A data-based intelligent education training management system

Abstract

The application discloses a data-based intelligent education training management system for guiding training subjects or trainees to make up for knowledge defects existing in a training process, comprising: a collection module for collecting multi-modal information of the trainees, a creation module in communication connection with the collection module and used for establishing multi-modal learning models from the multi-modal information, an analysis module in communication connection with the creation module and used for acquiring the multi-modal models, a detection module in communication connection with the analysis module, and a decision module in communication connection with the analysis module. The system has intelligent data information collection capability, data information analysis capability, diagnosis capability, feedback capability and improvement effect evaluation capability after being built, has the characteristics of full sample, multi-dimensional and multi-modal information, reality, accompanying, individualization, instant feedback and the like, efficiently serves the trainees to efficiently learn, reduces the burden of teachers and students, and helps the trainees to solve problems existing in a learning process.

Description

Technical Field

[0001] The present invention relates to the field of intelligent education technology, and in particular to a data-based intelligent education and training management system. Background Art

[0002] Education is a particularly complicated matter, because students will forget, students will grow, and their thinking will change. Teachers need to constantly test and verify the mastery of knowledge during the learning process. From a long-term perspective, the process and form of students' learning are secondary, and developing good learning habits is the most important thing.

[0003] With the continuous exploration and development of information technology, the education profession has also quickly integrated into the wave. Many people advocate the promotion of new information technologies represented by big data, cloud computing, artificial intelligence and the Internet in conjunction with education. At present, the education model that combines big data, cloud computing, artificial intelligence and the Internet is fixed on the education and teaching model. Specifically, big data is used to track the rules of students' learning content or learning needs, and to push suitable or close subject content to students to help students learn more deeply and more knowledge content that is completely consistent with or related to the subject or needs. However, in terms of analyzing the effect of students' learning knowledge, especially the students' mastery of knowledge, it is impossible to conduct comprehensive testing, analysis and summary, and it is impossible to provide students with comprehensive and more optimized recommended suggestions for each subject. As a result, students do not have an objective and accurate understanding of the effect of their learning in the subject, and cannot fully understand their own learning effects, so that students' grades cannot be comprehensively improved.

[0004] Therefore, the current problem is that the systems involved in current information technology have gaps in the detection, analysis and summary of the effectiveness of students' mastery of subject knowledge, and cannot provide students with optimized recommended suggestions suitable for improving their grades. Summary of the Invention

[0005] In order to solve the technical problem that the systems involved in the above-mentioned current information technology have gaps in the detection, analysis and summary of the students' mastery of subject knowledge, and cannot provide students with optimized recommended suggestions suitable for improving their grades, the present invention provides a data-based intelligent education and training management system, which collects, models and analyzes the multimodal learning status of the trainees during the training process, summarizes the deficient knowledge points in the subjects, and gives recommended suggestions for improvement.

[0006] In order to achieve the above object, the present invention is achieved through the following technical solutions:

[0007] A data-based intelligent education and training management system is used to guide trainees or trainees to make up for knowledge deficiencies in the training process, including:

[0008] The acquisition module is configured to acquire multi-modal information of the trainee, the multi-modal information being gestures, expressions and sounds of the trainee in the training process, so as to grasp the interaction degree of the trainee.

[0009] The creation module is in communication connection with the acquisition module and is configured to establish a multi-modal learning model for the multi-modal information.

[0010] The analysis module is in communication connection with the creation module and is configured to acquire the multi-modal model, calculate and analyze the probability of each modal according to a preset algorithm, and obtain modal probability information.

[0011] The detection module is in communication connection with the analysis module and is configured to acquire detection information of the trainee and transmit the detection information to the analysis module, the analysis module being configured to compare the detection information with target information to obtain defect information, and the creation module being configured to establish a defect information model according to the defect information to obtain a defect model, wherein the target information is knowledge information required for the trainee to achieve a training target in each subject in the analysis module, and the defect information is knowledge information different between the target information and the detection information.

[0012] The decision module is in communication connection with the analysis module and is configured to combine the modal probability information with the defect information, associate each modal probability with a defect model of a subject corresponding to the modal probability, so that the trainee understands the defect points in the training, and the decision module gives the trainee a recommendatory suggestion for making up the knowledge points of the subject or guides the training subject on how to train the trainee to master the defect information.

[0013] Compared with the prior art, the present application has the following advantages: the system composed of the acquisition module, the creation module, the analysis module, the detection module and the decision module can completely capture the learning situation of the trainee in the training process, the multi-modal model established by the creation module is analyzed and summarized by the analysis module, the decision module gives the trainee a recommendatory suggestion for making up the knowledge points of the subject, and also gives the training subject a decision method for helping the trainee to improve, the system has intelligent data information acquisition capability, data information analysis capability, diagnosis capability, feedback capability and improvement effect evaluation capability after being built, has the characteristics of full sample, multi-dimensional and multi-modal information, reality, accompanying, individualization, instant feedback, etc., efficiently serves the trainee to efficiently learn, and reduces the burden of teachers and students. Through big data and various analysis technologies, the knowledge mastery of the trainee, the difference in the ability level, the behavior characteristics and the character features of the trainee can be effectively diagnosed and analyzed and fed back, the teacher is helped to better teach students according to their aptitude, and the trainee is helped to solve the problems in the learning process.

[0014] Further preferably, the acquisition module comprises:

[0015] The gesture acquisition device is in communication connection with the creating module and is used to acquire gesture information of the training subject in the training process.

[0016] The sound acquisition device is in communication connection with the creating module and is used to acquire sound information communicated between the training subject and the training subject.

[0017] The expression acquisition device is in communication connection with the creating module and is used to acquire facial expressions of the training subject in the training process, so as to acquire the training state of the training subject in the training process.

[0018] The above technical solution is adopted to acquire the gesture, sound and expression information of the training subject in the learning process.

[0019] Further preferably, the gesture acquisition device selects a gesture recognition device.

[0020] The above technical solution can accurately identify the gesture of the training subject and provide an accurate basis for acquiring the interactive information of the training subject in the learning process.

[0021] Further preferably, the sound acquisition device selects a distance education sound acquisition device.

[0022] The above technical solution can acquire the sound emitted by the training subject when interacting with the training subject in the learning process, and provides an accurate basis for acquiring the information corresponding to the sound mode.

[0023] Further preferably, the expression acquisition device selects a facial expression capture device.

[0024] The above technical solution can acquire the expression information of the training subject when interacting with the training subject in the learning process, and provides an accurate basis for acquiring the information corresponding to the expression mode.

[0025] Further preferably, the creating module comprises:

[0026] The defect modeling submodule is in communication connection with the detecting module and is used to edit defect information through a matrix to create a defect model.

[0027] The multi-modal modeling submodule is in communication connection with the acquisition module and is used to edit multi-modal information into different modal matrices, create a multi-modal model through a nonlinear dynamic system, and obtain a gesture model, a sound model and an expression model respectively.

[0028] The above technical solution can complete the creation of the defect model and the modal model, and further realize the recording of the learning state of the simulated training subject in the learning stage.

[0029] Further preferably, the analysis module comprises:

[0030] The subject evaluation submodule is in communication connection with the detection module, configured to acquire the detection information, and compare the detection information with the target information to obtain the defect information.

[0031] The gesture analysis submodule is in communication connection with the multi-modal modeling submodule, configured to analyze the probability of the appearance of the gesture model to obtain the gesture probability information.

[0032] The sound analysis submodule is in communication connection with the multi-modal modeling submodule, configured to analyze the probability of the appearance of the sound to obtain the sound probability information.

[0033] The expression analysis submodule is in communication connection with the multi-modal modeling submodule, configured to analyze the probability of the appearance of the expression in the interactive state during the training to obtain the expression probability information.

[0034] By obtaining the defect information, the gesture probability information, the sound probability information and the expression probability information, the analysis result can be directly obtained, which provides a direct analysis basis for the training subject and a most direct method basis for the training subject to know the training subject.

[0035] Further optimization is that the detection module comprises:

[0036] The class detection submodule is in communication connection with the subject evaluation submodule, configured to collect the condition of the training subject mastering the subject knowledge during the class in the training process to obtain the class detection information, and transmit the class detection information to the subject evaluation submodule.

[0037] The stage detection submodule is in communication connection with the subject evaluation submodule, configured to collect the condition of the training subject mastering the subject knowledge in the current time after the stage training is completed to obtain the stage detection information, and transmit the stage detection information to the subject evaluation submodule.

[0038] The subject evaluation submodule is configured to compare the class detection information and the stage detection information with the target information to obtain the class defect information and the stage defect information respectively.

[0039] By the above technical solution, the detection of the training subject in the class learning effect and the stage learning is completed, and the direct analysis and summary of the knowledge points not mastered or not mastered comprehensively in the class learning are completed, and the same is true for the comparison and presentation of the missed knowledge points or the not solid knowledge points after the stage learning is completed.

[0040] Further optimization is that the decision module is in communication connection with the subject evaluation submodule, the gesture analysis submodule, the sound analysis submodule and the expression analysis submodule, configured to provide the class defect information, the stage defect information, the gesture probability information, the sound probability information and the expression probability information to the training subject or the training subject, and give the optimization suggestion according to the class defect information and the stage defect information.

[0041] With the above technical solutions, the defects existing in the learning process of the training subject are directly pointed out by the in-class detection and stage detection, and the most direct solution to the problems existing in the learning process is given to help the training subject complete the learning efficiently, and the most direct recommended suggestions to help the training subject improve are recommended for the training subject, thereby reducing the training work burden of the training subject. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 It is a functional module diagram of the embodiment.

[0043] Figure 2 It is a functional module diagram of the acquisition module in the embodiment.

[0044] Figure 3 It is a functional module diagram of the creation module in the embodiment.

[0045] Figure 4 It is a functional module diagram of the analysis module in the embodiment.

[0046] Figure 5 It is a functional module diagram of the detection module in the embodiment.

[0047] Figure 6 It is a system information transmission diagram in the embodiment.

[0048] Reference signs: 1-training subject; 2-training subject; 3-acquisition module; 31-gesture acquisition device; 32-sound acquisition device; 33-expression acquisition device; 4-creation module; 41-defect modeling sub-module; 42-multimodal modeling sub-module; 5-analysis module; 51-discipline evaluation sub-module; 52-gesture analysis sub-module; 53-sound analysis sub-module; 54-expression analysis sub-module; 6-detection module; in-class detection sub-module; 62-stage detection sub-module; 7-decision module. DETAILED DESCRIPTION

[0049] With the continuous exploration and development of information technology, the education profession is quickly integrated into the tide. Many people advocate the implementation of new information technologies such as big data, cloud computing, artificial intelligence, and the Internet to cooperate with education. At present, the education mode combining big data, cloud computing, artificial intelligence, and the Internet is fixed on the mode of education and teaching. Specifically, the student is pushed to the subject content suitable or close to the student by tracking the learning content rules or learning needs of the student through big data, so as to help the student learn deeper and more knowledge content that completely matches or is related to the subject or needs. However, in the analysis of the learning effect of the student, especially the situation that the student masters the knowledge, the student cannot be comprehensively detected, analyzed and summarized, and the student cannot be provided with comprehensive and more optimized recommendation suggestions for each subject, so that the student cannot objectively and accurately recognize the effect of learning the subject, and the student cannot fully understand the learning effect, so that the student's performance cannot be fully improved.

[0050] Therefore, the current problem is that information technology is lacking in the detection, analysis and summary of the effect of students mastering subject knowledge, and cannot provide students with optimized recommendation suggestions suitable for improving students' performance. The following is a further detailed description of the present application in combination with the accompanying Figure 1 、 Figure 2 、 Figure 3 、 Figure 4 、 Figure 5 and Figure 6 .

[0051] In view of the above technical problems, the present application is designed and conceived as follows: in order to comprehensively master the learning status of the student, i.e. the training subject, it is conceived that the multi-modal information of the classroom teaching is obtained from the language expression interaction, body movement and learning expression of the training subject, and the in-class detection information reflecting the training effect is obtained from the in-class detection sub-module of each subject. The stage detection information is obtained during the training process. A multi-modal learning model is created by creating a module. The occurrence frequency of the three modes is analyzed and summarized by a multi-modal analysis module. The in-class detection information is compared with the target information of the preset subject to obtain the defect information. Finally, the decision module gives the training subject or the training subject a recommendation suggestion for finding and filling the gaps.

[0052] A data-based intelligent education and training management system for guiding the training subject 2 or the training subject 1 to make up for the knowledge defects existing in the training process, as shown in Figure 1 , comprising:

[0053] The acquisition module 3 is used for acquiring the multi-modal information of the training subject 1, and the multi-modal information is the gesture, expression and sound of the training subject 1 during the training process, so as to master the interaction degree of the training subject 1.

[0054] The creating module 4 is in communication connection with the collecting module 3, and is configured to establish a learning model of the multi-modal information in multiple modes. In an embodiment, the creating module 4 respectively statistically aggregates the multi-modal information in a learning process, such as gesture, expression and sound, and generates a learning model corresponding to each mode of information. For example, the creating module 4 aggregates the gesture information of a subject in a certain learning process, and generates a learning model of the gesture mode. Similarly, a learning model of the expression and sound is generated. The creating module 4 can generate the above learning model of mathematical linearity by using simulation software. The simulation software can be selected from MATLAB (MATLAB is a commercial mathematical software produced by MathWorks Company in the United States, which is used in the fields of data analysis, wireless communication, deep learning, image processing and computer vision, signal processing, quantitative finance and risk management, robot, control system, etc.).

[0055] The analyzing module 5 is in communication connection with the creating module 4, and is configured to obtain the multi-modal model, calculate and analyze the probability of each mode according to a preset algorithm, and obtain modal probability information.

[0056] The detecting module 6 is in communication connection with the analyzing module 5, and is configured to collect detection information of the subject 1, and transmit the detection information to the analyzing module 5. The analyzing module 5 is configured to compare the detection information with target information, and obtain defect information. The creating module 4 is configured to establish a defect information model according to the defect information, and obtain a defect model. The target information is knowledge information required for the subject 1 to achieve a training target in each subject in the analyzing module 5. The defect information is knowledge information different from the target information. The target information includes class target information and stage target information.

[0057] The decision module 7 is in communication connection with the analyzing module 5, and is configured to combine the modal probability information and the defect information, and associate each mode probability with the defect model of the corresponding subject, so that the subject 1 can understand the defect point of the subject 1, and then give the subject 1 an optimization suggestion, or guide the training subject 2 to train the subject 1 to master the defect information.

[0058] Please refer to Figures 1-6The system composed of the acquisition module 3, the creation module 4, the analysis module 5, the detection module 6 and the decision module 7 can completely capture the learning situation of the trainee 1 in the training process, and the analysis module 5 analyzes and summarizes the multi-modal model established by the creation module 4, and the decision module 7 gives the trainee 1 a recommended suggestion to make up for the knowledge points missing in the subject, and also gives the decision method of how the training subject 2 helps the trainee 1 to improve. After the system is built, it will have intelligent data information acquisition ability, data information analysis ability, diagnosis ability, feedback ability, improvement effect evaluation ability, full sample, information multi-dimensional multi-modal, real, accompanying, personalized, feedback instant and other characteristics, efficient service for the trainee 1 to learn efficiently, and reduce the burden of teachers and students. Through big data and various analysis techniques, the knowledge mastery of students, the difference in ability level, and the behavior characteristics and personality characteristics of students can be effectively diagnosed and analyzed and fed back, helping teachers to teach students according to their aptitude and helping students to solve problems in the learning process.

[0059] Specifically, as shown in Figure 2 The acquisition module 3 in the embodiment includes:

[0060] The gesture acquisition device 31 is in communication connection with the creation module 4, and is used to acquire the gesture information of the trainee 1 in the training process.

[0061] The sound acquisition device 32 is in communication connection with the creation module 4, and is used to acquire the sound information communicated between the trainee 1 and the training subject 2.

[0062] The expression acquisition device 33 is in communication connection with the creation module 4, and is used to acquire the facial expression of the trainee 1 in the training process, so as to acquire the training state of the trainee 1 in the training process. In this way, the gesture, sound and expression information of the trainee 1 in the learning process are acquired.

[0063] Specifically, the gesture acquisition device 31 in the embodiment selects a gesture recognition device. Specifically, it is a published technology of Chinese patent with application number CN202111665973.3 and name "a gesture recognition method, device, electronic device and storage medium", which can accurately recognize the gestures of the trainee and provide accurate basis for acquiring the interaction information of the trainee in the learning process.

[0064] Specifically, the sound collecting device 32 in the embodiment is a remote education sound collecting device 32. In an embodiment, the sound collecting device 32 applies the disclosed technology of the Chinese patent with the application number CN201921617653.9 and the name “a remote education sound collecting device 32 for a smart campus”. In this way, the sound emitted by the trainee in the interaction with the training subject 2 in the learning process can be obtained, thereby providing an accurate basis for collecting information corresponding to the sound mode.

[0065] Specifically, the expression collecting device 33 in the embodiment is a facial expression capture device. In an embodiment, the expression collecting device 33 applies the disclosed technology of the Chinese patent with the application number CN201921288505.7 and the name “a VR device with facial expression capture function”. In this way, the expression information of the trainee in the interaction with the training subject 2 in the learning process can be obtained, thereby providing an accurate basis for collecting information corresponding to the expression mode.

[0066] Specifically, as shown in Figure 3 the creating module 4 in the embodiment includes:

[0067] The defect modeling submodule 41 is in communication connection with the detecting module 6, and is configured to create a defect model by editing defect information through a matrix. The defect modeling submodule 41 realizes the creation of the defect model by running the matrix operation function in the simulation software.

[0068] The multi-modal modeling submodule 42 is in communication connection with the collecting module 3, and is configured to edit multi-modal information into different modal matrices, create a multi-modal model through a nonlinear dynamic system, and obtain a gesture model, a sound model, and an expression model, respectively. In this way, the creation of the defect model and the modal model can be completed, and the recording of the learning state of the simulated trainee 1 in the learning stage can be further realized. The multi-modal modeling submodule 42 realizes the creation of the defect model by running the matrix operation, drawing function, and data function in the simulation software.

[0069] Specifically, as shown in Figure 4 the analyzing module 5 in the embodiment includes:

[0070] The subject evaluation submodule 51 is in communication connection with the detecting module 6, and is configured to obtain detection information, compare the detection information with target information, and obtain defect information.

[0071] The gesture analysis submodule 52 is in communication connection with the multi-modal modeling submodule 42, and is configured to analyze the probability of the appearance of the gesture model, and obtain gesture probability information.

[0072] The sound analysis submodule 53 is in communication connection with the multi-modal modeling submodule 42, and is configured to analyze the probability of the appearance of the sound, and obtain sound probability information.

[0073] The expression analysis submodule 54 is in communication connection with the multi-modal modeling submodule 42, and is configured to analyze the probability of the appearance of the expression in the interaction state during the training, to obtain expression probability information.

[0074] By obtaining the defect information, the gesture probability information, the sound probability information, and the expression probability information, the analysis result can be directly obtained, which provides a direct analysis basis for the trainee 1 and a most direct method basis for the trainer 2 to know the trainee 1.

[0075] Specifically, as shown in FIG. 6, the detection module 6 in the embodiment includes: Figure 5

[0076] The in-class detection submodule 61 is in communication connection with the subject evaluation submodule 51, and is configured to collect the situation of the trainee 1 mastering the subject knowledge in the training process, to obtain in-class detection information, and to transmit the in-class detection information to the subject evaluation submodule 51.

[0077] The stage detection submodule 62 is in communication connection with the subject evaluation submodule 51, and is configured to collect the situation of the trainee 1 mastering the subject knowledge in the current time after the stage training ends, to obtain stage detection information, and to transmit the stage detection information to the subject evaluation submodule 51.

[0078] The subject evaluation submodule 51 is configured to compare the in-class detection information and the stage detection information with target information, to obtain in-class defect information and stage defect information, respectively.

[0079] In this way, the detection of the learning effect of the trainee 1 in the in-class learning and the detection of the learning effect of the trainee 1 in the stage learning are completed, and the knowledge points that are not comprehensively mastered or not mastered in the in-class learning are directly analyzed and summarized. Similarly, the knowledge points that are not comprehensively mastered or not mastered after the stage learning ends are compared and presented.

[0080] Specifically, the decision module 7 in the embodiment is in communication connection with the subject evaluation submodule 51, the gesture analysis submodule 52, the sound analysis submodule 53, and the expression analysis submodule 54, respectively, and is configured to provide the in-class defect information, the stage defect information, the gesture probability information, the sound probability information, and the sound probability information to the trainee 1 or the trainer 2, and to give a promotion suggestion for the in-class defect information and the stage defect information. In this way, the defects of the trainee 1 in the learning process are directly pointed out by the in-class detection and the stage detection, and the most direct solution to the problems existing in the learning process is given, to help the trainee 1 complete the learning efficiently. At the same time, the most direct recommendation suggestion for helping the trainee 1 to be promoted is recommended to the trainer 2, to reduce the training work burden of the trainer 2.

[0081] ​A matrix submodule is in communication connection with the gesture collection device 31 and the sound collection device 32 respectively, used for obtaining the gesture information and the sound information corresponding to the expression, and encoding the gesture information and the sound information into a matrix to calculate a gesture matrix and a sound matrix respectively.

[0082] An emulation submodule is in data connection with the matrix submodule, used for emulating the gesture virtual model and the sound virtual model corresponding to the expression through an output mathematical model.

[0083] Taking a subject X learned by a trained subject 1 as an example, please combine Figures 1-6 The whole management process of the present application will be described in detail as follows:

[0084] The training subject 2 teaches the trainee subject 1, which can be in the form of audio-visual or artificial teaching. During the teaching process, the trainee subject 1 communicates with the training subject 2 through language expression, hand gestures and expressions to convey whether the knowledge content of a certain subject X is understood. In the process of interaction, the gesture acquisition device 31 of the acquisition module 3 obtains the hand gestures of the trainee subject 1 at several time points to obtain gesture modality information and transmit the gesture modality information to the multi-modal modeling sub-module 42 of the creation module 4. The sound acquisition device 32 collects the sound of the trainee subject 1 for language expression to obtain sound modality information and transmit the sound modality information to the multi-modal modeling sub-module 42. The expression acquisition device 33 collects the expression actions of the trainee subject 1 at several time points to obtain expression modality information and transmit the expression modality information to the multi-modal modeling sub-module 42. The multi-modal modeling sub-module 42 respectively collects the gesture modality information, the sound modality information and the expression modality information, creates simulation mathematical models corresponding to the three modalities of gestures, sounds and expressions through simulation software, and transmits the three simulation mathematical models to the analysis module 5. The gesture analysis sub-module 52 in the analysis module 4 reads the data in the simulation mathematical model associated with the gesture, analyzes the time ratio of the gesture action in the learning process, and obtains gesture probability information. The sound analysis sub-module 52 obtains the data in the simulation mathematical model associated with the sound, analyzes the time ratio of the sound in the learning process, and obtains sound probability information. The expression analysis sub-module 53 obtains the data in the simulation mathematical model associated with the expression, analyzes the time ratio of the expression in the learning process, and obtains expression probability information. At the same time, the class test sub-module 61 sends class test questions to the trainee subject 1 to detect the effect of the trainee subject 1 on the learning knowledge content, obtains class detection information, and transmits the class detection information to the subject evaluation sub-module 51 of the analysis module 5. After the stage learning is completed, the stage test sub-module 62 sends stage test questions to the trainee subject 1 to detect the effect of the trainee subject 1 on the knowledge content in the learning process in a certain stage, obtains stage detection information, and transmits the stage detection information to the subject evaluation sub-module 51. The subject evaluation sub-module 51 compares the target information preset therein with the class detection information and the stage detection information, obtains class defect information and stage defect information, and transmits the two to the decision module 7. The decision module 7 combines the gesture probability information, the sound probability information and the expression probability information with the class defect information and the stage defect information to provide the trainee subject 1 with knowledge points and reasons that exist in the class learning process and are not mastered or not mastered, and give suggestions for improvement. Or provide a method for the training subject 2 to help the trainee subject 1 to find and fill the gaps or improve, so as to achieve the purpose of efficient and high-quality learning.

[0085] In summary, the system composed of the acquisition module 3, the creation module 4, the analysis module 5, the detection module 6 and the decision module 7 can completely capture the learning situation of the trainee 1 in the training process, and the multi-modal model established by the creation module 4 is analyzed and summarized by the analysis module 5, the decision module 7 gives the trainee 1 the recommended suggestion of making up the knowledge points of the subject missing, and also gives the decision method of how the training subject 2 helps the trainee 1 to improve, after the system is built, it will have intelligent data information acquisition ability, data information analysis ability, diagnosis ability, feedback ability, improvement effect evaluation ability, with full sample, information multi-dimensional and multi-modal, real, accompanying, personalized, feedback instant and other characteristics, efficient service for the trainee 1 to learn efficiently, reduce the burden of teachers and students. Through big data and various analysis techniques, the knowledge mastery of students, the difference in ability level, and the behavior characteristics and personality characteristics of students can be effectively diagnosed and analyzed and fed back, helping teachers to teach students according to their aptitude and helping students to solve problems in the learning process.

[0086] The specific embodiments are only an explanation of the application, and are not a limitation of the application, and those skilled in the art can make modifications to the embodiments without creative contribution according to the needs after reading the specification, as long as the modifications are within the protection scope of the application.

Claims

1. A data-based intelligent education and training management system for guiding a training subject (2) or a trainee subject (1) to make up for a knowledge deficiency existing in a training process, characterized in that, The application relates to a training system for a trainee (1), which comprises: a collection module (3) for collecting multi-modal information of the trainee (1), the multi-modal information being gestures, expressions and sounds embodied by the trainee during the training process, so as to grasp the interaction degree of the trainee (1); a creation module (4) in communication connection with the collection module (3) and used for establishing a multi-modal learning model of the multi-modal information; the creation module (4) comprises: a defect modeling submodule (41) in communication connection with the detection module (6) and used for creating a defect model by editing the defect information through a matrix, the defect modeling submodule (41) realizing the creation of the defect model by running a matrix operation function in simulation software; a multi-modal modeling submodule (42) in communication connection with the collection module (3) and used for editing the multi-modal information into different modal matrices, creating a multi-modal model through a nonlinear dynamic system, and respectively obtaining a gesture model, a sound model and an expression model, the multi-modal modeling submodule (42) realizing the creation of the defect model by running a matrix operation, a drawing function and a data function in the simulation software; an analysis module (5) in communication connection with the creation module (4) and used for acquiring the multi-modal model, calculating and analyzing the probability of each mode according to a preset algorithm, and obtaining modal probability information; a detection module (6) in communication connection with the analysis module (5) and used for collecting detection information of the trainee (1) and transmitting the detection information to the analysis module (5), the analysis module (5) being used for comparing the detection information with target information, obtaining defect information, the creation module (4) being used for establishing a defect information model according to the defect information and obtaining a defect model; wherein the target information is knowledge information required by each subject for the trainee (1) to reach a training target and pre-set in the analysis module (5), and the defect information is knowledge information different between the target information and the detection information; a decision module (7) in communication connection with the analysis module (5) and used for combining the modal probability information with the defect information, associating each modal probability with the defect model of the corresponding subject, so that the trainee (1) understands the defect points of the training, and the decision module (7) gives the trainee (1) an improvement suggestion or guides the training subject (2) on how to train the trainee (1) to grasp the defect information.

2. The data-based intelligent education training management system according to claim 1, wherein The collection module (3) comprises: a gesture collection device (31) in communication connection with the creation module (4) and used for acquiring gesture information of the trainee (1) during the training process; a sound collection device (32) in communication connection with the creation module (4) and used for acquiring sound information communicated between the trainee and the training subject (2); an expression collection device (33) in communication connection with the creation module (4) and used for collecting facial expressions of the trainee (1) during the training process, so as to acquire the training state of the trainee during the training process.

3. The data-based intelligent education training management system according to claim 2, wherein The gesture collection device (31) is a gesture recognition device.

4. The data-based intelligent education training management system according to claim 2, wherein, The sound collection device (32) is a distance education sound collection device (32).

5. The data-based intelligent education training management system according to claim 2, wherein, The expression acquisition device (33) is selected from a facial expression capture device.

6. The data-based intelligent education training management system according to claim 1, wherein The analysis module (5) comprises: A subject evaluation submodule (51) in communication with the detection module (6) for obtaining the detection information and comparing the detection information with the target information to obtain the defect information; A gesture analysis submodule (52) in communication with the multi-modal modeling submodule (42) for analyzing the probability of the appearance of the gesture model to obtain gesture probability information; A sound analysis submodule (53) in communication with the multi-modal modeling submodule (42) for analyzing the probability of the appearance of the sound to obtain sound probability information; An expression analysis submodule in communication with the multi-modal modeling submodule (42) for analyzing the probability of the appearance of the expression in the interactive state during training to obtain sound probability information.

7. The data-based intelligent education training management system according to claim 6, wherein, The detection module (6) comprises: A class detection submodule (61) in communication with the subject evaluation submodule (51) for collecting the condition of the subject (1) mastering the subject knowledge during the training to obtain class detection information and transmitting the class detection information to the subject evaluation submodule (51); A stage detection submodule (62) in communication with the subject evaluation submodule (51) for collecting the condition of the subject (1) mastering the subject knowledge in the current time after the stage training to obtain stage detection information and transmitting the stage detection information to the subject evaluation submodule (51); The subject evaluation submodule (51) is configured to compare the class detection information and the stage detection information with the target information to obtain class defect information and stage defect information, respectively.

8. The data-based intelligent education training management system according to claim 7, wherein, The decision module (7) is in communication with the subject evaluation submodule (51), the gesture analysis submodule (52), the sound analysis submodule (53), and the expression analysis submodule, respectively, for providing the class defect information, the stage defect information, the gesture probability information, the sound probability information, and the sound probability information to the subject (1) or the trainer (2) and giving promotion suggestions for the class defect information and the stage defect information.

Images