A digital teaching material writing method and system based on action logic

Abstract

The present invention relates to the field of educational technology, and specifically to a method and system for writing digital teaching materials based on action logic. The present invention first obtains the teaching material demonstration page of the digital teaching material to be processed based on action logic at each demonstration moment, and all texts and all pictures in each teaching material demonstration page; then all teaching material demonstration pages are segmented to obtain all demonstration stages; then the learning stage of each demonstration stage is determined; finally, all teaching material demonstration pages in the corresponding demonstration stage are written and optimized according to different learning stages. The present invention divides the demonstration stage by capturing the content changes of different teaching material demonstration pages and the demonstration moments, and identifies and judges the learning stage for each demonstration stage, aiming to optimize the presentation of specific content in different learning stages based on the typical learning work links of the action logic of student learning, and improve the writing effect of the digital teaching material to be processed.

Description

Technical Field

[0001] The present invention relates to the field of educational technology, and in particular to a method and system for compiling digital teaching materials based on action logic. Background Art

[0002] Action logic (Psychomotor Logic) refers to the logical relationship between the steps and processes of actions, the physical and psychological conditions of actions, and the goals and results of actions in a specific situation; action logic is also fully applied to the compilation of education and teaching materials, specifically reflected in learning goals, learning paths, feedback mechanisms and decision points based on action principles. Digital teaching materials refer to educational resources produced and published through digital technology, usually presented in the form of multimedia. At present, the compilation of digital teaching materials based on students' action logic is usually compiled in sequence according to the three learning links in the specific learning typical work links of specific learning tasks, including the introduction learning stage, the deep learning stage and the consolidation learning stage; so that digital teaching materials can more effectively promote students' autonomous learning and deep learning, and improve learning efficiency and effectiveness.

[0003] In digital textbooks based on action logic, the presentation form of textbook content is crucial. Optimizing the presentation form of textbook content according to the specific characteristics and manifestations of action logic in different typical demonstration stages of specific learning typical work links of specific learning tasks can help guide students to gradually deepen their thinking logic and gradually improve their action ability, so as to achieve learning goals more effectively. However, the current digital textbooks have serious content homogeneity, and are prone to fall into the trap of copying and pasting or template application of simple concept descriptions, diagrams, and video demonstrations. They lack key identification or knowledge interaction such as concept interaction and action ability improvement, which in turn affects the writing effect of digital textbooks. Summary of the invention

[0004] In order to solve the technical problem of poor writing effect of digital teaching materials in the prior art, the purpose of the present invention is to provide a digital teaching material writing method and system based on action logic, and the technical solution adopted is as follows:

[0005] The present invention proposes a method for compiling digital teaching materials based on action logic, the method comprising:

[0006] Obtain the teaching material demonstration page of the digital teaching material to be processed at each demonstration moment based on the action logic, and obtain all texts and all pictures on the teaching material demonstration page at each demonstration moment;

[0007] According to the display dominance of the text and the picture in the teaching material demonstration page at each demonstration moment, combined with the corresponding demonstration moment, all the teaching material demonstration pages are segmented to obtain all the demonstration stages;

[0008] In each demonstration stage, the learning stage of the typical learning work link of each demonstration stage is determined based on the display differences of the text and pictures in the teaching material demonstration page at all demonstration moments and the pixel changes of the pictures at different demonstration moments;

[0009] According to the different learning stages, all the teaching material demonstration pages in the corresponding demonstration stage are written and optimized.

[0010] Furthermore, the acquisition method in the demonstration phase includes:

[0011] According to the total number of characters corresponding to the text and the total number of pixels corresponding to the picture in the teaching material demonstration page at each demonstration moment, the display dominance coefficient of the teaching material demonstration page at each demonstration moment is obtained;

[0012] Obtaining a time segmentation coefficient for each demonstration moment according to the time interval between each demonstration moment and each preset learning stage segmentation time node in the digital teaching material to be processed;

[0013] At each demonstration moment, the segmentation possibility of each demonstration moment is obtained according to the time segmentation coefficient and the display dominance coefficient corresponding to the teaching material demonstration page; the time segmentation coefficient and the display dominance coefficient are both positively correlated with the segmentation possibility;

[0014] The presentation moment at which the segmentation possibility is greater than a preset threshold is used as the segmentation moment, and the presentation stage is obtained according to the adjacent segmentation moments.

[0015] Furthermore, the method for obtaining the display dominance coefficient includes:

[0016] In the teaching material demonstration page at each demonstration moment, the total number of characters corresponding to the text is used as the text display coefficient, and the ratio of the total number of pixels corresponding to the picture to the overall resolution of the teaching material demonstration page is used as the picture display coefficient;

[0017] The text display coefficient plus a first preset non-zero normal number is used as a numerator, the picture display coefficient plus a second preset non-zero normal number is used as a denominator, and the fraction ratio is used as a text display dominant parameter; the reciprocal of the text display dominant parameter is used as a picture display dominant parameter;

[0018] The maximum value of the text display dominant parameter and the picture display dominant coefficient is normalized, and the normalized result is used as the display dominant coefficient at the corresponding demonstration moment.

[0019] Furthermore, the method for obtaining the time division coefficient includes:

[0020] In the future time domain of each demonstration moment, the time interval between the demonstration moment and each preset learning stage segmentation time node is obtained respectively, and the negative correlation mapping result of the minimum time interval is used as the time segmentation coefficient corresponding to the demonstration moment.

[0021] Furthermore, the method for acquiring the demonstration stage according to the adjacent segmentation moments includes:

[0022] The teaching material demonstration page corresponding to the first demonstration moment is used as the stage starting page of the first demonstration stage, and the first segmentation moment is obtained by traversing along the time sequence of the demonstration moments, and the teaching material demonstration page corresponding to the first segmentation moment is used as the stage ending page of the first demonstration stage;

[0023] The teaching material demonstration page corresponding to the next adjacent demonstration moment of the first segmentation moment is used as the stage starting page of the second demonstration stage, and the second segmentation moment is obtained by traversing along the time sequence of the demonstration moments, and the teaching material demonstration page corresponding to the second segmentation moment is used as the stage ending page of the second demonstration stage;

[0024] The teaching material demonstration page corresponding to the next adjacent demonstration moment of the second segmentation moment is used as the starting point page of the third demonstration stage, and all demonstration stages are obtained by continuous iteration.

[0025] Furthermore, the method for determining the learning stage of the learning typical work link in each demonstration stage includes:

[0026] The learning stages of the typical learning work links at least include an introduction learning stage, an in-depth learning stage, and a consolidation learning stage;

[0027] In each demonstration stage, according to the dominant parameters of the text display in the teaching material demonstration page at all demonstration moments, it is determined whether the demonstration stage is a text demonstration stage; if the demonstration stage is a text demonstration stage, it is determined that the learning stage corresponding to the typical learning work link of the demonstration stage is an introduction learning stage;

[0028] If the demonstration stage is a non-text demonstration stage, then in the corresponding demonstration stage, the dynamic coefficient of the corresponding demonstration stage is obtained according to the pixel changes of the picture at different demonstration moments; when the dynamic coefficient is greater than the preset dynamic threshold, it is judged that the corresponding demonstration stage is in the deep learning stage; when the dynamic coefficient is less than or equal to the preset dynamic threshold, it is judged that the corresponding demonstration stage is in the consolidation learning stage.

[0029] Furthermore, the method for obtaining the dynamic coefficient includes:

[0030] In each demonstration stage, between the teaching material demonstration pages at different demonstration moments, the sum of the absolute values ​​of the pixel differences between all the pixel points at the same position in all the pictures at the same position is used as the picture dynamic parameter corresponding to the different demonstration moments;

[0031] The dynamic parameters of the image between all different presentation moments are integrated to obtain the dynamic coefficient of the corresponding presentation stage.

[0032] Furthermore, the method for optimizing the writing of all the teaching material demonstration pages in the corresponding demonstration stage according to the different learning stages includes:

[0033] When the demonstration phase is in the introduction learning phase, all the texts in the teaching material demonstration page at each demonstration moment are taken as a text set, and the union of the text sets at all demonstration moments is taken as the text to be processed; semantic analysis and word frequency recognition are performed on the text to be processed, important concept texts are obtained, and special annotation processing is performed on the important concept texts;

[0034] When the demonstration stage is in the deep learning stage, a natural language description of each picture in the teaching material demonstration page at each demonstration moment is obtained based on a neural network model, and an interactive window is generated based on the natural language description result of each picture for dynamic interaction;

[0035] When the demonstration stage is in the consolidation learning stage, the natural language description of each video frame image corresponding to the teaching material demonstration page at each demonstration moment is obtained based on the neural network model, and an interactive window is generated based on the natural language description of each video frame image for dynamic interaction.

[0036] Furthermore, the preset learning stage segmentation time nodes are respectively located at the time corresponding to the 20% position and the time corresponding to the 70% position in the entire demonstration period of the digital teaching material to be processed.

[0037] The present invention also proposes a digital teaching material writing system based on action logic, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the digital teaching material writing method based on action logic.

[0038] The present invention has the following beneficial effects:

[0039] The present invention first obtains the teaching material demonstration page of the digital teaching material to be processed at each demonstration moment based on the action logic, and obtains all the texts and all the pictures in the teaching material demonstration page at each demonstration moment, so as to provide a basis for the subsequent analysis of the changes in the demonstration content of the teaching material demonstration page; according to the display dominance of the text and the picture in the teaching material demonstration page at each demonstration moment, combined with the corresponding demonstration moment, all the teaching material demonstration pages are segmented to obtain all the demonstration stages, so as to facilitate the subsequent analysis and evaluation according to the content expression form in each demonstration stage, and perform feature writing optimization after judging the learning link of each demonstration stage; in each demonstration stage, according to the display difference of the text and the picture in the teaching material demonstration page at all demonstration moments, combined with the pixel change of the picture at different demonstration moments, the learning link of each demonstration stage is judged; according to different learning links, all the teaching material demonstration pages in the corresponding demonstration stage are written and optimized to improve the content display effect of each learning stage, optimize the digital teaching materials according to the students' learning behavior logic, and improve the learning efficiency and effect. The present invention divides the demonstration stages by capturing the content changes of different teaching material demonstration pages and the demonstration moments, and identifies and judges the learning stages for each demonstration stage. It aims to optimize the presentation of specific content in different learning stages in specific learning typical work links of specific learning work tasks based on the action logic of student learning, and improve the writing effect of the digital teaching materials to be processed. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A flowchart of a method for compiling digital teaching materials based on action logic provided by one embodiment of the present invention;

[0042] Figure 2 A schematic diagram of a specific learning typical work link of a specific learning work task based on action logic provided by an embodiment of the present invention;

[0043] Figure 3 A flow chart of an acquisition method in the demonstration phase provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a digital teaching material writing method and system based on action logic proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0045] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0046] The following is a detailed description of a method and system for writing digital teaching materials based on action logic provided by the present invention in conjunction with the accompanying drawings.

[0047] See also Figure 1 , which shows a flow chart of a method for writing digital teaching materials based on action logic provided by an embodiment of the present invention, specifically comprising:

[0048] Step S1, obtaining the teaching material demonstration page of the digital teaching material to be processed at each demonstration moment based on the action logic, and obtaining all texts and all pictures in the teaching material demonstration page at each demonstration moment.

[0049] In order to optimize the writing of the digital teaching materials to be processed based on action logic, the embodiment of the present invention will first obtain the teaching material demonstration page at each demonstration moment, and then analyze and identify the learning stage of the specific learning typical work link of the specific learning work task in which each teaching material demonstration page is located, so as to perform special processing on each teaching material demonstration page, improve the content display effect of each learning stage, optimize the display of digital teaching materials according to the students' learning behavior logic, and improve learning efficiency and effect.

[0050] It should be noted that the digital teaching materials to be processed are digital teaching materials that have been preliminarily produced, and are mainly produced based on specific learning typical work links of specific learning work tasks based on action logic. Please refer to Figure 2 , which shows a schematic diagram of a specific learning typical work link of a specific learning work task based on action logic provided by an embodiment of the present invention; the learning objectives and learning methods targeted by the learning stage of each learning typical work link are different, so the subsequent special processing method for each teaching material demonstration page is also different.

[0051] In one embodiment of the present invention, the digital teaching materials to be processed are connected to a multimedia teaching platform for demonstration. The multimedia teaching platform can simultaneously support presentation forms such as PPT, PDF or HTML page forms, and can simultaneously support the learning participation of at least 30 user students; then during the demonstration process, each second is regarded as a demonstration moment, and the teaching material demonstration page at each demonstration moment is obtained, and all texts and all pictures in the teaching material demonstration page at each demonstration moment are obtained;

[0052] For different forms of digital teaching materials to be processed, implementers can use different Java language corresponding libraries for automatic extraction; for example, for PPT format, implementers can use Apache POI library for extraction; for PDF format, implementers can use Apache PDFBox or iText library; for HTML page format, it can be extracted based on JSoup and other libraries; all of them are existing technologies well known to technicians in this field and will not be elaborated here.

[0053] In other embodiments of the present invention, implementers may also combine screen reading technology, OCR technology, screenshots and professional capture software to obtain all text and all images in the teaching material demonstration page at each demonstration moment, which is already existing technology and will not be described in detail here.

[0054] Step S2, according to the display dominance of texts and pictures in the teaching material demonstration page at each demonstration moment, combined with the corresponding demonstration moment, all teaching material demonstration pages are segmented to obtain all demonstration stages.

[0055] Taking into account the different learning stages of the specific learning typical work links of specific learning tasks, the focus of the textbook demonstration page is different. For example, in the introductory learning stage, the digital textbooks to be processed tend to help students understand basic concepts, and the text content such as background information and definition introduction occupies a dominant position; in the deep learning stage, it is usually to realize the action guidance functions such as classroom organization, teacher-student teaching interaction, classroom summary and reflection through information technology, and guide students to gradually perform the six specific dimensions of learning actions such as collecting information, making plans, making decisions, specific implementation, checking details, and evaluating results through autonomous and interactive actions, so as to achieve the goals of knowledge learning, skill training, and reflection on learning effectiveness; in the consolidation learning stage, the digital textbooks to be processed usually use video demonstrations or case analysis to help students think deeply and consolidate their learning, conduct problem feedback and homework assignments, and prepare for the end of teaching;

[0056] At different demonstration moments of digital teaching materials in specific learning typical work links of specific learning work tasks, the learning stages of the teaching material demonstration pages are different, the expression forms in the digital teaching materials are different, and the special processing methods for each teaching material demonstration page are also different; therefore, the embodiment of the present invention will first divide all teaching material demonstration pages according to the display dominance of texts and pictures in the teaching material demonstration pages at each demonstration moment, combined with the corresponding demonstration moments, to obtain all demonstration stages; and then facilitate subsequent analysis and evaluation according to the content expression forms in each demonstration stage to determine the learning stage of each demonstration stage.

[0057] Preferably, in one embodiment of the present invention, the acquisition method in the demonstration phase includes:

[0058] See also Figure 3 , which shows a flow chart of an acquisition method in a demonstration phase provided by an embodiment of the present invention, specifically comprising:

[0059] Step S301, obtaining the display dominance coefficient of the teaching material demonstration page at each demonstration moment according to the total number of characters corresponding to the text and the total number of pixels corresponding to the picture in the teaching material demonstration page at each demonstration moment.

[0060] Taking into account the differences in the corresponding display contents of text and pictures in each textbook demonstration page of the specific learning typical work link of the specific learning work task, the display dominance of the current textbook demonstration page can be reflected. The display dominance reflects the content display situation of the current textbook demonstration page, which is convenient for the subsequent division of demonstration stages.

[0061] In a preferred embodiment of the present invention, it is considered that the text and pictures in the teaching material demonstration page are not displayed all at once, but are displayed step by step. For example, in the process of introducing the concept of turning shaft parts, the background of turning shaft parts is introduced first, and then the concept or definition of turning shaft parts is gradually introduced; then as the demonstration time changes, the display of text content is gradual; similarly, the display of pictures is also gradual; when the text or picture display content tends to be maximized, it means that it may be at the stage change junction of the demonstration stage; and considering that the closer the demonstration time is to the corresponding learning stage switching node in the learning typical work link, it means that it is more likely to be at the stage change junction of the demonstration stage;

[0062] Therefore, the method for obtaining the display dominance coefficient includes:

[0063] In the teaching material demonstration page at each demonstration time, the total number of characters corresponding to the text is used as the text display coefficient, and the ratio of the total number of pixels corresponding to the picture to the overall resolution of the teaching material demonstration page is used as the picture display coefficient;

[0064] The text display coefficient plus the first preset non-zero normal number is used as the numerator, the picture display coefficient plus the second preset non-zero normal number is used as the denominator, and the fraction ratio is used as the text display dominant parameter; the reciprocal of the text display dominant parameter is used as the picture display dominant parameter;

[0065] The maximum value of the text display dominant parameter and the image display dominant coefficient is normalized, and the normalized result is used as the display dominant coefficient at the corresponding demonstration moment.

[0066] As an example, the calculation formula for the dominance coefficient is shown as: ; Where t is the serial number of the demonstration moment; is the display dominance coefficient at the t-th demonstration moment; is the maximum value function; is the total number of characters corresponding to the text in the teaching material demonstration page at the tth demonstration time, also known as the text display coefficient; It is the ratio of the total number of pixels corresponding to the image in the teaching material demonstration page at the t-th demonstration moment to the overall resolution of the teaching material demonstration page, which is also the image display coefficient; is a first preset non-zero positive constant; is a second preset non-zero positive constant; is the standard normalization function.

[0067] In this example, linear normalization is specifically used for normalization; the first preset non-zero normal number is 1; the overall resolution of the multimedia demonstration interface of the teaching material demonstration page is 720p, and the second preset non-zero normal number is 0.0000001; the first preset non-zero normal number and the second preset non-zero normal number are both set to ensure that the denominator is not zero, and the implementer can also set the value or the normalization method by himself, which will not be elaborated here.

[0068] It should be noted that the method for obtaining the total number of characters in a text is also an existing technology well known to those skilled in the art, such as using regular expressions to obtain all Chinese characters and counting the number, or using other word counting tools, etc., which will not be repeated here; the method for obtaining the total number of pixels in a picture is also an existing technology, which will not be repeated here.

[0069] Step S302, obtaining the time segmentation coefficient of each demonstration moment according to the time interval between each demonstration moment and each preset learning stage segmentation time node in the digital teaching material to be processed.

[0070] Considering that the closer to the preset link split time node, the more likely it is to be at the stage change intersection of the demonstration stage; therefore, the time split coefficient can be evaluated according to the time interval between each demonstration moment and the learning stage split time node of each preset learning typical work link, so as to obtain the demonstration stage in the subsequent split.

[0071] Preferably, in one embodiment of the present invention, the method for obtaining the time slicing coefficient includes:

[0072] In the future time domain of each demonstration moment, the time interval between the demonstration moment and each preset learning stage segmentation time node is obtained respectively, and the negative correlation mapping result of the minimum time interval is used as the time segmentation coefficient corresponding to the demonstration moment.

[0073] Among them, the preset learning stage segmentation time nodes are respectively located at the time corresponding to the 20% position and the 70% position of the digital teaching materials to be processed in the entire demonstration period.

[0074] As an example, to facilitate understanding of the preset learning phase segmentation time nodes, assuming that the complete demonstration time or course time of the digital teaching material to be processed is set to 50 minutes, and the demonstration starts at 8:00 a.m., then the first preset learning phase segmentation time node of the learning typical work link is 8:10, and the second preset learning phase segmentation time node of the learning typical work link is 8:45. In other examples, the implementer can also set the preset learning phase segmentation time nodes according to the time allocation of the learning typical work link.

[0075] After obtaining all the preset learning stage segmentation time nodes of the typical learning work link, assuming that the current demonstration time is 8:15:00, then between 8:15:00 and 8:50:00, calculate the time interval between the current demonstration time and each preset learning stage segmentation time node of the typical learning work link, screen out a minimum time interval, add the minimum non-zero normal number 0.1 to the time interval, and then perform an inverse operation to perform a negative correlation mapping, and use the negative correlation mapping result as the time segmentation coefficient of the demonstration moment; the smaller the time interval, the larger the time segmentation coefficient.

[0076] In other examples, implementers may also adopt other negative correlation mapping methods, such as using the minimum time interval as x in an exponential function exp(-x) with the natural constant e as the base, which will not be described in detail here.

[0077] Step S303, at each demonstration moment, according to the time segmentation coefficient and the display dominance coefficient of the corresponding teaching material demonstration page, obtain the segmentation possibility of each demonstration moment; the time segmentation coefficient and the display dominance coefficient are both positively correlated with the segmentation possibility.

[0078] As an example, at each demonstration moment, the time segmentation coefficient is multiplied and combined with the display dominance coefficient of the corresponding teaching material demonstration page, and the product is used as the segmentation possibility at the corresponding demonstration moment. In other examples, the implementer may also combine the two by other basic mathematical operations such as addition or weighted combination or related mapping means, which will not be described in detail here.

[0079] Step S304: taking the presentation moment at which the segmentation possibility is greater than a preset threshold as the segmentation moment, and acquiring the presentation stage according to adjacent segmentation moments.

[0080] As an example, the preset threshold is 0.95, and the demonstration moment when the segmentation possibility is greater than 0.95 is used as the segmentation moment; the implementer can also set the preset threshold by himself.

[0081] In a preferred embodiment of the present invention, the teaching material demonstration page corresponding to the first demonstration moment is used as the stage starting page of the first demonstration stage, the first segmentation moment is obtained by traversing along the time sequence of the demonstration moments, and the teaching material demonstration page corresponding to the first segmentation moment is used as the stage ending page of the first demonstration stage;

[0082] The teaching material demonstration page corresponding to the next adjacent demonstration moment of the first segmentation moment is used as the stage starting page of the second demonstration stage, and the second segmentation moment is obtained by traversing along the time sequence of the demonstration moments, and the teaching material demonstration page corresponding to the second segmentation moment is used as the stage ending page of the second demonstration stage;

[0083] The teaching material demonstration page corresponding to the next adjacent demonstration moment of the second segmentation moment is used as the starting point page of the third demonstration stage, and the iterative traversal is continuously performed to obtain all demonstration stages.

[0084] At this point, all the teaching material demonstration pages of the entire mathematics teaching material to be processed are divided into different demonstration stages, so that the corresponding learning stage can be judged according to the content in each demonstration stage, and different special processing can be performed.

[0085] Step S3, in each demonstration stage, according to the display differences of texts and pictures in the teaching material demonstration page at all demonstration moments, combined with the pixel changes of the pictures at different demonstration moments, the learning stage of the typical learning work link of each demonstration stage is determined.

[0086] After obtaining all demonstration stages, the learning stage of the typical learning work link of each demonstration stage can be analyzed according to the specific demonstration content in each demonstration stage; therefore, the embodiment of the present invention makes a judgment based on the display differences of text and pictures in the teaching material demonstration page at all demonstration moments in each demonstration stage, combined with the pixel changes of the pictures at adjacent demonstration moments.

[0087] In a preferred embodiment of the present invention, the learning stage of the learning typical work link includes at least an introduction learning stage, a deep learning stage and a consolidation learning stage; the method for determining the learning stage of the learning typical work link in each demonstration stage includes:

[0088] In each demonstration stage, according to the dominant parameters of the text display in the teaching material demonstration page at all demonstration moments, it is determined whether the demonstration stage is a text demonstration stage; if the demonstration stage is a text demonstration stage, the learning stage of the learning typical work link corresponding to the demonstration stage is determined to be the introduction learning stage;

[0089] If the demonstration stage is a non-text demonstration stage, then in the corresponding demonstration stage, the dynamic coefficient of the corresponding demonstration stage is obtained according to the pixel changes of the picture at different demonstration moments; when the dynamic coefficient is greater than the preset dynamic threshold, it is judged that the corresponding demonstration stage is in the deep learning stage; when the dynamic coefficient is less than or equal to the preset dynamic threshold, it is judged that the corresponding demonstration stage is in the consolidation learning stage.

[0090] As an example, first, in each demonstration stage, the dominant parameters of text display in the teaching material demonstration page at all demonstration moments are averaged, and then the mean is linearly normalized to obtain the text coefficient of each demonstration stage; when the text coefficient of the demonstration stage is greater than a preset coefficient threshold, such as 0.75, the demonstration stage is regarded as the text demonstration stage, and the learning stage of the typical learning work link is determined to be the introduction learning stage;

[0091] If the text coefficient of the demonstration stage is less than or equal to the preset coefficient threshold of 0.75, the demonstration stage is regarded as a non-text demonstration stage. The non-text demonstration stage indicates that the corresponding display content may be a picture or a video, and the difference between the video and the picture is that the video is dynamically changing, and the picture pixels at different demonstration moments are changing; the dynamic coefficient is obtained by the pixel changes in the picture at all different demonstration moments; when the dynamic coefficient is less than the preset dynamic threshold such as 0.1, it means that the picture is more likely to be in a silent state, and the corresponding demonstration stage is judged to be in the deep learning stage; otherwise, it means that the picture is more likely to be the corresponding video frame at different demonstration moments, and the corresponding demonstration stage is judged to be in the consolidation learning stage.

[0092] Among them, in a preferred embodiment of the present invention, the method for obtaining the dynamic coefficient includes:

[0093] In each demonstration stage, between the teaching material demonstration pages at different demonstration moments, the sum of the absolute values ​​of the pixel differences between all the pixel points at the same position in the pictures at the same position is used as the picture dynamic parameters corresponding to the different demonstration moments; the picture dynamic parameters between all the different demonstration moments are combined to obtain the dynamic coefficient of the corresponding demonstration stage.

[0094] As an example, the dynamic parameters of the images between all different presentation moments are accumulated and linearly normalized, and the normalized result is used as the dynamic coefficient of the corresponding presentation stage.

[0095] In other examples, implementers may also set the preset coefficient threshold, preset dynamic threshold and normalization method by themselves, which will not be elaborated here.

[0096] Step S4, optimizing the writing of all teaching material demonstration pages in the corresponding demonstration stage according to different learning stages.

[0097] After determining which learning stage each demonstration stage in the digital teaching materials to be processed corresponds to, targeted editing and optimization can be performed on all the teaching material demonstration pages.

[0098] Preferably, in one embodiment of the present invention, the method for optimizing the writing of all teaching material demonstration pages in the corresponding demonstration stage according to different learning stages includes:

[0099] When the demonstration phase is in the introduction learning phase of a specific learning typical work link of a specific learning work task, all texts in the teaching material demonstration page at each demonstration moment are taken as a text set, and the union of the text sets at all demonstration moments is taken as the text to be processed; semantic analysis and word frequency recognition are performed on the text to be processed, important concept texts are obtained, and special annotation processing is performed on the important concept texts;

[0100] When the demonstration phase is in the deep learning phase of a specific learning typical work link of a specific learning work task, the natural language description of each picture in the teaching material demonstration page at each demonstration moment is obtained based on the neural network model, and an interactive window is generated based on the natural language description result of each picture for dynamic interaction;

[0101] When the demonstration stage is in the consolidation learning stage of a specific learning typical work link of a specific learning task, the natural language description of each video frame corresponding to the teaching material demonstration page at each demonstration moment is obtained based on the neural network model, and an interactive window is generated based on the natural language description of each video frame for dynamic interaction.

[0102] As an example, when the demonstration phase is in the import learning phase, the acquired text to be processed is first preprocessed by jieba word segmentation and stop word removal; then the TF-IDF value of each word is calculated, which represents the word frequency of each word. The higher the word frequency, the higher its corresponding importance; then semantic understanding is performed based on the BERT model to identify named entities or keywords in the text to be processed; all word segments are sorted in descending order according to the TF-IDF value, and the word segments with higher word frequency, such as the top 20% of the word segmentation in the TF-IDF value ranking, are selected to construct the first word segmentation set; the second word segmentation set is constructed based on the word segmentation corresponding to all keywords acquired by the BERT model; all the word segments in the intersection of the first word segmentation set and the second word segmentation set are taken as important concept texts, and the important concept texts are highlighted and bolded; for example, when introducing the concept of "turning shaft parts", the "definition of turning shaft parts" can be bolded or marked with different colors;

[0103] When the demonstration stage is in the deep learning stage of the specific learning typical work link of the specific learning work task, the recurrent neural network model is used to obtain the natural language description of each picture in the teaching material demonstration page at each demonstration moment, wherein the natural language description is used to explain the content in the picture; and a graphical user interface GUI framework is used to generate an interactive window, which is used to dynamically and interactively display the natural language description results of each picture, so that students can understand it in combination with the picture; for example, when explaining the turning of shaft parts, a dynamic interactive description window in the turning shaft parts process or the corresponding display picture of the internship training is provided, and the user can click or hover on certain pictures to pop up detailed information or annotations of the picture to provide additional explanations;

[0104] When the demonstration stage is in the consolidation learning stage of a specific learning typical work link of a specific learning work task, each video frame image corresponding to the teaching material demonstration page at each demonstration moment is processed based on the same processing method of the images in the deep learning stage of the specific learning typical work link of the specific learning work task, and the natural language description of each video frame image corresponding to the teaching material demonstration page at each demonstration moment is obtained, and then dynamic interaction is performed to provide a text description of the current video frame demonstration content during video playback; for example, when demonstrating a dynamic video of the process of turning shaft parts, a dynamic interactive description window of the display content such as shaft parts being turned in the corresponding video frame is provided at each processing stage, and real-time annotations are provided to assist learning and understanding.

[0105] It should be noted that the processing means for writing and optimizing all teaching material demonstration pages in the corresponding demonstration stage at different learning stages of specific learning typical work links of specific learning work tasks are all existing technologies well known to technical personnel in this field. Implementers can also set other writing optimization means on their own to perform writing optimization for different demonstration stages.

[0106] In another embodiment of the present invention, the deep learning stage may also include six action dimensions: collecting information, formulating plans, making decisions, specific implementation, checking details, and evaluating results; in different action dimensions, the issuance of information sheets, plan sheets, decision sheets, implementation sheets, check sheets, and evaluation sheets can be automatically triggered, and students can be asked and answered and interacted with in text, voice, video, graphics, etc. in a timely manner through the computer system; in the deep learning stage, if the demonstration stage is an information, intelligence, and data collection behavior, then the corresponding demonstration stage is judged to be in the information collection dimension; if the demonstration stage is an information, intelligence, and data sorting behavior, then the corresponding demonstration stage is judged to be in the plan formulation dimension; if the demonstration stage is a plan comparison behavior, then the corresponding demonstration stage is judged to be in the decision-making dimension; if the demonstration stage is an information, intelligence, and data action behavior, then the corresponding demonstration stage is judged to be in the specific implementation dimension; if the demonstration stage is an action of digital quantification of the action amplitude, then the corresponding demonstration stage is judged to be in the detail inspection dimension; if the demonstration stage is a comparison of action results, then the corresponding demonstration stage is judged to be in the result evaluation dimension.

[0107] At this point, after all the teaching material demonstration pages are optimized, the optimized digital teaching materials can be obtained.

[0108] In summary, the present invention first obtains the teaching material demonstration page of the digital teaching material to be processed at each demonstration moment based on the action logic, and all the texts and all the pictures in each teaching material demonstration page; then all the teaching material demonstration pages are segmented to obtain all the demonstration stages; then the learning stage of the typical learning work link in each demonstration stage is determined; finally, all the teaching material demonstration pages in the corresponding demonstration stage are written and optimized according to different learning stages. The present invention divides the demonstration stage by the content changes of different teaching material demonstration pages and the capture of the demonstration moment, and identifies and judges the learning stage for each demonstration stage, aiming to optimize the presentation of specific content in different learning stages based on the typical learning work link of the action logic of student learning, and improve the writing effect of the digital teaching material to be processed.

[0109] The present invention also proposes a digital teaching material writing system based on action logic. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, steps S1-S4 of a digital teaching material writing method based on action logic are implemented.

[0110] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0111] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for compiling digital teaching materials based on action logic, characterized in that: The method comprises: Obtain the teaching material demonstration page of the digital teaching material to be processed at each demonstration moment based on the action logic, and obtain all texts and all pictures on the teaching material demonstration page at each demonstration moment; According to the display dominance of the text and the picture in the teaching material demonstration page at each demonstration moment, combined with the corresponding demonstration moment, all the teaching material demonstration pages are segmented to obtain all the demonstration stages; In each demonstration stage, the learning stage of the typical learning work link of each demonstration stage is determined based on the display differences of the text and pictures in the teaching material demonstration page at all demonstration moments and the pixel changes of the pictures at different demonstration moments; According to the different learning stages, all the teaching material demonstration pages in the corresponding demonstration stage are optimized; the acquisition method of the demonstration stage includes: According to the total number of characters corresponding to the text and the total number of pixels corresponding to the picture in the teaching material demonstration page at each demonstration moment, the display dominance coefficient of the teaching material demonstration page at each demonstration moment is obtained; Obtaining a time segmentation coefficient for each demonstration moment according to the time interval between each demonstration moment and each preset learning stage segmentation time node in the digital teaching material to be processed; At each demonstration moment, the segmentation possibility of each demonstration moment is obtained according to the time segmentation coefficient and the display dominance coefficient corresponding to the teaching material demonstration page; the time segmentation coefficient and the display dominance coefficient are both positively correlated with the segmentation possibility; Taking the presentation moment at which the segmentation possibility is greater than a preset threshold as the segmentation moment, and acquiring the presentation stage according to the adjacent segmentation moments; The method for obtaining the display dominance coefficient includes: In the teaching material demonstration page at each demonstration moment, the total number of characters corresponding to the text is used as the text display coefficient, and the ratio of the total number of pixels corresponding to the picture to the overall resolution of the teaching material demonstration page is used as the picture display coefficient; The text display coefficient plus a first preset non-zero normal number is used as a numerator, the picture display coefficient plus a second preset non-zero normal number is used as a denominator, and the fraction ratio is used as a text display dominant parameter; the reciprocal of the text display dominant parameter is used as a picture display dominant parameter; The maximum value of the text display dominant parameter and the picture display dominant coefficient is normalized, and the normalized result is used as the display dominant coefficient at the corresponding demonstration moment.

2. According to the method of writing digital teaching materials based on action logic in claim 1, it is characterized in that: The method for obtaining the time division coefficient includes: In the future time domain of each demonstration moment, the time interval between the demonstration moment and each preset learning stage segmentation time node is obtained respectively, and the negative correlation mapping result of the minimum time interval is used as the time segmentation coefficient corresponding to the demonstration moment.

3. According to the method of writing digital teaching materials based on action logic in claim 1, it is characterized in that: The method for acquiring the demonstration stage according to the adjacent segmentation moments comprises: The teaching material demonstration page corresponding to the first demonstration moment is used as the stage starting page of the first demonstration stage, and the first segmentation moment is obtained by traversing along the time sequence of the demonstration moments, and the teaching material demonstration page corresponding to the first segmentation moment is used as the stage ending page of the first demonstration stage; The teaching material demonstration page corresponding to the next adjacent demonstration moment of the first segmentation moment is used as the stage starting page of the second demonstration stage, and the second segmentation moment is obtained by traversing along the time sequence of the demonstration moments, and the teaching material demonstration page corresponding to the second segmentation moment is used as the stage ending page of the second demonstration stage; The teaching material demonstration page corresponding to the next adjacent demonstration moment of the second segmentation moment is used as the starting point page of the third demonstration stage, and the iterative traversal is continuously performed to obtain all demonstration stages.

4. The method for compiling digital teaching materials based on action logic according to claim 1 is characterized in that: The method for determining the learning stage of the learning typical work link in each demonstration stage includes: The learning stages of the typical learning work links at least include an introduction learning stage, an in-depth learning stage, and a consolidation learning stage; In each demonstration stage, according to the dominant parameters of the text display in the teaching material demonstration page at all demonstration moments, it is determined whether the demonstration stage is a text demonstration stage; if the demonstration stage is a text demonstration stage, it is determined that the learning stage corresponding to the typical learning work link of the demonstration stage is an introduction learning stage; If the demonstration stage is a non-text demonstration stage, then in the corresponding demonstration stage, the dynamic coefficient of the corresponding demonstration stage is obtained according to the pixel changes of the picture at different demonstration moments; when the dynamic coefficient is greater than the preset dynamic threshold, it is judged that the corresponding demonstration stage is in the deep learning stage; when the dynamic coefficient is less than or equal to the preset dynamic threshold, it is judged that the corresponding demonstration stage is in the consolidation learning stage.

5. A method for compiling digital teaching materials based on action logic according to claim 4, characterized in that: The method for obtaining the dynamic coefficient includes: In each demonstration stage, between the teaching material demonstration pages at different demonstration moments, the sum of the absolute values ​​of the pixel differences between all the pixel points at the same position in all the pictures at the same position is used as the picture dynamic parameter corresponding to the different demonstration moments; The dynamic parameters of the image between all different presentation moments are integrated to obtain the dynamic coefficient of the corresponding presentation stage.

6. A method for compiling digital teaching materials based on action logic according to claim 4, characterized in that: The method for optimizing the writing of all the teaching material demonstration pages in the corresponding demonstration stage according to the different learning stages includes: When the demonstration phase is in the introduction learning phase, all the texts in the teaching material demonstration page at each demonstration moment are taken as a text set, and the union of the text sets at all demonstration moments is taken as the text to be processed; semantic analysis and word frequency recognition are performed on the text to be processed, important concept texts are obtained, and special annotation processing is performed on the important concept texts; When the demonstration stage is in the deep learning stage, a natural language description of each picture in the teaching material demonstration page at each demonstration moment is obtained based on a neural network model, and an interactive window is generated based on the natural language description result of each picture for dynamic interaction; When the demonstration stage is in the consolidation learning stage, the natural language description of each video frame image corresponding to the teaching material demonstration page at each demonstration moment is obtained based on the neural network model, and an interactive window is generated based on the natural language description of each video frame image for dynamic interaction.

7. The method for compiling digital teaching materials based on action logic according to claim 1 is characterized in that: The preset learning stage segmentation time nodes are respectively located at the time corresponding to the 20% position and the time corresponding to the 70% position of the digital teaching material to be processed in the entire demonstration period.

8. A digital teaching material writing system based on action logic, characterized in that: The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a method for writing digital teaching materials based on action logic as described in any one of claims 1 to 7 are implemented.

Images