CN117056538A

CN117056538A - Teaching data generation method, device, equipment and storage medium

Info

Publication number: CN117056538A
Application number: CN202311031180.5A
Authority: CN
Inventors: 王士进; 陈玉珏; 占进冬; 汪洋; 朱香; 赖学武; 郐吉丰
Original assignee: iFlytek Co Ltd
Current assignee: iFlytek Co Ltd
Priority date: 2023-08-14
Filing date: 2023-08-14
Publication date: 2023-11-14

Abstract

The invention provides a method, a device, equipment and a storage medium for generating teaching data, wherein the method for generating the teaching data comprises the following steps: the method comprises the steps of obtaining unit teaching design requirements input by a user, calling a preset field large model, obtaining teaching intention according to the unit teaching design requirements, retrieving knowledge information related to the teaching intention from a field knowledge base, calling the field large model, generating unit teaching design contents according to the unit teaching design requirements and the knowledge information related to the teaching intention, further generating lesson teaching activity contents according to the unit teaching design contents, and generating lesson teaching activity courseware according to the lesson teaching activity contents. The teaching data generation method provided by the embodiment of the invention can intelligently generate teaching data, and the generated teaching data can be directly utilized or utilized after being modified by a user, and can also be used as a reference basis for the user to design the teaching data, so that the burden of the user can be greatly reduced.

Description

Teaching data generation method, device, equipment and storage medium

Technical Field

The present invention relates to the field of information generation, and in particular, to a method, an apparatus, a device, and a storage medium for generating teaching data.

Background

The teaching data (such as teaching design) is data according to which a teacher performs teaching. At present, the teaching data is usually designed by teachers according to lessons, teaching contents and teaching objects, the design of the teaching data is a difficult and heavy work, if the teaching data can be intelligently generated, the burden of the teachers can be greatly lightened, however, how to intelligently generate the teaching data is a problem which needs to be solved at present.

Disclosure of Invention

In view of this, the invention provides a method, a device and a storage medium for generating teaching data, which are used for intelligently generating teaching data so as to reduce the burden of teachers, and the technical scheme is as follows:

in a first aspect, a method for generating teaching data is provided, including:

acquiring a unit teaching design requirement input by a user;

calling a preset large field model, and acquiring teaching intention according to the unit teaching design requirement;

retrieving knowledge information related to the teaching intent from a domain knowledge base;

invoking the large field model, and generating unit teaching design contents as target unit teaching design contents according to the unit teaching design requirements and knowledge information related to the teaching intention;

The large domain model is obtained by training data of education domains and combining specified tasks, and the specified tasks at least comprise an intention recognition task and a unit teaching design generation task.

Optionally, the training process of the domain big model includes:

training the initial large domain model by adopting unlabeled training data in the education field to obtain a first large domain model;

and training the first domain large model by adopting marked training data of the education domain in combination with the appointed task to obtain a second domain large model as a final domain large model.

Optionally, the teaching data generating method further includes:

acquiring first content to be modified selected from generated unit teaching design content by a user, and acquiring modification requirements of the user on the first content and contextual information of the first content;

and calling the large domain model, modifying the first content according to the modification requirement of the first content and the context information of the first content, and taking the modified unit teaching design content as a target unit teaching design content.

Optionally, the target unit teaching design content includes part or all of the following: the unit theme, the learning task group to which the unit belongs, the unit teaching content, the unit teaching target, the situation task design content and the learning task design content;

The learning task design content comprises a plurality of task information, and each task information comprises part or all of the following information: task topics, task targets, task activity information, the number of hours required for a task, and task activity information including the names of the teaching activities in the hours.

Optionally, the designating task further includes: generating tasks for teaching activities in class time;

the teaching data generation method further comprises the following steps:

acquiring a user-specified class time teaching activity name from the target unit teaching design content as a target class time teaching activity name;

and calling the large field model, and generating the time teaching activity content corresponding to the target time teaching activity name as target time teaching activity content according to the target time teaching activity name and the target unit teaching design content.

Optionally, the teaching data generating method further includes:

acquiring second content to be modified selected from generated lesson teaching activity content by a user, and acquiring modification requirements of the user on the second content and contextual information of the second content;

and calling the large domain model, and modifying the second content according to the modification requirement of the second content and the context information of the second content, wherein the modified lesson teaching activity content is used as the target lesson teaching activity content.

Optionally, the teaching data generating method further includes:

and generating a class time teaching activity courseware according to the target class time teaching activity content.

Optionally, the designating task further includes: generating tasks by courseware outline;

generating a lesson teaching activity courseware according to the target lesson teaching activity content, wherein the lesson teaching activity courseware comprises the following steps:

invoking the field big model, and generating a courseware outline according to the target teaching activity content in class, wherein the courseware outline comprises the key point content of each page of courseware;

and generating a class time teaching activity courseware according to the class of the courseware.

Optionally, the designating task further includes: courseware content planning tasks;

generating a courseware of the teaching activities according to the courseware outline, wherein the method comprises the following steps:

invoking the large domain model, and determining constituent elements of each courseware and element content requirements corresponding to the constituent elements according to the gist content of each courseware;

acquiring element content corresponding to the constituent elements of each courseware according to the element content requirements corresponding to the constituent elements of each courseware;

generating the lesson teaching activity courseware according to the constituent elements of each courseware and the element content corresponding to the constituent elements of each courseware.

Optionally, the designating task further includes: a keyword recognition task;

the obtaining the element content corresponding to the constituent elements of each courseware according to the element content requirement corresponding to the constituent elements of each courseware comprises the following steps:

for each page of courseware:

invoking the large domain model, and identifying keywords from element content requirements corresponding to the constituent elements of the courseware;

and retrieving the materials related to the keywords from a multi-mode material library based on the keywords, and taking the materials as element contents corresponding to the constituent elements of the courseware.

Optionally, the designating task further includes: generating a material;

the method for obtaining the element content corresponding to the constituent elements of each courseware according to the element content requirements corresponding to the constituent elements of each courseware further comprises the following steps:

and if the materials related to the keywords are not retrieved from the multi-mode material library, calling the large domain model, and generating element contents corresponding to the constituent elements of the page courseware according to the element content requirements corresponding to the constituent elements of the page courseware.

Optionally, the designating task further includes: a structured graph generation task and/or a keyword recognition task;

The teaching data generation method further comprises the following steps:

acquiring text content selected by a user from the generated lesson teaching activity courseware; calling the large domain model, and generating a structured chart according to the text content;

and/or the number of the groups of groups,

acquiring material acquisition requirements of users; invoking the large domain model, and identifying keywords from the material acquisition requirements; retrieving material related to the identified keywords from a multimodal material library based on the identified keywords; and adding the retrieved materials into the generated lesson teaching activity courseware.

In a second aspect, there is provided a teaching data generating apparatus, including: the system comprises a teaching design requirement acquisition module, a teaching intention acquisition module, a domain knowledge retrieval module and a unit teaching design content generation module;

the teaching design requirement acquisition module is used for acquiring unit teaching design requirements input by a user;

the teaching intention acquisition module is used for calling a preset large domain model and acquiring teaching intention according to the unit teaching design requirement;

the domain knowledge retrieval module is used for retrieving knowledge information related to the teaching intention from a domain knowledge base;

The unit teaching design content generation module is used for calling the field large model, and generating unit teaching design content serving as target unit teaching design content according to the unit teaching design requirement and knowledge information related to the teaching intention;

In a third aspect, there is provided a teaching data generating apparatus, comprising: a memory and a processor;

the memory is used for storing programs;

the processor is configured to execute the program to implement each step of the teaching data generating method described in any one of the above.

In a fourth aspect, there is provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the teaching data generation method of any of the above.

According to the teaching data generation method provided by the invention, firstly, the unit teaching design requirements input by a user are acquired, then, a preset field large model is called, the teaching intention is acquired according to the unit teaching design requirements, then, knowledge information related to the teaching intention is retrieved from a field knowledge base, and finally, the field large model is called, and unit teaching design contents are generated according to the unit teaching design requirements and the knowledge information related to the teaching intention. The teaching data generation method provided by the invention can intelligently generate the unit teaching design content according to the unit teaching design requirement of the user, and the generated unit teaching design content can be directly utilized or modified by the user and can also be used as a reference basis for teaching data design of the user, so that the burden of the user can be greatly reduced.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present invention, and that other drawings can be obtained according to the provided drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic flow chart of a teaching data generating method according to an embodiment of the present invention;

FIG. 2 is an example of a cell teaching design interface provided by an embodiment of the present invention;

FIG. 3 is another example of a cell teaching design interface provided by an embodiment of the present invention;

FIG. 4 is an example of a unit teaching design outline and specific unit teaching design content provided by an embodiment of the present invention;

FIG. 5 is a schematic flow chart of obtaining teaching activity content in class according to an embodiment of the present invention;

FIG. 6 is a schematic flow chart of generating a lesson teaching activity courseware according to the lesson teaching activity content provided by the embodiment of the invention;

FIG. 7 is an example of converting text in a class of a teaching activity at a time into a mind map provided by an embodiment of the present invention;

Fig. 8 is a schematic structural diagram of a teaching data generating device according to an embodiment of the present application;

fig. 9 is a schematic structural diagram of a teaching data generating device according to an embodiment of the present application.

Detailed Description

Before describing the inventive solution, the English language referred to in this text is explained first:

prompt: an instruction is indicated. In a dialogue with an AI (e.g. a large artificial intelligence model), an instruction needs to be sent to the AI, which may be a text description or a parameter description according to a certain format.

Large artificial intelligence model: the model is an artificial intelligent model based on a deep learning technology, consists of hundreds of millions of parameters, and can realize complex tasks through learning and training of a large amount of data. The large model in the field of the application is a large artificial intelligent model.

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.

Referring to fig. 1, a flow chart of a method for generating teaching data according to an embodiment of the present invention is shown, where the method may include:

step S101: and obtaining the unit teaching design requirements input by the user.

A user may input a unit teaching design requirement at a unit teaching design interface, please refer to fig. 2, which shows an example of a unit teaching design interface, and "i want" the time quality, craftsman spirit "teaching design" in fig. 2 is the unit teaching design requirement input by the user.

In this embodiment, the unit teaching design requirement input by the user may be, but not limited to, some or all of the following information: teaching material information (such as the necessary repair of high school Chinese), unit information (such as the unit of what number), the names of lessons contained in the unit, unit subjects (the teaching subjects of the unit defined by the teaching material), unit teaching targets, unit teaching methods and the like.

The unit teaching design requirement input by the user can be "generate a second unit which is necessary to be repaired by the high school Chinese", the teaching design of the time quality and craftsman spirit theme is required to embody the unit teaching concept ", in this example," the second unit "is unit information," the time quality and craftsman spirit "is unit theme, and" embody the unit teaching concept "is unit teaching method.

In order to improve the teaching design generation effect, the unit teaching design interface can display input guide information, and the input guide information is used for guiding a user to input key information so as to improve the quality of generated teaching design content. Referring to fig. 3, another example of a unit teaching design interface is shown, where the unit teaching design interface shown in fig. 3 includes an input box, in which input guiding information "please fill in a unit theme of the unit teaching design, a contained text name, and a unit teaching target …" is displayed, and a user can input the unit teaching design requirement in the input box according to the input guiding information, "the unit theme" shown in fig. 3: dream of red blood cells "," lesson names: the whole book reads "dream of Red mansions", "Unit teaching goal: reading the full book of the dream of the red blood cell, the experience author … is the unit teaching design requirement of the user for inputting the guiding information according to the input.

In addition, it should be noted that, the user may input the teaching design requirement through the text input mode, or may input the teaching design requirement through the voice input mode, as shown in fig. 3, the user may directly input the text in the input box in fig. 3, or may input the voice through long-press the voice input identifier in the lower right corner of the input box. If the unit teaching design requirement obtained in the step S101 is input by the user through a text input mode, the subsequent step S102 is directly executed after the user input is obtained, and if the unit teaching design requirement obtained in the step S101 is input by the user through a voice input mode, the voice is converted into the text, and then the subsequent step S102 is executed.

Step S102: and calling a preset large field model, and acquiring teaching intention according to the unit teaching design requirement.

The preset large domain model is called, and the acquired teaching intention can be, but is not limited to, part or all of a teaching material version, a unit name, a keyword and the like according to the unit teaching design requirement.

For example, if the unit teaching design requirement is "please help me generate a large unit teaching design of a seventh unit which must be repaired by a high school language and read" dream of red blood cells "in the whole book," a preset field large model is called, and according to the unit teaching design requirement, the obtained teaching intention can be the teaching material version "must be repaired by a high school language", the unit name "seventh unit", the keyword "dream of red blood cells" and the like.

Specifically, the process of calling a preset large domain model and obtaining the teaching intention according to the design requirement of the unit teaching can comprise the following steps: acquiring a preconfigured first template in a template format, wherein the first template in the template format comprises a unit teaching design requirement information slot, and the first template in the template format is used for indicating a large domain model to identify teaching intention for information in the unit teaching design requirement information slot; filling the unit teaching design requirements input by the user into a unit teaching design requirement information groove to obtain a first edited instruction prompt; inputting the edited first instruction prompt into the large domain model to obtain the teaching intention output by the large domain model.

Step S103: knowledge information related to the teaching intent is retrieved from the domain knowledge base.

The domain knowledge base is a knowledge base of the education domain, at least comprising knowledge of the teaching scene, and also comprising knowledge of other scenes of the education domain. The knowledge of the teaching scene in the domain knowledge base may include a plurality of fields and contents corresponding to the plurality of fields respectively.

The knowledge of the teaching scene in the domain knowledge base may include fields such as a unit guide, a unit theme, a course name, a keyword, and contents corresponding to the fields respectively, and accordingly, a search result obtained by searching knowledge information related to the teaching intention from the domain knowledge base is content information corresponding to the fields such as the unit guide, the unit theme, the course name, and the keyword.

Each piece of knowledge in the domain knowledge base is provided with a vector representation, when knowledge information related to the teaching intention is retrieved from the domain knowledge base, the vector representation of the teaching intention can be obtained first, then the similarity between the vector representation of the teaching intention and the vector representation of each piece of knowledge in the teaching scene in the domain knowledge base is calculated, and further the knowledge information related to the teaching intention is determined according to the calculated similarity.

Step S104: and calling a large field model, and generating unit teaching design contents according to the unit teaching design requirements and knowledge information related to teaching intention.

And generating the unit teaching design content by using knowledge information related to the teaching intention as background knowledge according to the unit teaching design requirement by the large domain model.

Alternatively, the generated element tutorial design content may include element tutorial design outline and specific element tutorial design content. Referring to fig. 4, an example of a unit teaching design outline and specific unit teaching design contents is shown, wherein the left side content in fig. 4 is the unit teaching design outline, and the right side is the specific unit teaching design contents.

The process of generating the unit teaching design content according to the unit teaching design requirement and knowledge information related to the teaching intention may include: acquiring a preconfigured second template in a real-time format, wherein the second template in the real-time format comprises a unit teaching design requirement information groove and a domain knowledge information groove, and is used for indicating a domain big model to generate unit teaching design contents according to information in the unit teaching design requirement information groove and combining the information in the domain knowledge information groove; filling the unit teaching design requirements into a unit teaching design requirement information groove, and filling knowledge information related to teaching intention into a domain knowledge information groove to obtain an edited second instruction prompt; inputting the edited second instruction command prompt into the field large model to obtain the structured content of the unit teaching design output by the field large model, and after obtaining the structured content of the unit teaching design, analyzing the structured content of the unit teaching design to further call a teaching design plug-in to generate the unit teaching design content which can be finally presented to a user.

Alternatively, the cell teaching design content may include some or all of the following: the unit theme, a learning task group to which the unit belongs (the learning task group refers to sixteen language learning capability defined by a new lesson standard), unit teaching content, a unit teaching target, situation task design content and learning task design content, wherein the learning task design content comprises a plurality of task information, and each task information can comprise part or all of the following information: task topics, task targets, task activity information, the number of hours required for a task, and task activity information including the name of the time teaching activity for each hour.

The following table shows an example of the unit teaching design contents generated by the above steps S101 to S104:

table 1 cell teaching design content example

Optionally, after generating the unit teaching design content, the first content to be modified selected by the user from the generated unit teaching design content can be obtained, the modification requirement of the user on the first content and the context information of the first content are obtained, the large domain model is further called, and the first content is modified according to the modification requirement of the first content and the context information of the first content, so that the modified unit teaching design content is obtained. That is, after generating the unit teaching design content, the user may select a part of the content from the generated unit teaching design content, and input a modification requirement for the selected content, thereby triggering modification for the selected content. In addition, the user can directly and autonomously modify the generated unit teaching design content.

The large domain model in this embodiment is obtained by training data (at least including training data of a teaching scene) of an educational domain in combination with a specified task. Specifically, first, training an initial large domain model by adopting unlabeled training data of the education domain (i.e. performing non-supervision training) so that the large domain model learns relevant knowledge of the education domain to obtain a large first domain model, and then, combining an appointed task, training the large first domain model by adopting labeled training data of the education domain (i.e. performing supervision training) so that the large domain model has the capability of executing the appointed task to obtain a large second domain model, and taking the large domain model as a final large domain model.

The specific tasks at least comprise an intention recognition task and a unit teaching design generation task, and it is required to say that the training is performed by combining the intention recognition task so as to enable the large domain model to have the capability of acquiring teaching intention according to the unit teaching design requirement, and the training is performed by combining the unit teaching design generation task so as to enable the large domain model to have the capability of generating unit teaching design content.

Optionally, during unsupervised training, portions of the training data may be masked or replaced, allowing the domain large model to predict the masked or replaced data. When the supervision training is performed, the training data corresponding to the appointed task can be adopted for training, wherein the training data corresponding to the intention recognition task comprises unit teaching design requirements and instruction prompt for instructing the large domain model to execute the intention recognition task, the labeling information is standard teaching intention, the training data corresponding to the intention recognition task is input into the large domain model during training, the large domain model is trained by aiming at the teaching intention acquired by input to approach the standard teaching intention, the training data corresponding to the unit teaching design task comprises unit teaching design requirements and instruction prompt for instructing the large domain model to execute the unit teaching design task, and the training data corresponding to the unit teaching design task is input into the large domain model during training, so that the unit teaching design content generated based on the large domain model approaches the standard unit teaching design content to train.

According to the teaching data generation method provided by the embodiment of the invention, firstly, the unit teaching design requirement input by a user is acquired, then, a preset domain big model is called, the teaching intention is acquired according to the unit teaching design requirement, then, knowledge information related to the teaching intention is retrieved from a domain knowledge base, and finally, the domain big model is called, and unit teaching design content is generated according to the unit teaching design requirement and the knowledge information related to the teaching intention. The teaching data generation method provided by the embodiment of the invention can intelligently generate the unit teaching design content according to the unit teaching design requirement of the user, and the generated unit teaching design content can be directly utilized or modified by the user and can also be used as a reference basis for teaching data design of the user, so that the burden of the user can be greatly reduced.

An embodiment of the present invention provides another teaching data generating method, which is different from the teaching data generating method provided in the foregoing embodiment in that, in addition to the process of obtaining the unit teaching design content (i.e., each step provided in the foregoing embodiment), the method further includes a process of obtaining the teaching activity content in class, and referring to fig. 5, a schematic flow diagram of obtaining the teaching activity content in class is shown, and may include:

Step S501: and acquiring the class time teaching activity name specified by the user from the unit teaching design content as a target class time teaching activity name.

The unit teaching design content may be the unit teaching design content obtained by adopting the steps S101 to S105 in the above embodiment, or may be the unit teaching design content obtained by modifying the unit teaching design content obtained by adopting the steps S101 to S105 in the above embodiment.

The above embodiment mentions that the unit teaching design content may include learning task design content, the learning task design content may include task activity information, the task activity information may include a class time teaching activity name of each class time, and the user may select the class time teaching activity name from the unit teaching design content, thereby triggering generation of the class time teaching activity content.

Step S502: and calling a preset field large model, and generating the lesson teaching activity content corresponding to the target lesson teaching activity name according to the target lesson teaching activity name and the target unit teaching design content.

Specifically, the process of calling a preset field large model and generating the lesson teaching activity content corresponding to the target lesson teaching activity name according to the target lesson teaching activity name and the unit teaching design content may include: acquiring a preconfigured third template in a prompty format, wherein the third template in the prompty format comprises an activity name information slot and a unit teaching design information slot, and the third template in the prompty format is used for indicating the large model in the field to generate teaching activity contents in class according to the information in the activity name information slot and combining the information in the unit teaching design information slot; filling the target lesson teaching activity names into an activity name information slot, and filling the unit teaching design contents into a unit teaching design information slot to obtain a edited third instruction prompt; and inputting the edited third instruction prompt into the field large model to obtain the structured content of the lesson teaching activity output by the field large model, and after obtaining the structured content of the lesson teaching activity, analyzing the structured content of the lesson teaching activity to further call the lesson teaching activity plug-in to generate the lesson teaching activity content which can be finally presented to the user.

Optionally, after selecting the name of the teaching activity at the time of the lesson from the unit teaching design content, the user can supplement and input some requirements, such as requirements on the theme of the activity, learning condition of students, teaching habit and the like, and after selecting the name of the teaching activity at the time of the lesson and inputting the requirements, the user triggers the generation of the content of the teaching activity at the time of the lesson. If some requirements are input by the user, a preset large domain model is called, and according to the target class time teaching activity name, the unit teaching design content and the requirements input by the user, class time teaching activity content corresponding to the target class time teaching activity name is generated (the third template in the template format also comprises a user requirement information slot, and further the requirements input by the user are required to be filled into the user requirement information slot).

By way of example, the lesson activity content may include one or more of the following: activity importing, activity target, activity process, activity nodule, activity expanding and arranging operation. Table 2 below gives an example of the content of a lesson teaching activity generated based on the above method, which is "task one" in table 1 above: "Activity one" of the red building story: identification of content of "dream of Red mansions" corresponding to the content of teaching activities in class time:

Table 2 lesson teaching activity content example

Optionally, after generating the lesson teaching activity content corresponding to the target lesson teaching activity name, the second content to be modified selected by the user from the generated lesson teaching activity content can be obtained, the modification requirement of the user on the second content and the context information of the second content are obtained, then the field big model is called, and the second content is modified according to the modification requirement of the second content and the context information of the second content, namely, after generating the lesson teaching activity content, the user can select part of the content from the generated lesson teaching activity content, input the modification requirement for the selected content, and further trigger the modification of the selected content. In addition, the user can also directly modify the generated teaching activity content in class.

It should be noted that, the large domain model in this embodiment is also obtained by training data (at least including training data of a teaching scene) of an educational domain in combination with a specified task (performing unsupervised training first and then performing supervised training in combination with the specified task), where the specified task includes, in addition to an intention recognition task and a unit teaching design generation task, a time teaching activity generation task. The training is performed by combining the class teaching activity generating task to enable the field large model to have the capability of generating class teaching activity content.

The training data corresponding to the class teaching activity generation task is an instruction prompt which comprises class teaching activity names and unit teaching design contents and is used for instructing the domain big model to execute the class teaching activity generation task, the marking information is standard class teaching activity contents corresponding to the class teaching activity names in the prompt, and the training data corresponding to the class teaching activity generation task is input into the domain big model during training, so that the class teaching activity contents generated based on the domain big model approach to the standard class teaching activity contents and are used as targets for training.

According to the teaching data generation method provided by the embodiment of the invention, firstly, the unit teaching design content is generated according to the unit teaching design requirement input by the user, and after the unit teaching design content is obtained, the time teaching activity content corresponding to the time teaching activity name appointed by the user from the unit teaching design content can be generated. The teaching data generation method provided by the embodiment of the invention not only can intelligently generate the unit teaching design content according to the unit teaching design requirement of the user, but also can intelligently generate the corresponding time teaching activity content aiming at the time teaching activity name in the unit teaching design content, and the generated unit teaching design content and time teaching activity content can be directly utilized or modified by the user for utilization, and can also be used as a reference basis for the user to conduct teaching data design, so that the burden of the user can be greatly reduced.

An embodiment of the present invention provides still another teaching data generating method, which is different from the teaching data generating method provided in the above second embodiment in that it includes, in addition to a process of obtaining unit teaching design contents (a step of obtaining unit teaching design contents in the first embodiment) and a process of obtaining teaching activity contents in class (a step of obtaining teaching activity contents in class in the second embodiment), a process of obtaining teaching activity contents in class: and generating a lesson teaching activity courseware according to the acquired lesson teaching activity content.

Referring to fig. 6, a flow diagram of generating a lesson course from lesson course content is shown, which may include:

step S601: and calling a field outline model, and generating a courseware outline according to the target teaching activity content.

The class of the teaching activities at class time comprises the key point content of each class.

Specifically, the process of calling the domain big model and generating the courseware outline according to the target teaching activity content may include: acquiring a preconfigured fourth template in a real-time teaching activity information slot, wherein the fourth template in the real-time teaching activity information slot is used for indicating a large field model to generate a courseware outline according to information in the real-time teaching activity information slot; filling the target teaching activity content in a teaching activity information slot to obtain a fourth instruction prompt after editing; inputting the edited fourth instruction prompt into the field large model to obtain a courseware outline output by the field large model.

Step S602: generating the courseware of the teaching activities according to the courseware outline.

Specifically, according to the courseware outline, the process of generating the courseware of the teaching activities in class may include:

and step S6021, calling a large domain model, and determining the constituent elements of each courseware and the element content requirements corresponding to the constituent elements according to the point content of each courseware.

It should be noted that, the constituent elements of a page of courseware are used to indicate what mode of the page of courseware is required, for example, the constituent elements of a page of courseware may be text and pictures, i.e. the constituent elements of a page of courseware may be one or a combination of more of text, pictures, video, audio, etc. The element content requirement is used for indicating the required element content, for example, a text corresponding to the text is used for indicating what kind of content is required, a picture corresponding to the element content requirement is used for indicating what kind of content is required, a video corresponding to the element content requirement is used for indicating what kind of content is required, and a video corresponding to the audio is used for indicating what kind of content is required.

Step S6022, obtaining the element content corresponding to the constituent elements of each courseware according to the element content requirements corresponding to the constituent elements of each courseware.

Optionally, for each courseware, firstly, calling a domain big model, identifying keywords from element content requirements corresponding to the constituent elements of the page courseware, and then, based on the identified keywords, retrieving materials related to the keywords from a multi-mode material library to serve as element contents corresponding to the constituent elements of the page courseware.

The multi-mode material library comprises materials of multiple modes, such as text materials, picture materials, video materials, audio materials and the like. And if the constituent elements of one class are texts and pictures, identifying keywords from element content requirements corresponding to the texts, retrieving text materials related to the keywords from a multi-mode material library based on the identified keywords, identifying keywords from element content requirements corresponding to the pictures, and retrieving picture materials related to the keywords from the multi-mode material library based on the identified keywords.

The process of calling the domain big model and identifying the keywords from the element content requirements corresponding to the constituent elements of the courseware may include: acquiring a preconfigured fifth template in a prompty format, wherein the fifth template in the prompty format comprises a requirement information groove, and the fifth template in the prompty format is used for indicating a large domain model to identify keywords from information in the requirement information groove; filling element content requirements corresponding to the constituent elements of the page courseware into a requirement information groove in a fifth promt format template to obtain a edited fifth instruction promt; inputting the edited fifth instruction prompt into the large domain model to obtain keywords output by the large domain model.

In some cases, the situation that related materials cannot be retrieved from the multi-mode material library based on the identified keywords may occur, at this time, a large domain model may be invoked, element content is generated according to element content requirements, for example, constituent elements of a courseware are text and pictures, related text is not retrieved from the multi-mode material library based on the keywords identified from the element content requirements corresponding to the text, related pictures are not retrieved from the multi-mode material library based on the keywords identified from the element content requirements corresponding to the pictures, the large domain model may be invoked, text is generated according to the element content requirements corresponding to the text, the large domain model may be invoked, and pictures are generated according to the element content requirements corresponding to the pictures.

Of course, after obtaining the element content requirements corresponding to the constituent elements of each courseware through step S6021, the domain big model may be directly called, and the element content corresponding to the constituent elements of each courseware may be generated according to the element content requirements corresponding to the constituent elements of each courseware.

Step S6023, generating the lesson teaching activity courseware according to the constituent elements of each courseware and the element content corresponding to the constituent elements of each courseware.

Specifically, a courseware module (such as a PPT template) is selected from a courseware template library according to the constituent elements of each courseware and the element content corresponding to the constituent elements of each courseware, and then a courseware for teaching activities in class is generated according to the element content corresponding to the constituent elements of each courseware and the selected courseware template. The process for generating the courseware of the teaching activities in class according to the element content corresponding to the constituent elements of each page of courseware and the selected courseware template comprises the following steps: and filling element contents corresponding to the constituent elements of each page of courseware into the selected courseware templates for typesetting and beautifying.

Optionally, after the generated courseware of the teaching activity in class is generated, the generated courseware of the teaching activity in class can be modified, when the generated courseware of the teaching activity in class is modified, the user can independently modify the generated courseware of the teaching activity in class, and the content to be modified can be selected from the generated courseware of the teaching activity in class, so that the modification of the selected content is triggered. For user-selected content, the domain big model may be invoked to modify it, e.g., the user feels that a piece of text is too long to be converted into a structured chart (e.g., a mind map), the user may select the piece of text, and for user-selected text, the domain big model may be invoked to generate a structured chart (e.g., a mind map) from the selected text. Referring to fig. 7, an example of converting text in a class of teaching activities at class into a mind map is shown.

Optionally, after generating the class of teaching activities, the large domain model can be invoked to modify the content selected by the user, and the material acquisition requirement input by the user can be acquired, the large domain model is invoked to identify keywords from the material acquisition requirement, and the materials are retrieved from the multi-mode material library according to the identified keywords, so that the retrieved materials are added to the class of teaching activities.

For example, the user may input "please provide me with several pictures describing Gu Fu gate in the text" dream of red building ", after obtaining the material acquisition requirement input by the user, the domain big model may be invoked to identify keywords from the material acquisition requirement, and if the keywords" dream of red building "and" Gu Fu gate "are identified, further according to" dream of red building "and" Gu Fu gate ", the picture materials related to" dream of red building "and" Gu Fu gate "are retrieved from the multi-mode material library, and then the retrieved picture materials are added to the lesson teaching activity courseware. If the picture materials related to the 'dream of the red blood cell' and the 'Gu Fu gate' are not retrieved from the multi-mode material library, the field big model can be called to generate the picture materials related to the 'dream of the red blood cell' and the 'Gu Fu gate'.

It should be noted that, the large domain model in this embodiment is also obtained by training data of the education domain (at least including training data of a teaching scene) together with a designated task (performing unsupervised training first and then performing supervised training together with the designated task), where the designated task may include, in addition to an intention recognition task, a unit teaching design generation task, and a class teaching activity generation task, a courseware outline generation task, a courseware content planning task, a keyword recognition task, a material generation task (such as a text generation task, a picture generation task, an audio generation task, a video generation task, and the like), and a structured chart generation task (such as a thought guide generation task, a table generation task, and the like).

The training modes of the tasks are the same, except that the training data are different, for example, the training data corresponding to the courseware outline generating task are the teaching activity content containing the lesson time, the instruction promtt for instructing the domain big model to execute the courseware outline generating task, the marking information is the standard courseware outline, the training data corresponding to the courseware content planning task are the instruction promtt containing the courseware outline and used for instructing the domain big model to execute the courseware content planning task, the marking information is the standard courseware content planning information, the training data corresponding to the keyword recognition task are the real keywords in the material content requirement and used for instructing the domain big model to execute the keyword recognition task, the training data corresponding to the material generation task are the instruction promt containing the material content requirement and used for instructing the domain big model to execute the material generating task, the marking information is the standard data corresponding to the material content requirement, the training data corresponding to the structured chart generating task are the promt containing a section of text and used for instructing the domain big model to execute the structured chart generating task, and the marking information is the corresponding standard chart structure.

According to the teaching data generation method provided by the embodiment of the invention, firstly, the unit teaching design content is generated according to the unit teaching design requirement input by the user, after the unit teaching design content is obtained, the time teaching activity content corresponding to the time teaching activity name appointed by the user from the unit teaching design content can be generated, and after the time teaching activity content is obtained, the time teaching activity courseware can be generated according to the time teaching activity content. The teaching data generation method provided by the embodiment of the invention not only can intelligently generate the unit teaching design content according to the unit teaching design requirement of the user, but also can intelligently generate the corresponding class time teaching activity content according to the class time teaching activity name in the unit teaching design content, and can also intelligently generate class time teaching activity courseware according to the class time teaching activity content, and the generated teaching data can be directly utilized or modified by the user and then utilized, and can also be used as a reference basis for the teaching data design of the user, so that the burden of the user can be greatly reduced.

The embodiment of the invention provides a teaching data generating device, which is described below, and the teaching data generating device described below and the teaching data generating method described above can be correspondingly referred to each other.

Referring to fig. 8, a schematic structural diagram of a teaching data generating apparatus provided in an embodiment of the present invention is shown, where the teaching data generating apparatus may include: a teaching design requirement acquisition module 801, a teaching intention acquisition module 802, a domain knowledge retrieval module 803 and a unit teaching design content generation module 804.

The teaching design requirement acquisition module 801 is configured to acquire a unit teaching design requirement input by a user.

The teaching intention obtaining module 802 is configured to invoke a preset domain big model, and obtain a teaching intention according to the unit teaching design requirement.

A domain knowledge retrieval module 803, configured to retrieve knowledge information related to the teaching intention from a domain knowledge base.

The unit teaching design content generating module 804 is configured to invoke the domain big model, and generate unit teaching design content as target unit teaching design content according to the unit teaching design requirement and knowledge information related to the teaching intention.

Optionally, the teaching data generating device provided by the embodiment of the present invention may further include: and a model training module. Model training module for:

Optionally, the teaching data generating device provided by the embodiment of the present invention may further include: the system comprises a first content acquisition module, a first modification requirement acquisition module, a first context information acquisition module and a first modification module.

And the first content acquisition module is used for acquiring first content to be modified, which is selected from the generated unit teaching design content by a user.

And the first modification requirement acquisition module is used for acquiring the modification requirement of the user on the first content.

And the first context information acquisition module is used for acquiring the context information of the first content.

The first modification module is used for calling the large domain model, modifying the first content according to the modification requirement of the first content and the context information of the first content, and taking the modified unit teaching design content as a target unit teaching design content.

Optionally, the designating task further includes: the lesson teaching activities generate tasks.

The teaching data generation device may further include: and a lesson teaching activity content generation module.

The teaching activity content generation module is used for:

Optionally, the teaching data generating apparatus may further include: the system comprises a second content acquisition module, a second modification requirement acquisition module, a second context information acquisition module and a second modification module.

And the second content acquisition module is used for acquiring second content to be modified, which is selected from the generated lesson teaching activity content by the user.

And the second modification requirement acquisition module is used for acquiring the modification requirement of the user on the second content.

A second context information acquisition module, configured to acquire context information of the second content;

and the second modification module is used for calling the large domain model, modifying the second content according to the modification requirement of the second content and the context information of the second content, and taking the modified lesson teaching activity content as the target lesson teaching activity content.

Optionally, the teaching data generating apparatus may further include: and a lesson teaching activity courseware generating module.

And the lesson teaching activity courseware generating module is used for generating a lesson teaching activity courseware according to the target lesson teaching activity content.

the time teaching activity courseware generating module is specifically used for generating a time teaching activity courseware according to the target time teaching activity content:

the courseware generating module is specifically used for generating the courseware of the teaching activities in the class according to the outline of the courseware when the courseware is generated:

Optionally, the designating task further includes: a keyword recognition task;

the courseware generating module is specifically used for acquiring the element content corresponding to the constituent elements of each courseware according to the element content requirements corresponding to the constituent elements of each courseware:

For each page of courseware:

Optionally, the designating task further includes: generating a material;

the courseware generating module is also used for:

the teaching data generation device may further include: a structured graph generation module and/or a material acquisition module.

And the structured chart generation module is used for acquiring text contents selected from the generated lesson teaching activity courseware by a user, calling the large field model and generating a structured chart according to the text contents.

The material acquisition module is used for acquiring the material acquisition requirement of the user, calling the large domain model, identifying keywords from the material acquisition requirement, retrieving materials related to the identified keywords from the multi-mode material library, and adding the retrieved materials into the generated lesson teaching activity courseware.

The teaching data generating device provided by the embodiment of the invention can generate the unit teaching design content according to the unit teaching design requirement input by the user, can generate the time teaching activity content corresponding to the time teaching activity name appointed by the user from the unit teaching design content after obtaining the unit teaching design content, and can generate the time teaching activity courseware according to the time teaching activity content after generating the time teaching activity content. The teaching data generating device provided by the embodiment of the invention can intelligently generate the unit teaching design content, the time teaching activity content and the time teaching activity courseware, and the generated teaching data can be directly utilized or modified by a user and can also be used as a reference basis for the user to design the teaching data, so that the burden of the user can be greatly reduced.

An embodiment of the present invention provides a teaching data generating device, referring to fig. 9, which shows a schematic structural diagram of the teaching data generating device, where the teaching data generating device may include: a processor 901, a communication interface 902, a memory 903, and a communication bus 904;

in the embodiment of the present invention, the number of the processor 901, the communication interface 902, the memory 903 and the communication bus 904 is at least one, and the processor 901, the communication interface 902 and the memory 903 complete communication with each other through the communication bus 904;

Processor 901 may be a central processing unit CPU, or a specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present invention, etc.;

the memory 903 may include a high-speed RAM memory, and may further include a non-volatile memory (non-volatile memory), etc., such as at least one disk memory;

wherein the memory stores a program, the processor is operable to invoke the program stored in the memory, the program operable to:

acquiring a unit teaching design requirement input by a user;

Alternatively, the refinement function and the extension function of the program may be described with reference to the above.

The embodiment of the present invention also provides a readable storage medium storing a program adapted to be executed by a processor, the program being configured to:

acquiring a unit teaching design requirement input by a user;

Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A teaching data generation method, characterized by comprising:

acquiring a unit teaching design requirement input by a user;

2. The teaching data generation method according to claim 1, wherein the training process of the domain big model includes:

3. The teaching data generation method according to claim 1, characterized by further comprising:

4. A teaching data generation method according to any of claims 1-3, characterized in that said target unit teaching design content comprises part or all of: the unit theme, the learning task group to which the unit belongs, the unit teaching content, the unit teaching target, the situation task design content and the learning task design content;

5. The teaching data generation method according to claim 4, wherein the specifying task further comprises: generating tasks for teaching activities in class time;

the teaching data generation method further comprises the following steps:

6. The teaching data generation method according to claim 5, characterized by further comprising:

7. The teaching data generation method according to claim 5, characterized by further comprising:

8. The teaching data generation method according to claim 7, wherein the specifying task further comprises: generating tasks by courseware outline;

9. The teaching data generation method according to claim 8, wherein the specifying task further comprises: courseware content planning tasks;

10. The teaching data generation method according to claim 9, wherein the specifying task further comprises: a keyword recognition task;

for each page of courseware:

11. The teaching data generation method according to claim 10, characterized in that the specifying task further comprises: generating a material;

12. The teaching data generation method according to claim 8, wherein the specifying task further comprises: a structured graph generation task and/or a keyword recognition task;

the teaching data generation method further comprises the following steps:

and/or the number of the groups of groups,

13. A teaching data generation device, characterized by comprising: the system comprises a teaching design requirement acquisition module, a teaching intention acquisition module, a domain knowledge retrieval module and a unit teaching design content generation module;

14. A teaching data generating apparatus, characterized by comprising: a memory and a processor;

the memory is used for storing programs;

the processor is configured to execute the program to implement the respective steps of the teaching data generation method according to any one of claims 1 to 12.

15. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the respective steps of the teaching data generation method according to any of claims 1-12.