Teaching model training method and device based on artificial intelligence, equipment and medium

Abstract

The invention discloses a teaching model training method and device based on artificial intelligence, computer equipment and a storage medium. The method comprises the steps of collecting teaching material data; obtaining training data from the teaching material data; and inputting the training data into a teaching model for model training, enabling the teaching model to extract the relationship between the problem solving elements according to the explanation method of the problem through the model training, and asking a question according to the relationship between wrong and / or missing problem solving elements. Through training the teaching model, the teaching model can be converted into a student role, a teaching mode of replacing learning by teaching is realized, and the teaching model can help the students to improve the learning effect and the expression ability of the students.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an artificial intelligence-based teaching model training method, device, equipment and medium. Background Art

[0002] Driven by the development of artificial intelligence, AI big models are gradually being used in the field of education, which can help improve the quality of education, improve teaching efficiency, and increase the utilization rate of higher education resources.

[0003] In the field of education, the Feynman learning method is a very efficient learning method. It can effectively improve students' learning outcomes by "teaching instead of learning", but it relies more on "one-to-one" teaching methods and the teaching cost is too high.

[0004] In view of the above problems, there is a lack of an effective solution in the prior art. Summary of the invention

[0005] Based on this, in order to solve the above problems, this paper proposes a teaching model training method, device, equipment and medium based on artificial intelligence. The present invention trains the teaching model so that it can be transformed into a "student" role, realizing the teaching method of "teaching instead of learning", which can help students improve their learning effects and their expression ability.

[0006] According to a first aspect of the present invention, there is provided a teaching model training method based on artificial intelligence, comprising:

[0007] Data collection steps: collecting teaching material data;

[0008] A training data acquisition step, acquiring training data from the teaching material data, wherein the training data includes: questions acquired from the teaching material data, elements for solving the questions, relationships between the elements for solving the questions, and methods for explaining the questions;

[0009] The teaching model training step is to input the training data into the teaching model for model training. Through the model training, the teaching model can extract the relationship between the problem-solving elements according to the explanation method of the question, and ask questions according to the relationship between the wrong and / or missing problem-solving elements. By asking questions, the teaching model can provide feedback to the teaching model user on the knowledge points that the user has not mastered, thereby improving the teaching effect. In some embodiments, asking questions based on the relationship between the wrong and / or missing problem-solving elements can be transformed into pushing corresponding learning materials based on the relationship between the wrong and / or missing problem-solving elements.

[0010] In some embodiments, the training data includes: a correspondence between the relationship between problem-solving elements and the explanation method of the problem;

[0011] The method for explaining the questions includes: a method for explaining a correct question: being able to obtain complete and accurate relationships between elements for solving the question, a method for explaining an incorrect question: obtaining incorrect relationships between elements for solving the question, and a method for explaining an incomplete question: obtaining incomplete relationships between elements for solving the question.

[0012] In some embodiments, the problem-solving elements are entities and entity attribute values ​​in the relevant knowledge graph obtained based on the question.

[0013] In some embodiments, the relationship between the problem-solving elements is an entity relationship in a knowledge graph.

[0014] In some embodiments, the knowledge graph is constructed by performing entity extraction, relationship extraction, and attribute extraction on teaching material data.

[0015] In some embodiments, the teaching model is an AI model; the teaching model outputs the correct answer to the question when the relationship between the problem-solving elements is complete and accurate.

[0016] In some embodiments, in the teaching model training step, the relationships between erroneous problem-solving elements extracted by the teaching model are manually corrected.

[0017] In some embodiments, the teaching model uses an LSTM model when judging whether the student's explanation method is accurate. For a trained LSTM model, a question is input through an input gate, and the output gate outputs the correct answer to the question when the relationship between the elements of the problem is complete and accurate. Whether the student's explanation method is accurate is judged by whether the correct answer can be output.

[0018] According to a second aspect of the present invention, there is provided a teaching model training device based on artificial intelligence, comprising:

[0019] Data collection module, used to collect teaching material data;

[0020] A training data acquisition module, used to acquire training data from teaching material data, wherein the training data includes: questions acquired from the teaching material data, elements for solving the questions, relationships between the elements for solving the questions, and methods for explaining the questions;

[0021] The teaching model training module is used to input the training data into the teaching model for model training. Through the model training, the teaching model can extract the relationship between the problem-solving elements according to the explanation method of the question, and ask questions according to the relationship between the wrong and / or missing problem-solving elements.

[0022] According to a third aspect of the present invention, there is provided a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above-mentioned method embodiments when executing the computer program.

[0023] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method of any of the above embodiments are implemented.

[0024] The technical solution of the present invention trains the teaching model to transform it into a "student" role. The teaching model can judge whether the relationship between problem-solving elements is complete and accurate, and provide feedback to students on knowledge points that they have not mastered, thereby realizing a teaching method of "teaching instead of learning". In addition, the wrong relationship between problem-solving elements extracted by the teaching model is manually corrected, and the training effect of the teaching model is improved in a "human-in-the-loop" manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 is a flow chart of some embodiments of the teaching model training method based on artificial intelligence of the present invention;

[0026] Figure 2 It is a schematic diagram of a horizontal projectile motion problem;

[0027] Figure 3 It is a schematic diagram of the structure of some embodiments of the teaching model training device based on artificial intelligence of the present invention;

[0028] Figure 4 It is a diagram of the internal structure of a computer device used to implement some embodiments of the present invention. DETAILED DESCRIPTION

[0029] Embodiments of the present invention will be described more fully below with reference to the accompanying drawings, in which embodiments of the present invention are shown. However, the present invention can be embodied in many different forms and should not be construed as limited to the embodiments set forth herein.

[0030] The terms used herein are used only for the purpose of describing specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a", "an", "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that when used herein, the term "comprising" specifies the presence of the claimed features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof.

[0031] Unless otherwise defined, the terms (including technical and scientific terms) used herein have the same meanings as commonly understood by those of ordinary skill in the art to which the present disclosure belongs. The terms used herein should be interpreted as having the meanings consistent with their meanings in the context of this specification and in the relevant field, and should not be interpreted in an idealized or overly formal sense, unless specifically defined as such herein.

[0032] Problem-solving elements: the knowledge points that need to be involved in the process of solving the problem, including: specific names (corresponding numerical values).

[0033] The relationship between problem-solving elements: that is, the relationship between the knowledge points involved in the problem-solving process. Through the relationship between the knowledge points, the logical deduction, numerical calculation and other problem-solving steps between the knowledge points can be completed.

[0034] The corresponding relationship between the problem-solving elements and the explanation method of the question: The corresponding relationship between the explanation method of the question and the relationship between the problem-solving elements obtained by the explanation method of the question. For example, the explanation method of the question is: Since the length is 5 and the width is 4, the area of ​​the rectangle is the length multiplied by the width, and the area is equal to 5 times 4, which is equal to 20. The relationship between the problem-solving method and the problem-solving elements (length × width) is established, and the correct relationship between the problem-solving elements obtained when the problem-solving method is input into the teaching model is length × width.

[0035] Figure 1 Flowcharts showing some embodiments of the artificial intelligence-based teaching model training method of the present invention.

[0036] like Figure 1 As shown, the method includes:

[0037] Data collection step S102, collecting teaching material data;

[0038] The training data acquisition step S104 is to acquire training data from the teaching material data, wherein the training data includes: questions acquired from the teaching material data, elements for solving the questions, relationships between the elements for solving the questions, and methods for explaining the questions;

[0039] The teaching model training step S106 inputs the training data into the teaching model for model training. Through the model training, the teaching model can extract the relationship between the problem-solving elements according to the explanation method of the question, and ask questions according to the relationship between the wrong and / or missing problem-solving elements.

[0040] In some embodiments, the training data includes: a correspondence between the relationship between problem-solving elements and the explanation method of the problem;

[0041] The method for explaining the questions includes: a method for explaining a correct question: being able to obtain complete and accurate relationships between elements for solving the question, a method for explaining an incorrect question: obtaining incorrect relationships between elements for solving the question, and a method for explaining an incomplete question: obtaining incomplete relationships between elements for solving the question.

[0042] In some embodiments, the problem-solving elements are entities and entity attribute values ​​in the relevant knowledge graph obtained based on the question.

[0043] In some embodiments, the relationship between the problem-solving elements is an entity relationship in a knowledge graph.

[0044] In some embodiments, the knowledge graph is constructed by performing entity extraction, relationship extraction, and attribute extraction on teaching material data.

[0045] In some embodiments, the teaching model is an AI model; the teaching model outputs the correct answer to the question when the relationship between the problem-solving elements is complete and accurate.

[0046] In some embodiments, in the teaching model training step, the relationships between erroneous problem-solving elements extracted by the teaching model are manually corrected.

[0047] In some embodiments, the teaching model uses an LSTM model when judging whether the student's explanation method is accurate. For a trained LSTM model, a question is input through an input gate, and the output gate outputs the correct answer to the question when the relationship between the elements of the problem is complete and accurate. Whether the student's explanation method is accurate is judged by whether the correct answer can be output.

[0048] When the teaching model obtained by the teaching model training method is used for student teaching in the present invention, the specific steps include:

[0049] 1. Get the topic

[0050] The problem is a horizontal projectile motion problem. Problem content: At a height of H = 45m above the ground, a small ball A moves with an initial velocity v 0 =10m / s 2 Throw it horizontally, A can be represented as a point mass, air resistance is ignored, and the acceleration due to gravity is g = 10m / s 2 . Find the time when ball A hits the ground and the horizontal distance it is thrown. Figure 2 This is a schematic diagram of a projectile motion problem.

[0051] 2. Import questions into the teaching model, and the teaching model obtains the corresponding knowledge graph based on the questions (including: entities in the knowledge graph, knowledge point graph entity relationships)

[0052] Import the question into the AI ​​(teaching model) (when the AI ​​uses the LSTM model, import the question through the input gate) to obtain or create a knowledge graph related to the question (including knowledge graph entities and entity relationships).

[0053] The entities in the knowledge graph include: A's initial velocity v 0 (Attribute value: 10m / s 2 ), height from the ground H (attribute value: 45m), gravitational acceleration g (attribute value: 10m / s 2 ), landing time TA, horizontal distance SA.

[0054] The entity relationship of the knowledge point graph is as follows: SA = V 0 *TA;

[0055] 3. The teaching model separates the entity relationships from the entities in the knowledge graph to obtain a knowledge graph without entity relationships.

[0056] 4. Students input the explanation method of the question into the teaching model. The teaching model extracts the relationship between the problem-solving elements (entity relationships in the knowledge graph) based on the explanation method of the question and fills in the entity relationships in the knowledge graph.

[0057] For example, students say: Since the height is 45m, the acceleration due to gravity is 10m / s 2 According to the free fall formula, in free fall, the height is equal to the acceleration due to gravity multiplied by the square of time divided by two, so the time is equal to 2 multiplied by the height divided by the square root of the acceleration due to gravity, and the landing time is 3 seconds. Since the air resistance is negligible, the horizontal distance is equal to the time multiplied by the initial velocity, that is, 3 seconds multiplied by 10 meters per second equals 30 meters. At this time, AI can understand what the students are saying and fill in the knowledge graph relationship, that is: SA = V 0 *TA; At this point, AI determines that the entity relationships in the knowledge graph are complete and accurate, and ends the teaching. If the student does not explain the time when the ball lands, then the knowledge graph The entity relationship is missing, and AI will ask students based on the incomplete relationship: "How to calculate the time it takes for the ball to fall to the ground?" If the student answers correctly (for example: in free fall, the height is equal to the acceleration due to gravity multiplied by the square of the time divided by two, so the time is equal to 2 times the height divided by the square root of the acceleration due to gravity, ), then AI will continue to supplement the knowledge graph The teaching ends when the student answers incorrectly or does not answer. If the student does not have a good grasp of the free fall motion knowledge point, the AI ​​will inform the student that the free fall motion knowledge point is insufficient and push the corresponding learning materials. After the student has completed the relevant knowledge point learning, the AI ​​will continue to explain the solution to the problem.

[0058] Figure 3 A structural schematic diagram of some embodiments of the teaching model training device based on artificial intelligence of the present invention is shown.

[0059] like Figure 3 As shown, including:

[0060] The data collection module 100 is used to collect teaching material data;

[0061] The training data acquisition module 200 is used to acquire training data from the teaching material data, wherein the training data includes: questions acquired from the teaching material data, elements for solving the questions, relationships between the elements for solving the questions, and methods for explaining the questions;

[0062] The teaching model training module 300 is used to input the training data into the teaching model for model training. Through the model training, the teaching model can extract the relationship between the problem-solving elements according to the explanation method of the question, and ask questions based on the relationship between the wrong and / or missing problem-solving elements.

[0063] For the specific definition of a teaching model training device based on artificial intelligence, please refer to the definition of a teaching model training method based on artificial intelligence above, which will not be repeated here. Each module in the above-mentioned teaching model training device based on artificial intelligence can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0064] The present invention also provides a computer device, which may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above-mentioned teaching model training method based on artificial intelligence is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc. Those skilled in the art can understand that Figure 4The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0065] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the above-mentioned teaching model training method based on artificial intelligence is implemented.

[0066] Those skilled in the art can understand that all or part of the processes in the above method embodiments can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus), direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0067] So far, the embodiments of the present invention have been described in detail. In order to avoid obscuring the concept of the present invention, some details known in the art are not described. Based on the above description, those skilled in the art can fully understand how to implement the technical solutions disclosed herein.

[0068] Although some specific embodiments of the present invention have been described in detail by way of examples, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It should be understood by those skilled in the art that the above embodiments may be modified or some technical features may be replaced by equivalents without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the appended claims.

Claims

1. A teaching model training method based on artificial intelligence, characterized in that: The following steps are involved: Data collection steps: collecting teaching material data; A training data acquisition step, acquiring training data from the teaching material data, wherein the training data includes: questions acquired from the teaching material data, elements for solving the questions, relationships between the elements for solving the questions, and methods for explaining the questions; The teaching model training step is to input the training data into the teaching model for model training. Through the model training, the teaching model can extract the relationship between the problem-solving elements according to the explanation method of the question, and ask questions according to the relationship between the wrong and / or missing problem-solving elements.

2. The teaching model training method based on artificial intelligence according to claim 1, characterized in that: The training data includes: the corresponding relationship between the relationship between the problem-solving elements and the explanation method of the problem.

3. The teaching model training method based on artificial intelligence according to claim 1, characterized in that: The problem-solving elements are entities and entity attribute values ​​in the relevant knowledge graph obtained based on the question. The relationship between the problem-solving elements is the entity relationship in the knowledge graph.

4. The teaching model training method based on artificial intelligence according to claim 3 is characterized in that: The knowledge graph is constructed by performing entity extraction, relationship extraction, and attribute extraction on teaching material data.

5. The teaching model training method based on artificial intelligence according to claim 1, characterized in that: The asking of questions based on the relationship between the wrong and / or missing problem-solving elements is to push corresponding learning materials based on the relationship between the wrong and / or missing problem-solving elements.

6. The teaching model training method based on artificial intelligence according to claim 1, characterized in that: The teaching model is an AI model; when the relationship between the problem-solving elements is complete and accurate, the teaching model outputs the correct answer to the question.

7. The teaching model training method based on artificial intelligence according to claim 1, characterized in that: In the teaching model training step, the relationship between erroneous problem-solving elements extracted by the teaching model is manually corrected.

8. A teaching model training device based on artificial intelligence, characterized in that: include: Data collection module, used to collect teaching material data; A training data acquisition module, used to acquire training data from teaching material data, wherein the training data includes: questions acquired from the teaching material data, elements for solving the questions, relationships between the elements for solving the questions, and methods for explaining the questions; The teaching model training module is used to input the training data into the teaching model for model training. Through the model training, the teaching model can extract the relationship between the problem-solving elements according to the explanation method of the question, and ask questions based on the relationship between the wrong and / or missing problem-solving elements.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images