Course resource intelligent generation system and method

Abstract

The invention discloses an intelligent course resource generation system and method.The intelligent course resource generation system comprises a demand perception module, a knowledge fusion engine, an intelligent generation core and a quality optimization ring, and the demand perception module is used for analyzing user portraits and course standard metadata; and the knowledge fusion engine is connected with the multi-source knowledge base and constructs a cross-modal knowledge graph. According to the method, personalized demands of learners are accurately adapted, and resource content depth and presentation forms are dynamically adjusted based on a deep cognitive model; the bottleneck of multi-modal teaching resource collaboration is broken through, and a teaching logic closed loop is formed by text elaboration, visual demonstration and interactive operation by means of a semantic-level cross-modal alignment technology; the cognitive overload risk is intelligently avoided, and an abstract concept is converted into a stepped cognitive path through an embedded load monitoring mechanism; and finally reconstructing an educational resource generation normal form, converting teaching experience into an iterable generation strategy, and forming an intelligent educational ecology with an adaptive evolution characteristic.

Description

Technical Field

[0001] This invention belongs to the field of course resource generation technology, specifically relating to an intelligent course resource generation system and method. Background Technology

[0002] Curriculum resources are the cornerstone of high-quality education development. Their construction needs to take into account both "systematicity" (classification and integration) and "dynamics" (technological iteration). Through policy guidance, technological innovation and multi-party collaboration, resources can achieve a leap from "quantity" to "quality", thus contributing to the achievement of educational equity and lifelong learning goals.

[0003] The current development of course resources has the following problems: 1. Low efficiency: Manually creating a single lesson resource takes a long time on average; 2. Poor adaptability: Standardized courseware cannot meet the needs of differentiated instruction, and the coverage rate of special education groups is low; 3. Limited modality: Traditional tools only support text / image separation generation and lack dynamic interactive capabilities.

[0004] Therefore, it is necessary to provide a system and method for intelligent generation of course resources to address the aforementioned technical issues.

[0005] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent course resource generation system and method that can solve three core problems in educational resource generation: personalized adaptation, multimodal collaboration, and cognitive load control.

[0007] To achieve the above objectives, a specific embodiment of the present invention provides the following technical solution: The intelligent course resource generation system includes a demand perception module, a knowledge fusion engine, an intelligent generation core, and a quality optimization loop. The demand perception module is used to parse user profiles and course standard metadata. The knowledge fusion engine connects to multi-source knowledge bases and constructs a cross-modal knowledge graph. The intelligent generation core includes a text generation unit, a visual generation unit, and an interaction design unit. The quality optimization loop dynamically adjusts the output resources based on an educational effectiveness evaluation model.

[0008] In one or more embodiments of the present invention, the demand-aware module includes: The user profiling unit extracts cognitive level and subject preference characteristics from historical learning data; The curriculum standard analysis unit uses NLP entity recognition technology to extract knowledge nodes from the teaching syllabus.

[0009] In one or more embodiments of the present invention, the knowledge fusion engine includes: A unified encoder for heterogeneous data sources, based on the Transformer-XL architecture, to vectorize textbook / paper / video data; A graph neural network relation inferencer constructs a semantic association graph of text, image, and video entities.

[0010] In one or more embodiments of the present invention, the intelligent generation core includes: The text generation unit uses the GPT-4 architecture to generate lesson plans and test questions. The visual generation unit generates principle diagrams and 3D animations based on the Stable Diffusion model; The interaction design unit automatically builds AR / VR experimental scenarios and integrates H5 interactive controls.

[0011] In one or more embodiments of the present invention, the quality optimization loop includes: Cognitive load monitor calculates resource complexity metrics in real time; The A / B test feedback unit triggers resource restructuring based on student answer data.

[0012] In one or more embodiments of the present invention, the operating logic of the cognitive load monitor includes: When the resource information density is detected to exceed a preset threshold, the simplified reconstruction module is automatically triggered. Based on the Bayesian knowledge tracing algorithm, students' common mistakes are identified, which drives the generation of enhanced practice resources.

[0013] In one or more embodiments of the present invention, a cross-modal alignment unit is provided between the knowledge fusion engine and the intelligent generation core. This unit achieves semantic consistency control between text descriptions and visual elements through a multi-head attention mechanism.

[0014] In one or more embodiments of the present invention, the generation system further includes a blockchain evidence storage unit for storing key parameters of the generated resources on the blockchain and generating an immutable copyright certificate.

[0015] The intelligent generation method for course resources includes the following steps: S1. Receive the learning objectives and constraints input by the user; S2. Construct a cross-modal knowledge graph through multi-source knowledge extraction; S3: Parallel startup of text resource generation, visual resource synthesis, and interactive scene construction; S4. Dynamically optimize resources based on cognitive load assessment results.

[0016] In one or more embodiments of the present invention, the dynamic optimization step specifically includes: Calculate resource cognitive load using an education effectiveness assessment model; When the load value exceeds the threshold, perform one of the following operations: a) Break down complex knowledge points into tiered learning segments b) Replace abstract descriptions with visual examples c) Insert interactive cognitive scaffolding control.

[0017] Compared with existing technologies, this invention accurately adapts to learners' personalized needs, dynamically adjusts the depth and presentation of resource content based on a deep cognitive model; breaks through the bottleneck of multimodal teaching resource collaboration, and uses semantic-level cross-modal alignment technology to form a closed loop of teaching logic through textual explanation, visual demonstration, and interactive operation; intelligently avoids the risk of cognitive overload by transforming abstract concepts into a step-by-step cognitive path through an embedded load monitoring mechanism; and finally reconstructs the paradigm of educational resource generation, transforming teaching experience into iterative generation strategies to form an intelligent education ecosystem with adaptive evolutionary characteristics. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of a course resource intelligent generation system according to an embodiment of the present invention; Figure 2 This is a flowchart of an embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions in this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this disclosure.

[0021] like Figure 1As shown in one embodiment of the present invention, the intelligent course resource generation system includes a demand perception module, a knowledge fusion engine, an intelligent generation core, and a quality optimization loop. The demand perception module is used to parse user profiles and course standard metadata. The knowledge fusion engine connects to multi-source knowledge bases and constructs a cross-modal knowledge graph. The intelligent generation core includes a text generation unit, a visual generation unit, and an interaction design unit. The quality optimization loop dynamically adjusts the output resources based on the education effectiveness evaluation model.

[0022] The intelligent resource generation system for this course utilizes several key principles. First, the demand perception module analyzes user profiles, mimicking teacher experience to construct a cognitive model. This transforms historical learning trajectories into a heatmap of knowledge mastery. Second, the curriculum standard analysis employs Semantic Web technology to automatically establish a tree-like structure linking teaching objectives and knowledge points. Third, the knowledge fusion engine constructs a cross-modal knowledge graph by introducing a semantic bridging mechanism. Textual concepts are automatically associated with keyframes in experimental videos, and the explanatory power weights of visual elements are calculated using three-dimensional spatial vectors. For conflicting knowledge, an authoritative source arbitration strategy is initiated, prioritizing frequently cited academic expressions. The core of intelligent generation, based on text generation, uses an educational context constraint algorithm to embed a teaching terminology filter in the GPT model output layer, ensuring that the generated content conforms to subject-specific expression standards. Fourth, visual generation incorporates a teaching logic controller: based on the causal chain described in the text, it automatically plans the sequence of animation keyframes and the path of visual focus movement. Fifth, the quality optimization loop, based on a cognitive load monitoring cognitive flow model, analyzes the information density and cognitive path complexity of resources in real time. When a knowledge gap is detected, a transitional explanation module is automatically inserted. Sixth, feedback optimization employs an error pattern feedback mechanism: extracting knowledge misunderstanding patterns from student errors to drive the generation of targeted corrective cases.

[0023] Specifically, the demand perception module includes a user profiling unit and a curriculum standard analysis unit. The user profiling unit extracts cognitive level and subject preference characteristics from historical learning data; the curriculum standard analysis unit uses NLP entity recognition technology to extract knowledge nodes from the teaching syllabus.

[0024] Specifically, the knowledge fusion engine includes a heterogeneous data source unified encoder and a graph neural network relation inferencer. The heterogeneous data source unified encoder vectorizes textbook / paper / video data based on the Transformer-XL architecture; the graph neural network relation inferencer constructs a semantic association graph of text-image-video entities.

[0025] Specifically, the intelligent generation core includes a text generation unit, a visual generation unit, and an interaction design unit. The text generation unit uses the GPT-4 architecture to generate lesson plans and test questions; the visual generation unit generates principle diagrams and 3D animations based on the Stable Diffusion model; and the interaction design unit automatically constructs AR / VR experimental scenarios and integrates H5 interactive controls.

[0026] Specifically, the quality optimization loop includes a cognitive load monitor and an A / B test feedback unit. The cognitive load monitor calculates resource complexity indicators in real time, while the A / B test feedback unit triggers resource reconfiguration based on student answer data.

[0027] Meanwhile, the cognitive load monitor's operating logic includes: automatically triggering the simplified reconstruction module when the resource information density exceeds a preset threshold; and using the Bayesian knowledge tracing algorithm to locate students' common mistakes and drive the generation of enhanced exercise resources.

[0028] Preferably, a cross-modal alignment unit is provided between the knowledge fusion engine and the intelligent generation core. This unit achieves semantic consistency control between text descriptions and visual elements through a multi-head attention mechanism.

[0029] Furthermore, the generation system also includes a blockchain evidence storage unit, which is used to store the key parameters of the generated resources on the blockchain and generate an immutable copyright certificate.

[0030] Example: like Figure 2 As shown, the example of generating the "Advanced 3Ds Max Polygon Modeling" course is... 1. Requirements Analysis Phase: The system identifies common "bottlenecks in surface modeling" among students, such as the upturned eaves of pavilion roofs, and automatically activates the basic module of "Decomposition of Ancient Building Components". Extracting the collaborative design relationship of "topography-building-vegetation" from the landscape design curriculum standards, and setting three-tiered knowledge anchor points: Basics: Polygonal terrain shaping; Advanced: Parametric modeling of architectural features; Advanced: Creating the atmosphere through lighting and shadows in the scene.

[0031] 2. Knowledge Integration Stage: 1) Link to classical garden database: Extracting patterns from Suzhou garden roof tiles as texture material library; Match the CAD structural drawing and the actual photograph of the "hipped roof pavilion".

[0032] 2) Constructing a cross-modal knowledge graph 3. Resource Generation Stage: 1) Text Unit Generate tiered lesson plans: Basics: Create the stone bench base using polygon extrusion; The scanning tool generates garden fence vines.

[0033] Advanced Section: Surface optimization: Transform the cube into a Taihu stone using turbo smoothing; Parametric components: Array generation of pavilion mortise and tenon structure.

[0034] 2) Visual unit 3D animation demonstration: Dynamic disassembly of pavilion roof modeling process: planar polygon → add bend modifier → edge segmentation optimization → tile array distribution.

[0035] The gradient semi-transparent material demonstrates the refraction effect of the water feature glass curtain wall.

[0036] 3) Interactive Unit AR hands-on sand table: Adjust the intensity of the artificial rockery displacement map by dragging and dropping the image. The density of the bridge railings can be changed in real time by sliding the parameter bar.

[0037] 4. Dynamic optimization stage: When a high error rate is detected in the "Terrain Shaping" operation, a comparison demonstration will be automatically inserted: To address the issue of "distorted material texture," a physical rendering parameter comparison table is automatically generated.

[0038] The intelligent generation method for course resources in one embodiment of the present invention includes the following steps: S1. Receive the learning objectives and constraints input by the user; S2. Construct a cross-modal knowledge graph through multi-source knowledge extraction; S3: Parallel startup of text resource generation, visual resource synthesis, and interactive scene construction; S4. Dynamically optimize resources based on cognitive load assessment results.

[0039] The dynamic optimization steps specifically include: Calculate resource cognitive load using an education effectiveness assessment model; When the load value exceeds the threshold, perform one of the following operations: a) Break down complex knowledge points into tiered learning segments; b) Replace the abstract description with a visual case; c) Insert interactive cognitive scaffolding control.

[0040] It will be apparent to those skilled in the art that this disclosure is not limited to the details of the exemplary embodiments described above, and that this disclosure can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of this disclosure is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within this disclosure. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0041] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A course resource intelligent generation system, characterized in that, include: The demand awareness module is used to parse user profiles and course standard metadata; A knowledge fusion engine connects multiple knowledge bases and constructs a cross-modal knowledge graph; The intelligent generation core includes a text generation unit, a visual generation unit, and an interaction design unit. The quality optimization loop dynamically adjusts output resources based on the education effectiveness evaluation model.

2. The intelligent course resource generation system according to claim 1, characterized in that, The demand perception module includes: The user profiling unit extracts cognitive level and subject preference characteristics from historical learning data; The curriculum standard analysis unit uses NLP entity recognition technology to extract knowledge nodes from the teaching syllabus.

3. The intelligent course resource generation system according to claim 1, characterized in that, The knowledge fusion engine includes: A unified encoder for heterogeneous data sources, based on the Transformer-XL architecture, to vectorize textbook / paper / video data; A graph neural network relation inferencer constructs a semantic association graph of text, image, and video entities.

4. The intelligent course resource generation system according to claim 1, characterized in that, The intelligent generation core includes: The text generation unit uses the GPT-4 architecture to generate lesson plans and test questions. The visual generation unit generates principle diagrams and 3D animations based on the Stable Diffusion model; The interaction design unit automatically builds AR / VR experimental scenarios and integrates H5 interactive controls.

5. The intelligent course resource generation system according to claim 1, characterized in that, The quality optimization loop includes: Cognitive load monitor calculates resource complexity metrics in real time; The A / B test feedback unit triggers resource restructuring based on student answer data.

6. The intelligent course resource generation system according to claim 5, characterized in that, The operating logic of the cognitive load monitor includes: When the resource information density is detected to exceed a preset threshold, the simplified reconstruction module is automatically triggered. Based on the Bayesian knowledge tracing algorithm, students' common mistakes are identified, which drives the generation of enhanced practice resources.

7. The intelligent course resource generation system according to claim 1, characterized in that, A cross-modal alignment unit is provided between the knowledge fusion engine and the intelligent generation core. This unit achieves semantic consistency control between text descriptions and visual elements through a multi-head attention mechanism.

8. The intelligent course resource generation system according to claim 1, characterized in that, The generation system also includes a blockchain evidence storage unit, which is used to store key parameters of the generated resources on the blockchain and generate an immutable copyright certificate.

9. A method for intelligently generating course resources, used in the intelligent course resource generation system as described in any one of claims 1 to 8, characterized in that, Includes the following steps: S1. Receive the learning objectives and constraints input by the user; S2. Construct a cross-modal knowledge graph through multi-source knowledge extraction; S3: Parallel startup of text resource generation, visual resource synthesis, and interactive scene construction; S4. Dynamically optimize resources based on cognitive load assessment results.

10. The intelligent generation method for course resources according to claim 9, characterized in that, The dynamic optimization steps specifically include: Calculate resource cognitive load using an education effectiveness assessment model; When the load value exceeds the threshold, perform one of the following operations: a) Break down complex knowledge points into tiered learning segments b) Replace abstract descriptions with visual examples c) Insert interactive cognitive scaffolding control.

Images