a classified pottery course resource data set, the classified pottery course resource data set is divided into a course grid according to a preset difficulty level in combination with the task system of the pottery course, and a number of pottery course grids are obtained, and the learning of each pottery course grid is progressive, and the pottery course is divided into a number of pottery course grids through the task system of the pottery course, so that the pottery course can be studied from shallow to deep according to the subject classification and difficulty level, so that children with autism can better understand the course content and improve their interest in learning. ;
[0079] The learning state recognition module 22 is used to continuously collect real-time images of the learning object in the process of learning the ceramic course grid based on the image acquisition device, and perform learning state recognition processing based on the real-time image using a data fusion algorithm. If it is recognized that the learning object is in an abnormal learning state, a reminder message is issued to the learning object;
[0080] In the specific implementation process of the present invention, the learning state recognition processing based on the real-time image using the data fusion algorithm includes: performing face detection processing on the learning object based on the real-time image, if the face of the learning object is not detected, then judging the learning state to be an absent learning state, if the face of the learning object is detected, then locating the face area of the learning object; locating facial feature points and eye feature points of the face area based on the neural domain feature point positioning algorithm to obtain facial feature points and eye feature points; calculating fatigue state parameters and distraction state parameters based on the facial feature points and eye feature points using the data fusion algorithm combined with the proportional coefficient and the correction parameter; comparing the fatigue state parameter with a first preset threshold value, if the fatigue state parameter is greater than or equal to the first preset threshold value, then judging the learning state to be a fatigue state, comparing the distraction state parameter with a second preset threshold value, if the distraction state parameter is greater than or equal to the second preset threshold value, then judging the learning state to be a distraction state.
[0081] Specifically, the learning object is an autistic child. The learning object learns each pottery course grid on the learning platform. The real-time image of the learning object in the process of learning the pottery course grid is continuously collected by an image acquisition device. The face detection processing is performed on the learning object based on the real-time image. The face detection is performed using an opencv algorithm. If the face of the learning object is not detected, the learning state is judged to be an absent learning state. If the face of the learning object is detected, a two-dimensional face shape model and a local texture model are established. The posture parameters and shape parameters are determined based on the two-dimensional face shape model and the local texture model. The face area image is accurately located according to the posture parameters and the shape parameters to obtain the face area of the learning object; the facial feature points and eye feature points of the face area are located based on the neural domain feature point positioning algorithm, and the face area image is similarly transformed, that is, the face area image is rotated, scaled and translated, and the features in the face area image after the similarity transformation are detected by the neural domain feature points. points, respectively set facial feature constraints and eye feature constraints, frame the facial image feature points according to the facial feature constraints to obtain facial feature points, frame the eye feature points according to the eye feature constraints to obtain eye feature points; based on the facial feature points and eye feature points, use the data fusion algorithm in combination with the proportional coefficient and the correction parameter to calculate the fatigue state parameter and the distraction state parameter, calculate the mouth opening degree, eye opening degree, blinking frequency, yaw angle of the head in the vertical direction and horizontal direction, and horizontal deflection angle of the eye gaze direction through the facial feature points and the eye feature points in combination with the correction parameter and the proportional coefficient, and construct the fatigue state contribution value and the distraction state contribution value through the mouth opening degree, eye opening degree, blinking frequency, yaw angle of the head in the vertical direction and horizontal direction, and horizontal deflection angle of the eye gaze direction through the data fusion algorithm. The data fusion algorithm is to fuse different data to obtain more useful information, and calculate the fatigue state parameter through the fatigue state contribution value. The calculation expression of the fatigue state parameter is:
[0082] T=N1*t N +D T *t T ,
[0083] Among them, T is the fatigue state parameter, N1 is the normal state contribution value, t N is the normal state time, D T is the fatigue state contribution value, t Tis the fatigue state time, and the distraction state parameter is also calculated according to the distraction state contribution value; the fatigue state parameter is compared with the first preset threshold value, if the fatigue state parameter is greater than or equal to the first preset threshold value, the learning state is judged to be a fatigue state, if the fatigue state parameter is less than the first preset threshold value, the learning state is judged to be a normal learning state, the distraction state parameter is compared with the second preset threshold value, if the distraction state parameter is greater than or equal to the second preset threshold value, the learning state is judged to be a distraction state, if the distraction state parameter is less than the second preset threshold value, the learning state is judged to be a non-distracted state, when it is judged that the learning object is in a distracted state or a fatigue state, the learning object can be reminded through voice prompts, and when it is identified that the learning state of the learning object is an abnormal learning state, the current recognition time and the teaching resource learning position in the corresponding pottery course grid, that is, the video viewing position and the text browsing position, etc., will be marked, and the learning state recognition is performed through the data fusion algorithm, so as to further improve the learning concentration, and the learning behavior of autistic children can be known through learning state recognition, and then it can be judged which knowledge point they are more focused on and interested in.
[0084] The learning mastery score rate calculation module 23 is used to detect the learning progress of each ceramic course grid. When it is detected that the learning progress of the corresponding ceramic course grid reaches a preset standard value, a course test is generated by combining the learning interaction data between the learning object and the corresponding ceramic course grid, and the learning mastery score rate of the learning object is calculated based on the test answer data generated by the learning object according to the course test;
[0085] In the specific implementation process of the present invention, the learning interaction data of the learning object and the corresponding pottery course grid is combined to generate a course test, and the learning mastery score rate of the learning object is calculated based on the test answer data generated by the learning object according to the course test, including: constructing a correspondence between the learning interaction data and the course test type, and generating a first target test question based on the correspondence; generating a second target test question based on the knowledge points of the corresponding pottery course grid, and constructing a course test based on the first target test question and the second target test question; obtaining the test answer data generated by the learning object according to the course test, and calculating the learning mastery score rate of the learning object based on the test answer data using the test score and difficulty coefficient of each question in the course test.
[0086] Specifically, the learning progress of each ceramic course grid is detected, and the learning progress of each ceramic course grid is detected through the video viewing progress and the text reading progress. When it is detected that the learning progress of the corresponding ceramic course grid reaches the preset standard value, a course test is generated by combining the learning interaction data between the learning object and the corresponding ceramic course grid. The learning interaction data is the interaction behavior data between the learning object and the corresponding ceramic course grid, such as the operation behavior data on the video text and the operation behavior data on the teaching text, etc., and a corresponding relationship between the learning interaction data and the course test type is constructed. The number of operations on the video teaching resources of the ceramic course grid or the reading time of the teaching text in the learning interaction data can be used to roughly understand the learning object's interested knowledge points or difficult to understand knowledge points. According to the interested knowledge points or difficult to understand knowledge points, a correspondence between the course test type can be established. A corresponding relationship is generated, and a number of corresponding test questions can be searched in the test question database based on the corresponding relationship to form a first target test question; a second target test question is generated based on the knowledge point of the corresponding ceramic course grid, and the knowledge point of the corresponding ceramic course grid also has a number of corresponding test questions in the test question database, and the number of test questions are combined to form a second target test question, and a course test is constructed based on the first target test question and the second target test question; the course test is sent to the learning object, and the learning object answers according to the course test, and test answer data is generated, and the test answer data is obtained, and the learning mastery score rate of the learning object is calculated based on the test answer data using the test score and difficulty coefficient of each test question in the course test, the test score and difficulty coefficient of each test question are calculated, and then the calculation results of all the test questions are comprehensively counted to obtain the learning mastery score rate.
[0087] The consolidation learning path generation module 24 is used to continue the study of the ceramic art course grid if the learning mastery score rate reaches a preset threshold value, and if the learning mastery score rate does not reach the preset threshold value, obtain a number of weak knowledge points based on the learning mastery score rate using a knowledge base with a tree structure, and generate a consolidation learning path based on the weak knowledge points, and send the consolidation learning path to the corresponding learning object;
[0088] In the specific implementation process of the present invention, the method of obtaining a number of weak knowledge points based on the learning mastery score rate using a knowledge base with a tree structure includes: calculating association weights for relevant knowledge points in each pottery course grid, constructing a knowledge tree based on the association weights, and constructing a knowledge base with a tree structure based on the knowledge tree; constructing a mastery matrix of relevant knowledge points based on the learning mastery score rate; and obtaining a number of weak knowledge points of a learning object based on the mastery matrix using a knowledge base with a tree structure.
[0089] Furthermore, the generation of a consolidation learning path based on the weak knowledge points includes: counting the proportion of each weak knowledge point; obtaining corresponding preceding knowledge points and succeeding knowledge points for the weak knowledge points using a knowledge base with a tree structure, and constructing a learning path relationship between the weak knowledge points based on the preceding knowledge points and succeeding knowledge points; and constructing a consolidation learning path based on the learning path relationship and the proportion of each weak knowledge point.
[0090] Specifically, the learning mastery score is compared with a preset threshold value. If the learning mastery score rate reaches the preset threshold value, the learning of the ceramic course grid is continued. If the learning mastery score rate does not reach the preset threshold value, the relevant knowledge points in each ceramic course grid are traversed in turn, and the frequency of occurrence of two adjacent relevant knowledge points in the corresponding ceramic course grid is calculated. The frequency of occurrence is used as an initial weight value, and then the initial weight value is adjusted by a preset ratio to obtain an associated weight. The relevant knowledge points are added to the knowledge tree in turn according to the associated relationship, starting from the root node of the knowledge tree. The first related knowledge point added is connected to the root node, and its connection strength is set to zero. The associated weight is used as the connection strength between the remaining related knowledge points. After all knowledge points are added to the corresponding knowledge tree, a knowledge base with a tree structure is established according to the constructed knowledge tree; a mastery matrix of the relevant knowledge points is constructed based on the learning mastery score rate, and the learning ability of the corresponding knowledge point can be obtained according to the learning mastery score rate. The relevant knowledge points are constructed based on the learning ability of the corresponding knowledge point combined with the complexity of the knowledge point. A mastery matrix; based on the mastery matrix, a knowledge base with a tree structure is used to obtain several weak knowledge points of the learning object. Based on the mastery matrix, the overall mastery rating of the learning object can be determined. The knowledge points designed in this course test are sorted according to the overall mastery rating, and the knowledge points with the last digit in the sorting are used as the initial weak points. The knowledge base with a tree structure is used to obtain the association relationship between the initial weak points, and several knowledge weak points are finally determined through the association relationship to prevent omissions in the search for knowledge weak points due to accidental factors; the proportion of each weak knowledge point is counted, and the proportion of each weak knowledge point in the corresponding ceramic course grid is counted, that is, the number of occurrences of the corresponding weak knowledge point in the corresponding ceramic course grid is counted; the knowledge base with a tree structure is used to obtain the corresponding pre-knowledge point and post-knowledge point for the weak knowledge point. The knowledge base with a tree structure stores the association relationship of each knowledge point, so the corresponding pre-knowledge point and post-knowledge point can be quickly found, and the learning path relationship between the weak knowledge points is constructed based on the pre-knowledge point and post-knowledge point;Based on the learning path relationship and the proportion of each weak knowledge point, a consolidation learning path is constructed. Each weak knowledge point is connected in series through the learning path relationship. At the same time, the proportion of teaching resources and learning time for consolidation learning can be determined according to the proportion of weak knowledge points. The learning object re-learns according to the consolidation learning path. Since autistic children have certain obstacles in social interaction and communication, they usually lack communication with teachers about their own weak links in the course, resulting in insufficient learning of the pottery course. Therefore, a number of weak knowledge points are obtained by using a knowledge base with a tree structure through learning mastery score rate, and a consolidation learning path is generated based on the weak knowledge points, which can effectively avoid the problem of insufficient learning. In addition, the learning detection of each pottery course grid can not only improve the learning quality, but also scientifically and objectively reflect the psychological and learning changes of autistic children. ;
[0091] Learning evaluation and course improvement module 25: is used to evaluate the learning mastery score rate of all pottery course grids when it is detected that the learning progress of all pottery course grids has reached a preset total value, determine the course improvement plan based on the evaluation result combined with the learning state recognition result of the learning object and the consolidation learning path, and visualize the course improvement plan.
[0092] In the specific implementation process of the present invention, the learning mastery score rates of all pottery course grids are evaluated, and a course improvement plan is determined based on the evaluation results combined with the learning state recognition results and consolidation learning paths of the learning objects, including: calculating the learning mastery score rate of each pottery course grid according to a preset weight, and evaluating the learning effect of the learning object based on the calculation results using preset hierarchical categories to obtain evaluation results; obtaining the learning object's knowledge points of interest based on the evaluation results combined with the learning state recognition results and learning interaction data of the learning object, and determining the course improvement plan based on the knowledge points of interest and the consolidation learning paths.
[0093] Specifically, when it is detected that the learning progress of all ceramics course grids has reached a preset total value, the learning mastery score rate of each ceramics course grid is calculated according to the preset weight, the learning mastery score rate of each ceramics course grid is multiplied by the preset weight, the multiplication results are accumulated to obtain the calculation result, and based on the calculation result, the learning effect of the learning object is evaluated using the preset hierarchical category to obtain the evaluation result; based on the evaluation result, the learning state recognition result and the learning interaction data of the learning object are combined to obtain the learning object's interested knowledge points, and the course improvement plan is determined based on the interested knowledge points and the consolidation learning path, and the learning state recognition result and the learning interaction data of the learning object are combined to obtain the learning object's interested knowledge points, and the course improvement plan is determined based on the interested knowledge points and the consolidation learning path. The data can be used to obtain the interested knowledge points of the learning objects. The learning status identification results of the learning objects can tell which knowledge points the learning objects are more likely to be distracted or produce learning fatigue. The learning interaction data can tell which knowledge points the learning objects have more operation behaviors in the resource data. The initial curriculum plan can be improved through the interested knowledge points and the consolidation learning path to make the curriculum improvement plan more in line with the interests of autistic children. The curriculum improvement plan can be visualized in the form of charts to increase the learning interest of autistic children. This not only ensures the effectiveness of the implementation of pottery course education for autistic children, but also helps promote the transformation of autistic children's pottery activities into realistic psychological functions and social functions.
[0094] In the embodiment of the present invention, the pottery course is divided into several pottery course grids through the task system of the pottery course, so that the pottery course can be studied from shallow to deep according to the subject classification and difficulty level, so that autistic children can better understand the course content and improve their learning interest. At the same time, the learning state is identified through the data fusion algorithm, so as to further improve the learning concentration. The learning behavior of the autistic children can be known through the learning state identification, and then it can be judged which knowledge point they are more focused on and interested in. After the study of the pottery course at each stage, the learning mastery score rate is used to obtain several weak knowledge points using a knowledge base with a tree structure, and a consolidation learning path is generated based on the weak knowledge points, which improves the learning quality and can scientifically and objectively reflect the psychological and learning change process of autistic children, which not only ensures the effect of the implementation of the pottery course education for autistic children, but also is beneficial to promote the transformation of autistic children's pottery activities into realistic psychological functions and social functions.
[0095] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, an interactive learning method for a pottery course in any of the above embodiments is implemented. The computer-readable storage medium includes, but is not limited to, any type of disk (including floppy disk, hard disk, optical disk, CD-ROM, and magneto-optical disk), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic card or optical card. That is, the storage device includes any medium that can store or transmit information in a readable form by a device (for example, a computer, a mobile phone), which can be a read-only memory, a disk or an optical disk, etc.
[0096] Embodiment 3
[0097] See also Figure 3 , Figure 3 It is a schematic diagram of the structure of an electronic device in an embodiment of the present invention.
[0098] The embodiment of the present invention further provides an electronic device, such as Figure 3 As shown, the electronic device includes a memory 31, a processor 33, and a computer program 32 stored in the memory 31 and executable on the processor 33. Those skilled in the art will appreciate that Figure 3The electronic device shown does not constitute a limitation on all devices, and may include more or fewer components than shown, or combine certain components. The memory 31 can be used to store the computer program 32 and various functional modules, and the processor 33 runs the computer program 32 stored in the memory 31, thereby executing various functional applications and data processing of the device. The memory can be an internal memory or an external memory, or include both internal and external memories. The internal memory can include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a flash memory, or a random access memory. The external memory can include a hard disk, a floppy disk, a ZIP disk, a U disk, a magnetic tape, etc. The processor 33 can 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 gates or transistor logic devices, discrete hardware components, etc. The general processor may be a microprocessor, a single chip microcomputer or the processor 33 may also be any conventional processor, etc. The processor and memory disclosed in the present invention include but are not limited to these types of processors and memories. The processor and memory disclosed in the present invention are only examples and not limitations.
[0099] As an embodiment, the electronic device includes: one or more processors 33, a memory 31, and one or more computer programs 32, wherein the one or more computer programs 32 are stored in the memory 31 and are configured to be executed by the one or more processors 33, and the one or more computer programs 32 are configured to execute the interactive learning method for pottery courses in any of the above embodiments. For the specific implementation process, please refer to the above embodiments and will not be repeated here.
[0100] In the embodiment of the present invention, the pottery course is divided into several pottery course grids through the task system of the pottery course, so that the pottery course can be studied from shallow to deep according to the subject classification and difficulty level, so that autistic children can better understand the course content and improve their learning interest. At the same time, the learning state is identified through the data fusion algorithm, so as to further improve the learning concentration. The learning behavior of the autistic children can be known through the learning state identification, and then it can be judged which knowledge point they are more focused on and interested in. After the study of the pottery course at each stage, the learning mastery score rate is used to obtain several weak knowledge points using a knowledge base with a tree structure, and a consolidation learning path is generated based on the weak knowledge points, which improves the learning quality and can scientifically and objectively reflect the psychological and learning change process of autistic children, which not only ensures the effect of the implementation of the pottery course education for autistic children, but also is beneficial to promote the transformation of autistic children's pottery activities into realistic psychological functions and social functions.
[0101] In addition, the above is a detailed introduction to an interactive learning method and related devices for a pottery course provided by an embodiment of the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.
Claims
1. An interactive learning method for a pottery course, characterized in that: The method comprises: Obtain a task system for a pottery course, divide the pottery course into grids based on the task system for the pottery course, and obtain a number of pottery course grids; Continuously collecting real-time images of the learning object in the process of learning the ceramic art course grid based on the image acquisition device, performing learning state recognition processing based on the real-time image using a data fusion algorithm, and sending a reminder message to the learning object if it is recognized that the learning object is in an abnormal learning state; Detecting the learning progress of each ceramics course grid, and when it is detected that the learning progress of the corresponding ceramics course grid reaches a preset standard value, generating a course test by combining the learning interaction data between the learning object and the corresponding ceramics course grid, and calculating the learning mastery score rate of the learning object based on the test answer data generated by the learning object according to the course test; If the learning mastery score rate reaches the preset threshold, the learning of the ceramic art course grid is continued; if the learning mastery score rate does not reach the preset threshold, a number of weak knowledge points are obtained based on the learning mastery score rate using a knowledge base with a tree structure, and a consolidation learning path is generated based on the weak knowledge points, and the consolidation learning path is sent to the corresponding learning object; When it is detected that the learning progress of all ceramic course grids has reached a preset total value, the learning mastery score rate of all ceramic course grids is evaluated, and a course improvement plan is determined based on the evaluation result combined with the learning state recognition result of the learning object and the consolidation learning path, and the course improvement plan is visualized; Among them, the task system of the ceramics course is obtained, and the ceramics course is grid-divided based on the task system of the ceramics course to obtain a number of ceramics course grids, including: obtaining the task system of the ceramics course and the total ceramics course resource data set based on a preset platform; performing data format conversion on the total ceramics course resource data set to obtain a standardized total ceramics course resource data set; performing data stream classification on the standardized total ceramics course resource data set to obtain a ceramics course teaching video data set and a ceramics course teaching text data set; performing word embedding encoding processing on the ceramics course teaching text data set to obtain a bidirectional autoregressive vocabulary, performing vocabulary expansion on the bidirectional autoregressive vocabulary to obtain an expanded bidirectional autoregressive vocabulary, pre-training the expanded bidirectional autoregressive vocabulary to obtain a vocabulary embedding matrix, and performing word embedding on the ceramics course teaching text data set through the vocabulary embedding matrix to obtain a text semantic feature data set; performing feature extraction processing on the ceramics course teaching video data set, extracting frames on the ceramics course teaching video data set to obtain a number of ceramics course teaching video frames, and for each ceramic The ceramic art course teaching video frames are subjected to multi-level pooling processing, and the corresponding feature matrix is determined based on each ceramic art course teaching video frame, and the corresponding feature matrix is subjected to multi-level pooling processing, that is, the corresponding feature matrix is gradually reduced until it is reduced to a feature matrix containing only one point, and a plurality of ceramic art course teaching video frames after multi-level pooling processing are obtained, and a video feature data set is obtained by splicing the ceramic art course teaching video frames after multi-level pooling processing; the text semantic feature data set and the video feature data set are subjected to the same theme classification integration processing, and a likelihood function is constructed, and the text semantic feature data set and the video feature data set are input into the likelihood function to obtain a theme feature vector, and the similarity of each two feature data sets is calculated according to the theme feature vector, and two data sets whose similarity reaches a preset threshold value are classified into the same theme classification, and a classified ceramic art course resource data set is obtained, and the classified ceramic art course resource data set is combined with the task system of the ceramic art course to perform course grid division according to a preset difficulty level, and a plurality of ceramic art course grids are obtained; The method of obtaining a plurality of weak knowledge points based on the learning and mastering score rate by using a knowledge base with a tree structure includes: calculating an association weight for the relevant knowledge points in each ceramic course grid, constructing a knowledge tree based on the association weight, adding the relevant knowledge points to the knowledge tree in sequence according to the association relationship, and adding them in sequence starting from the root node of the knowledge tree. The first relevant knowledge point added is connected to the root node, and its connection strength is set to zero. The association weight is used as the connection strength between the remaining relevant knowledge points, and a knowledge base with a tree structure is constructed based on the knowledge tree; the learning ability of the corresponding knowledge point is obtained according to the learning and mastering score rate, and a mastery matrix of the relevant knowledge point is constructed based on the learning ability of the corresponding knowledge point combined with the complexity of the knowledge point; and the weak knowledge points of the learning object are obtained based on the mastery matrix by using a knowledge base with a tree structure; Generating a consolidation learning path based on the weak knowledge points includes: counting the proportion of each weak knowledge point in the corresponding ceramics course grid, that is, counting the number of occurrences of the corresponding weak knowledge point in the corresponding ceramics course grid; obtaining corresponding pre-knowledge points and post-knowledge points for the weak knowledge points using a knowledge base with a tree structure, and constructing a learning path relationship between the weak knowledge points based on the pre-knowledge points and post-knowledge points; and constructing a consolidation learning path based on the learning path relationship and the proportion of each weak knowledge point; The method of using a data fusion algorithm to perform learning state recognition processing based on the real-time image includes: performing face detection processing on the learning object based on the real-time image, if the face of the learning object is not detected, judging the learning state to be an absent learning state, if the face of the learning object is detected, establishing a two-dimensional face shape model and a local texture model, determining posture parameters and shape parameters based on the two-dimensional face shape model and the local texture model, accurately locating the face area image according to the posture parameters and the shape parameters, and obtaining the face area of the learning object; locating facial feature points and eye feature points of the face area based on a neural domain feature point positioning algorithm, performing similarity transformation on the face area image, that is, rotating, scaling and translating the face area image, and obtaining the face area of the learning object through neural domain feature point positioning algorithm. The method comprises the following steps: detecting feature points in the face area image after similarity transformation by point detection, respectively setting facial feature constraint conditions and eye feature constraint conditions, framing the feature points of the facial image according to the facial feature constraint conditions to obtain facial feature points, and framing the eye feature points according to the eye feature constraint conditions to obtain eye feature points; calculating fatigue state parameters and distraction state parameters based on the facial feature points and eye feature points by using a data fusion algorithm in combination with a proportional coefficient and a correction parameter; comparing the fatigue state parameter with a first preset threshold value, and if the fatigue state parameter is greater than or equal to the first preset threshold value, determining that the learning state is a fatigue state; and comparing the distraction state parameter with a second preset threshold value, and if the distraction state parameter is greater than or equal to the second preset threshold value, determining that the learning state is a distraction state.
2. The interactive learning method for pottery courses according to claim 1, characterized in that: The step of generating a course test by combining the learning interaction data of the learning object and the corresponding ceramic course grid, and calculating the learning mastery score rate of the learning object based on the test answer data generated by the learning object according to the course test, comprises: Constructing a correspondence between the learning interaction data and the course test type, and generating a first target test question based on the correspondence; Generate a second target test question based on the knowledge points of the corresponding ceramics course grid, and construct a course test based on the first target test question and the second target test question; The test answer data generated by the learning object according to the course test is obtained, and the learning mastery score rate of the learning object is calculated based on the test answer data using the test score and difficulty coefficient of each test question in the course test.
3. The interactive learning method for pottery courses according to claim 1, characterized in that: The learning mastery score rate of all ceramic course grids is evaluated, and a course improvement plan is determined based on the evaluation results combined with the learning state identification results of the learning objects and the consolidation learning path, including: The learning mastery score rate of each ceramics course grid is calculated according to the preset weight, and the learning effect of the learning object is evaluated based on the calculation result using the preset hierarchical category to obtain the evaluation result; Based on the evaluation results combined with the learning state recognition results and learning interaction data of the learning object, the learning object's knowledge points of interest are obtained, and a course improvement plan is determined based on the knowledge points of interest and the consolidation learning path.
4. An interactive learning device for pottery courses, characterized in that: The device comprises: Course grid division module: used to obtain the task system of the pottery course, divide the pottery course into grids based on the task system of the pottery course, and obtain several pottery course grids; Learning state recognition module: used for continuously collecting real-time images of the learning object in the process of learning the ceramic course grid based on the image acquisition device, and performing learning state recognition processing based on the real-time image using the data fusion algorithm. If the learning object is identified to be in an abnormal learning state, a reminder message is issued to the learning object; Learning mastery score rate calculation module: used to detect the learning progress of each ceramic course grid. When it is detected that the learning progress of the corresponding ceramic course grid reaches a preset standard value, a course test is generated by combining the learning interaction data of the learning object and the corresponding ceramic course grid, and the learning mastery score rate of the learning object is calculated based on the test answer data generated by the learning object according to the course test; A consolidation learning path generation module: used for continuing the learning of the ceramic art course grid if the learning mastery score rate reaches a preset threshold; if the learning mastery score rate does not reach the preset threshold, obtaining a number of weak knowledge points based on the learning mastery score rate using a knowledge base with a tree structure, and generating a consolidation learning path based on the weak knowledge points, and sending the consolidation learning path to the corresponding learning object; Learning evaluation and course improvement module: used to evaluate the learning mastery score rate of all ceramic course grids when it is detected that the learning progress of all ceramic course grids has reached the preset total value, determine the course improvement plan based on the evaluation result combined with the learning state recognition result of the learning object and the consolidation learning path, and visualize the course improvement plan; Among them, the task system of the ceramics course is obtained, and the ceramics course is grid-divided based on the task system of the ceramics course to obtain a number of ceramics course grids, including: obtaining the task system of the ceramics course and the total ceramics course resource data set based on a preset platform; performing data format conversion on the total ceramics course resource data set to obtain a standardized total ceramics course resource data set; performing data stream classification on the standardized total ceramics course resource data set to obtain a ceramics course teaching video data set and a ceramics course teaching text data set; performing word embedding encoding processing on the ceramics course teaching text data set to obtain a bidirectional autoregressive vocabulary, performing vocabulary expansion on the bidirectional autoregressive vocabulary to obtain an expanded bidirectional autoregressive vocabulary, pre-training the expanded bidirectional autoregressive vocabulary to obtain a vocabulary embedding matrix, and performing word embedding on the ceramics course teaching text data set through the vocabulary embedding matrix to obtain a text semantic feature data set; performing feature extraction processing on the ceramics course teaching video data set, extracting frames on the ceramics course teaching video data set to obtain a number of ceramics course teaching video frames, and for each ceramic The ceramic art course teaching video frames are subjected to multi-level pooling processing, and the corresponding feature matrix is determined based on each ceramic art course teaching video frame, and the corresponding feature matrix is subjected to multi-level pooling processing, that is, the corresponding feature matrix is gradually reduced until it is reduced to a feature matrix containing only one point, and a plurality of ceramic art course teaching video frames after multi-level pooling processing are obtained, and a video feature data set is obtained by splicing the ceramic art course teaching video frames after multi-level pooling processing; the text semantic feature data set and the video feature data set are subjected to the same theme classification integration processing, and a likelihood function is constructed, and the text semantic feature data set and the video feature data set are input into the likelihood function to obtain a theme feature vector, and the similarity of each two feature data sets is calculated according to the theme feature vector, and two data sets whose similarity reaches a preset threshold value are classified into the same theme classification, and a classified ceramic art course resource data set is obtained, and the classified ceramic art course resource data set is combined with the task system of the ceramic art course to perform course grid division according to a preset difficulty level, and a plurality of ceramic art course grids are obtained; The method of obtaining a plurality of weak knowledge points based on the learning and mastering score rate by using a knowledge base with a tree structure includes: calculating an association weight for the relevant knowledge points in each ceramic course grid, constructing a knowledge tree based on the association weight, adding the relevant knowledge points to the knowledge tree in sequence according to the association relationship, and adding them in sequence starting from the root node of the knowledge tree. The first relevant knowledge point added is connected to the root node, and its connection strength is set to zero. The association weight is used as the connection strength between the remaining relevant knowledge points, and a knowledge base with a tree structure is constructed based on the knowledge tree; the learning ability of the corresponding knowledge point is obtained according to the learning and mastering score rate, and a mastery matrix of the relevant knowledge point is constructed based on the learning ability of the corresponding knowledge point combined with the complexity of the knowledge point; and the weak knowledge points of the learning object are obtained based on the mastery matrix by using a knowledge base with a tree structure; Generating a consolidation learning path based on the weak knowledge points includes: counting the proportion of each weak knowledge point in the corresponding ceramics course grid, that is, counting the number of occurrences of the corresponding weak knowledge point in the corresponding ceramics course grid; obtaining corresponding pre-knowledge points and post-knowledge points for the weak knowledge points using a knowledge base with a tree structure, and constructing a learning path relationship between the weak knowledge points based on the pre-knowledge points and post-knowledge points; and constructing a consolidation learning path based on the learning path relationship and the proportion of each weak knowledge point; The method of using a data fusion algorithm to perform learning state recognition processing based on the real-time image includes: performing face detection processing on the learning object based on the real-time image, if the face of the learning object is not detected, judging the learning state to be an absent learning state, if the face of the learning object is detected, establishing a two-dimensional face shape model and a local texture model, determining posture parameters and shape parameters based on the two-dimensional face shape model and the local texture model, accurately locating the face area image according to the posture parameters and the shape parameters, and obtaining the face area of the learning object; locating facial feature points and eye feature points of the face area based on a neural domain feature point positioning algorithm, performing similarity transformation on the face area image, that is, rotating, scaling and translating the face area image, and obtaining the face area of the learning object through neural domain feature point positioning algorithm. The method comprises the following steps: detecting feature points in the face area image after similarity transformation by point detection, respectively setting facial feature constraint conditions and eye feature constraint conditions, framing the feature points of the facial image according to the facial feature constraint conditions to obtain facial feature points, and framing the eye feature points according to the eye feature constraint conditions to obtain eye feature points; calculating fatigue state parameters and distraction state parameters based on the facial feature points and eye feature points by using a data fusion algorithm in combination with a proportional coefficient and a correction parameter; comparing the fatigue state parameter with a first preset threshold value, and if the fatigue state parameter is greater than or equal to the first preset threshold value, determining that the learning state is a fatigue state; and comparing the distraction state parameter with a second preset threshold value, and if the distraction state parameter is greater than or equal to the second preset threshold value, determining that the learning state is a distraction state.
5. An electronic device, comprising a processor and a memory, characterized in that: The memory is used to store instructions, and the processor is used to call the instructions in the memory, so that the electronic device executes the interactive learning method of pottery courses according to any one of claims 1 to 3.
6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed on an electronic device, the electronic device executes the interactive learning method of a ceramics course according to any one of claims 1 to 3.
