An ideological and political individualized teaching system based on AI intelligent recommendation

Abstract

This invention discloses an AI-based personalized ideological and political education teaching system, specifically in the field of personalized teaching demonstrations, designed to address the problem that existing ideological and political education classrooms, in large-class settings, struggle to cater to students from different majors. The system constructs a logical display structure for the classroom, comprising a main lecturer demonstration layer and a major-specific adaptation demonstration layer. It generates a case structure tree containing core demonstration nodes and major-specific adaptation sub-nodes. Using a graph neural network model, it calculates the matching relationship between major-specific adaptation sub-nodes and student course system tags, generating corresponding adaptation content sequences for different major groups. Based on a unified teaching timeline, the system generates a collaborative control instruction set, driving the main lecturer demonstration layer and each major-specific adaptation demonstration layer to synchronously present case content during classroom teaching. This achieves differentiated teaching demonstrations across multiple majors while ensuring consistency in the main teaching line, thereby improving the effectiveness of ideological and political education classroom teaching and student participation.

Description

Technical Field

[0001] This invention relates to the field of personalized teaching demonstration technology, and more specifically, to a personalized ideological and political education teaching system based on AI intelligent recommendation. Background Technology

[0002] As the status of ideological and political education courses in the talent cultivation system of higher education institutions continues to rise, large-class and interdisciplinary blended teaching have become common organizational forms for ideological and political education. In actual teaching, the same classroom often contains students from different professional backgrounds such as engineering, economics and management, and literature and history. Students from different majors have significant differences in knowledge structure, curriculum system, and focus.

[0003] However, current ideological and political education classrooms often employ standardized courseware and content presentation. Lecturers typically present teaching cases and materials synchronously to all students via a single display terminal, making it difficult to differentiate presentations based on the cognitive characteristics of students from different majors. This results in some students having insufficient understanding of the case content, limiting their participation and the overall teaching effectiveness. While some teaching systems have introduced resource recommendations or after-class learning push mechanisms, these technologies are mostly focused on pre-class preparation or post-class support, failing to address the issue of adapting real-time presentations to a diverse student population. Especially in ideological and political education that emphasizes value guidance, case analysis, and contextual integration, there remains a lack of effective and systematic solutions for coordinating the main lecture content with the supplementary explanations, background materials, and presentation formats needed by students from different majors within the same teaching timeline.

[0004] Therefore, it is necessary to propose a personalized teaching technology solution for ideological and political education that can combine classroom space structure with multi-display terminal collaborative control, so as to realize differentiated teaching demonstrations for different majors in the same classroom environment and improve the overall teaching quality and teaching relevance of ideological and political education classes. Summary of the Invention

[0005] In order to overcome the above-mentioned deficiencies of the prior art, embodiments of the present invention provide an AI-based intelligent recommendation-based personalized ideological and political education system to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A personalized ideological and political education teaching system based on AI intelligent recommendation includes:

[0008] The display logic module is used to construct a classroom logical display structure based on the professional attributes in the objective profile data of students in the class. It includes a lecturer demonstration layer and a professional adaptation demonstration layer, with each demonstration layer associated with an independent display terminal.

[0009] The case analysis module is used to analyze and extract core teaching knowledge points from ideological and political teaching cases, map each core teaching knowledge point to a core demonstration node, and generate corresponding professional-adaptive sub-nodes for each core demonstration node to construct a case structure tree.

[0010] The filtering module is used to calculate the matching score between each major adaptation sub-node and the major attributes in the student's objective profile through a pre-set graph neural network model, and generate an adaptation content sequence for each major adaptation demonstration layer based on the matching score.

[0011] The collaborative alignment module is used to unify the teaching timeline with the presentation sequence of the main presentation layer as the main line, establish content synchronization relationship between the main presentation layer and the professional adaptation presentation layers, and generate a collaborative control instruction set containing time encoding and content mapping relationship.

[0012] The demonstration module is used to control the main presentation layer display terminal to present the core presentation node content based on the collaborative control instruction set, and to simultaneously control the various professional adaptation presentation layer display terminals to present the corresponding adaptation content sequence.

[0013] The switching monitoring module is used to regenerate the appropriate content sequence and collaborative control instruction set for each major's appropriate presentation layer based on the updated case structure tree if the main lecturer's presentation layer triggers a case switching instruction during the teaching process.

[0014] As a further aspect of the present invention, the display logic module constructs a classroom logic display structure based on the professional attributes in the objective portrait data of students in the class, including a lecturer demonstration layer and a professional adaptation demonstration layer. Each demonstration layer is associated with an independent display terminal, specifically including:

[0015] The objective profile data of students in the class includes the students' major attributes and each student's unique number;

[0016] Based on the professional attributes in the objective profile data of students in the class, all students are divided into multiple non-overlapping professional groups. The professional attributes include professional tags and curriculum system tags.

[0017] Create a corresponding professional adaptation demonstration layer for each professional group area, and create a lecturer demonstration layer for the lecturer. Integrate the lecturer demonstration layer with all professional adaptation demonstration layers into a classroom logical display structure.

[0018] Based on the classroom's logical display structure, the main presentation layer is mapped to the main display terminal in the classroom, and the professional adaptation presentation layers are mapped to the designated auxiliary display terminals in the classroom.

[0019] As a further aspect of the present invention, the case analysis module analyzes and extracts core teaching knowledge points from ideological and political education cases, maps each core teaching knowledge point to a core demonstration node, and generates corresponding professionally adapted sub-nodes for each core demonstration node. The construction of the case structure tree specifically includes:

[0020] The ideological and political education teaching cases are taken from the original material library of ideological and political education teaching cases. The materials include text, image resources or video clips. The correspondence between the materials and the corresponding course system tags is established and information is extracted to identify and mark the core teaching knowledge points of the teaching cases.

[0021] Using the ideological and political education teaching case identifier as the case root node, each core teaching knowledge point is mapped to a core demonstration node under the case root node. Based on the coverage relationship of all professional attribute keywords entered in the teaching system, a corresponding professional adaptation sub-node is generated for each core demonstration node to construct a case structure tree.

[0022] The case structure tree is expanded based on the course system tags corresponding to the professional adaptation sub-nodes, and the corresponding material index list is matched.

[0023] As a further aspect of the present invention, the expansion of the case structure tree based on the course system tags corresponding to the professional adaptation sub-nodes specifically includes:

[0024] Summarize the set of course system tags associated with each professional adaptation sub-node under each core demonstration node and evaluate the overlap. When the overlap meets the set conditions, activate the corresponding cross-professional adaptation sub-node.

[0025] The cross-major adaptation sub-node is associated with two or more professional attributes at the same time, and selects materials from the original material library that are consistent with the overlapping course system tags as a cross-material index list and associates them with the cross-major adaptation sub-node.

[0026] As a further aspect of the present invention, in the screening module, the preset graph neural network model structure adopts a heterogeneous graph structure composed of professional groups and professional matching sub-nodes, and performs training iterations in an offline environment. The model inputs a case structure tree and outputs the matching score of each professional matching sub-node.

[0027] As a further aspect of the present invention, the filtering module calculates the matching score between each professional adaptation sub-node and the professional attributes in the student's objective profile, and generates an adaptation content sequence for each professional adaptation demonstration layer based on the matching score, specifically including:

[0028] The professional adaptation sub-nodes under each core demonstration node in the case structure tree are used as candidate nodes of the graph neural network, and professional labels and curriculum system labels are used as node features.

[0029] A graph structure is constructed based on the relationship between the professional attributes of students in different professional grouping areas and the professional matching sub-node labels. The matching score of each professional matching sub-node is calculated by graph neural network.

[0030] Select the professional adaptation sub-node with the highest score and establish the control channel of the professional adaptation demonstration layer auxiliary display terminal corresponding to the professional group area.

[0031] As a further aspect of the present invention, the collaborative alignment module uses the presentation sequence of the main presentation layer as the main line to unify the teaching timeline, establishes a content synchronization relationship between the main presentation layer and the professional adaptation presentation layers, and generates a collaborative control instruction set containing time encoding and content mapping relationships, specifically including:

[0032] Using the presentation sequence of the main presentation layer as the main thread, the timeline is divided into a sequence containing multiple ordered content segments, and a time code is defined for the starting point of each content segment.

[0033] Based on the hierarchical relationship between the core demonstration node and the professional adaptation sub-node in the case structure tree, a synchronization strategy is set between the professional adaptation demonstration layer and the main presentation layer. The synchronization strategy includes immediate synchronization, delayed synchronization, and early synchronization.

[0034] Based on the time encoding and synchronization strategy, a playback schedule associated with the timeline of the main presentation layer is generated for the professional adaptation presentation layer corresponding to each professional group area;

[0035] Based on the playback schedule and the material index associated with the professionally adapted presentation layer, a collaborative control instruction set is synthesized, which includes the target auxiliary display terminal identifier, trigger time, and content index.

[0036] As a further aspect of the present invention, the demonstration module, based on a collaborative control instruction set, controls the main presentation layer display terminal to present the core presentation node content, and simultaneously controls the various professional adaptation presentation layer display terminals to present the corresponding adaptation content sequence, specifically including:

[0037] The collaborative control instruction set is parsed, and the trigger time, target display terminal identifier, and associated content index contained therein are extracted. Based on the trigger time, the instruction entries in the collaborative control instruction set are inserted into the rendering task queue of the corresponding display terminal.

[0038] Based on the content index, the corresponding material resources are scheduled from the material library and encapsulated into a data package to be rendered. When the trigger time arrives, the data package to be rendered is sent to the display terminal pointed to by the target display terminal identifier, driving the main presentation layer display terminal to present the core presentation node content, and simultaneously driving each professional adaptation presentation layer display terminal to present the corresponding adaptation content sequence.

[0039] As a further aspect of the present invention, the regeneration of the adaptation content sequence and collaborative control instruction set for each professional adaptation demonstration layer in the switching monitoring module specifically includes:

[0040] Receive a manual case switching command sent from the main presentation layer display terminal, and parse the target case identifier from the command;

[0041] Based on the target case identifier, recalculate the adaptation content sequence of each professional adaptation demonstration layer, and update the collaborative control instruction set in conjunction with the progress of the current unified teaching timeline.

[0042] As a further aspect of the present invention, it includes a central teaching controller, a data transmission interface, and an interactive input trigger:

[0043] The central teaching controller integrates the display logic module, case analysis module, filtering module, collaborative alignment module, demonstration module, and switching monitoring module, and connects to the independent display terminals of each demonstration layer through a data transmission interface;

[0044] The data transmission interface is used to transmit the sequence of adapted content output by the demonstration module in the central teaching controller to the independent display terminal.

[0045] The interactive input trigger is used to receive operation instructions from the main lecturer during the teaching process and is connected to the switching monitoring module of the central teaching controller.

[0046] The technical effects and advantages of the AI-based intelligent recommendation-based personalized ideological and political education teaching system of this invention are as follows:

[0047] By constructing a classroom logic display structure that combines a main lecture demonstration layer with multiple professionally adapted demonstration layers, differentiated teaching demonstrations for students from different majors are achieved within the same ideological and political education class. This effectively solves the problem of difficulty in catering to diverse professional backgrounds in large-class teaching. Through structured analysis of ideological and political education teaching cases and the construction of a case structure tree containing core demonstration nodes and professionally adapted sub-nodes, teaching cases can be organized and invoked in a unified main thread and in parallel with professionally adapted content. This ensures consistency in the main teaching thread while enhancing the relevance of teaching content to students' professional curriculum system. By introducing a graph neural network to comprehensively calculate the matching relationship between professionally adapted sub-nodes and student curriculum system tags, intelligent selection of professionally adapted content is achieved, making teaching demonstrations more aligned with the knowledge structure characteristics of different professional groups. Furthermore, by unifying the teaching timeline and collaborative control instruction set, synchronous or staggered display between the main lecture demonstration layer and various professionally adapted demonstration layers is achieved, avoiding disruptions in the classroom rhythm and improving overall teaching coherence.

[0048] This invention can improve the teaching relevance and presentation flexibility of ideological and political education classes without increasing the extra operational burden on teachers, and has good teaching adaptability and promotion value. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the structure of a personalized ideological and political education system based on AI intelligent recommendation according to the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0051] Example 1

[0052] Figure 1 This invention presents a personalized ideological and political education system based on AI intelligent recommendation, comprising:

[0053] The display logic module is used to construct a classroom logical display structure based on the professional attributes in the objective profile data of students in the class. It includes a lecturer demonstration layer and a professional adaptation demonstration layer, with each demonstration layer associated with an independent display terminal.

[0054] The case analysis module is used to analyze and extract core teaching knowledge points from ideological and political teaching cases, map each core teaching knowledge point to a core demonstration node, and generate corresponding professional-adaptive sub-nodes for each core demonstration node to construct a case structure tree.

[0055] The filtering module is used to calculate the matching score between each major adaptation sub-node and the major attributes in the student's objective profile through a pre-set graph neural network model, and generate an adaptation content sequence for each major adaptation demonstration layer based on the matching score.

[0056] The collaborative alignment module is used to unify the teaching timeline with the presentation sequence of the main presentation layer as the main line, establish content synchronization relationship between the main presentation layer and the professional adaptation presentation layers, and generate a collaborative control instruction set containing time encoding and content mapping relationship.

[0057] The demonstration module is used to control the main presentation layer display terminal to present the core presentation node content based on the collaborative control instruction set, and to simultaneously control the various professional adaptation presentation layer display terminals to present the corresponding adaptation content sequence.

[0058] The switching monitoring module is used to regenerate the appropriate content sequence and collaborative control instruction set for each major's appropriate presentation layer based on the updated case structure tree if the main lecturer's presentation layer triggers a case switching instruction during the teaching process.

[0059] In the display logic module, a classroom logic display structure is constructed based on the professional attributes in the objective profile data of students in the class. It includes a lecturer demonstration layer and a professional adaptation demonstration layer, with each demonstration layer associated with an independent display terminal.

[0060] For specific teaching classes, objective student profile data was acquired and organized. This data originated from objective data records already generated during the teaching management process. The data included at least a unique ID for each student and clearly marked professional attribute information. The unique ID ensured the traceability and non-confusing nature of student identities during data processing, while the professional attribute served as the core basis for subsequent grouping and demonstration adaptation. The professional attribute consisted of two levels: firstly, a professional tag, identifying the student's academic discipline or specialization, such as engineering, economics and management, or humanities and social sciences; secondly, a curriculum system tag, reflecting the core course structure and knowledge modules corresponding to the major in the curriculum, such as engineering principles, industrial economics, and policy theory. After collecting the objective student profile data, all students were categorized according to their professional tags. A one-to-one mapping method was used to group students with the same professional tag into the same professional group area, while students with different professional tags or curriculum system tags were assigned to different professional group areas, ensuring no overlap or intersection between the professional group areas.

[0061] For each defined professional group, a corresponding professional-adaptive demonstration layer is created sequentially. Each professional-adaptive demonstration layer exists logically independently, internally associated only with the student group within its corresponding professional group, and is used to carry teaching demonstration content that matches the professional background and curriculum system. Simultaneously, a lecturer demonstration layer is created for the main lecturer undertaking the unified teaching task. This lecturer demonstration layer is used to present common content, core case nodes, and main teaching information in ideological and political education. Subsequently, the lecturer demonstration layer and all professional-adaptive demonstration layers are integrated according to unified classroom structure rules to form a complete logical classroom display structure. This logical classroom display structure clearly distinguishes the subordinate and collaborative relationships between the lecturer layer and multiple professional-adaptive layers, describing how different demonstration layers are organized within the same classroom environment. After logical integration, a display terminal mapping operation is performed according to the classroom's logical display structure. The main lecture presentation layer is fixedly mapped to the main display terminal in the classroom used for overall teaching presentation, such as the main screen located at the front of the classroom. Simultaneously, the presentation layers adapted to each major are mapped to pre-designated auxiliary display terminals in the classroom. These auxiliary display terminals can be distributed in different areas of the classroom, forming a spatial correspondence with the corresponding major group areas. Through this mapping method, different presentation layers are clearly distinguished at the physical display level, thereby achieving a parallel teaching structure in the same classroom where the main lecture content is presented uniformly and the professional content is presented differently.

[0062] In the case analysis module, the core teaching knowledge points of ideological and political teaching cases are analyzed and extracted. Each core teaching knowledge point is mapped to a core demonstration node, and a corresponding professional adaptation sub-node is generated for each core demonstration node to construct a case structure tree.

[0063] The original material library for ideological and political education teaching cases is uniformly compiled and entered by teaching management personnel or curriculum development personnel. The materials cover teaching syllabus texts, policy document excerpts, teaching images, and video clips in the form of classroom lectures or documentaries. Each piece of material undergoes manual review during entry to determine its corresponding curriculum system tag. These tags are derived from publicly available professional training programs and course structure catalogs from universities, ensuring the stability and reusability of the tag system. After establishing the correspondence between materials and curriculum system tags, textual materials are directly decomposed into semantic units, while image and video materials are converted into structured text descriptions using existing text descriptions or subtitles, forming a unified set of structured texts. Subsequently, information extraction is performed on the structured texts, analyzing each policy theme, theoretical viewpoint, practical case, and value guidance element within the teaching content. Through keyword positioning and contextual association, the recurring and dominant teaching themes in each teaching case are identified and marked as core teaching knowledge points.

[0064] After identifying and labeling the core teaching knowledge points, a root node of the case structure tree is constructed using the unique case identifier of each ideological and political education teaching case as the logical starting point. This root node is only used to uniformly manage all demonstration content under that teaching case and does not directly carry specific teaching materials. Subsequently, each labeled core teaching knowledge point is mapped one by one to a core demonstration node under the root node. Each core demonstration node corresponds to a teaching content unit that can be independently presented and invoked during the teaching process. For each core demonstration node, a coverage relationship analysis is performed based on a set of professional attribute keywords that have been pre-entered and maintained in the teaching system. The set of professional attribute keywords comes from the professional training objectives, course names, and professional description texts, such as "mechanical design," "economic management," and "ideological and political theory." By comparing the semantic coverage relationship between the text content of the core teaching knowledge points and the professional attribute keywords, several professionally adapted sub-nodes are generated for each core demonstration node. Each sub-node is clearly associated with a professional attribute and structurally belongs to the corresponding core demonstration node. The resulting case structure tree is logically presented as a hierarchical structure of "case root node - core demonstration node - professional adaptation sub-node". This structure does not contain chronological information and is only used to describe the organizational relationship of teaching content from different professional perspectives.

[0065] After constructing the core demonstration node and its subordinate professional adaptation sub-nodes in the case structure tree, the system centrally summarizes and processes the course system tag sets associated with each professional adaptation sub-node that has been generated under the same core demonstration node. Each professional adaptation sub-node is bound to at least one explicit course system tag during generation. These course system tags originate from standardized course classification information in university training programs, course outlines, or teaching plans. The tags are stored as strings and managed in a standardized tag dictionary. For all professional adaptation sub-nodes under the same core demonstration node, the system reads their corresponding course system tag sets one by one and performs pairwise comparisons between tag sets within the same node. The overlap is assessed by counting the number of overlapping course system tags between different professional adaptation sub-nodes. The overlap is controlled using explicit threshold rules. For example, under a core demonstration node, if any two or more professional adaptation sub-nodes have at least two completely identical tag items in their course system tag sets, the professional combination is considered to have cross-expansion value. To avoid excessively broad or narrow overlap, the overlap threshold is set to 2. This threshold is set by teaching administrators during system initialization and can be adjusted based on the actual teaching scale. For example, in teaching scenarios with fewer professional categories, the threshold can be adjusted to one label, while in comprehensive classes with more professional categories, two or more labels can be maintained. Through the above overlap evaluation rules, the system activates cross-professional adaptation sub-nodes for professional combinations that meet the conditions in the case structure tree. This achieves a horizontal expansion of the case structure tree based on the original professional adaptation structure, allowing both single-professional adaptation sub-nodes and cross-professional adaptation sub-nodes to exist simultaneously under the same core demonstration node.

[0066] After activating the cross-disciplinary adaptation sub-nodes, specific node construction and material association operations are performed for each activated sub-node. Each cross-disciplinary adaptation sub-node is bound to two or more professional attribute information simultaneously, such as simultaneously associating "Engineering Technology" and "Economic Management," or simultaneously associating "Ideological and Political Theory" and "Public Administration." This professional attribute information is directly inherited from the original professional adaptation sub-nodes participating in the cross-over assessment, thus ensuring the integrity and consistency of the cross-nodes in terms of professional semantics. After completing the professional attribute binding of the cross-disciplinary adaptation sub-nodes, the system performs targeted filtering operations on the original material library based on the overlapping course system tags identified during the aforementioned overlap assessment process. Specifically, the system retrieves material entries in the material library that are completely consistent with the overlapping course system tags. These materials include text materials, image resources, and video clips, and each material entry has already established a corresponding relationship with the corresponding course system tag during the database entry stage. For example, when the overlapping course system tags of the cross-disciplinary adaptation sub-nodes are "Engineering Management" and "Industrial Chain Fundamentals," the system will automatically filter out engineering management case texts, industrial chain structure diagrams, and related video materials that are simultaneously labeled with the above tags or the corresponding course system tags. The system will then compile the index identifiers of these materials into a unified cross-material index list. This cross-material index list is stored in list form and bound to the cross-disciplinary adaptation sub-nodes, serving as a direct data source for generating subsequent professional adaptation content sequences. In this embodiment, to ensure the integrity and continuity of teaching, the system stipulates that each cross-material index list must contain at least one text material and one non-text material, such as an image or video. If a material combination meeting this condition is not found in the original material library, the system will activate a material supplementation mechanism through the teaching management configuration item, retrieving suitable materials from adjacent teaching cases under the same course system tag to supplement the content. This ensures that the cross-disciplinary adaptation sub-nodes have a complete and demonstrable content structure during actual teaching demonstrations.

[0067] In the filtering module, the matching score between each professional adaptation sub-node and the professional attributes in the student's objective profile is calculated, and an adaptation content sequence is generated for each professional adaptation demonstration layer based on the matching score.

[0068] The graph neural network model was trained offline, with training data derived from the long-term accumulated ideological and political education teaching case structure data and historical teaching demonstration records of the teaching system. Training samples were constructed using the matching relationship between "core demonstration nodes—professionally compatible sub-nodes—professionally group profiles" as the basic unit. All professionally compatible sub-nodes under each core demonstration node were extracted from the case structure tree, and each professionally compatible sub-node, along with its associated profession tag, curriculum system tag, and material type tag, constituted candidate node description information. Secondly, teaching demonstration processes that had occurred were collected from historical teaching records, and the professionally compatible sub-nodes actually used by different professional groups under specific core demonstration nodes were extracted as the basis for positive sample labeling. Sample labeling was completed using a "successfully matched, unselected" approach. Professionally compatible sub-nodes that were actually used in teaching demonstrations and confirmed by teachers and students were labeled as high-match samples, while other sub-nodes that were not selected but possessed the same candidate qualifications were labeled as low-match samples. To avoid the influence of single teacher or single classroom bias on the model, the training samples were evenly sampled according to teaching class, course type, and case theme during the construction process. This ensures that the matching rules learned by the model come from real teaching practices in multiple batches and scenarios. The training set constructed in this way can fully reflect the real correlation between the professional grouping course system labels and professional matching sub-nodes, providing a stable and repeatable sample foundation for the effective training of the graph neural network model.

[0069] After constructing the training data, the graph neural network model undergoes multiple rounds of training iterations in an offline environment. The model structure adopts a heterogeneous graph structure composed of professional group nodes and professional matching sub-nodes. Node features are encoded by professional labels, curriculum system labels, and material attributes, while edge relationships describe the degree of label overlap and correlation strength between professional groups and professional matching sub-nodes. During training, the model uses multiple rounds of feature propagation and aggregation to gradually acquire a comprehensive representation reflecting its matching degree with the professional group for each professional matching sub-node, and outputs a corresponding matching score. After each round of training, the model updates its parameters based on the annotation results in the training samples. The update strategy aims to improve the scores of high-matching samples and reduce the scores of low-matching samples, iterating repeatedly until the overall matching ranking results stabilize on the validation sample set. The number of training rounds is set according to the size of the training set; for example, dozens of complete iterations are performed when there are thousands of samples to ensure sufficient model convergence. After the model training is completed, it is deployed in the teaching system with fixed parameters and used only for matching calculations during the online inference phase. No further parameter updates are performed during online use. During online inference, the system only needs to convert the professional matching sub-nodes and professional group profiles in the current case's structure tree into a graph structure input model to quickly output the matching scores of each professional matching sub-node, enabling real-time selection of teaching demonstration content. By separating offline training from online inference, the stability and response efficiency of the model's inference process are ensured, while avoiding the impact of frequent changes in the model structure on the continuity of teaching during online teaching.

[0070] The pre-constructed case structure tree serves as the structural input source for the graph neural network model. The candidate nodes of the graph neural network are explicitly limited to the professionally adapted sub-nodes already generated and activated under each core demonstration node in the case structure tree, avoiding invalid computations within irrelevant content. Specifically, for each core demonstration node, the system reads all its attached professionally adapted sub-nodes, including basic professionally adapted sub-nodes and cross-professionally adapted sub-nodes activated by the overlap of curriculum system tags, and maps these sub-nodes one by one to node entities in the graph neural network. Each node entity is bound to a fixed set of node attributes, which includes at least professional tag information and curriculum system tag information. The professional tag represents the subject category to which the professionally adapted sub-node belongs, such as engineering technology, economics and management, or ideological and political theory; the curriculum system tag represents the curriculum system direction covered by the professionally adapted sub-node, such as engineering principles, supply chain analysis, policy theory, or engineering ethics. The aforementioned tag information comes from a pre-maintained professional attribute keyword library and curriculum system tag library in the teaching system, and has been standardized and stored in a structured manner during the case structure tree generation stage. Through the above methods, the node set and node attribute sources of the graph neural network model are clear and stable, and the model input does not depend on the real-time text semantic parsing process. This ensures that the graph neural network performs calculations only around the teaching demonstration structure during the running phase, enabling technicians in the relevant technical field to complete the graph structure initialization according to the above node construction method.

[0071] For each major group region, an input graph structure for a graph neural network is constructed based on the student objective profile data corresponding to that major group region. Specifically, the system first extracts the set of major labels and the set of curriculum system labels corresponding to that major group region from the student objective profile data, and maps this set as a whole to group attribute nodes in the graph neural network. These group attribute nodes represent the overall curriculum system characteristics of the current major group, rather than the individual characteristics of a single student. Subsequently, the system constructs edge relationships in the graph structure based on the label association relationships between the group attribute nodes and each major-matching sub-node. When the curriculum system labels contained in a group attribute node are consistent with or partially overlap with the curriculum system labels associated with a certain major-matching sub-node, an association edge is established between the group attribute node and that major-matching sub-node, and the corresponding label overlap strength level is recorded in the edge attributes. After the graph structure is constructed, it is input into a pre-trained graph neural network model, which performs multi-layer feature propagation and aggregation processing on the node attributes and association edge relationships. By leveraging the neighborhood information transfer mechanism within the graph neural network, each professional matching sub-node simultaneously integrates its own label attributes and the strength of its association with professional group attribute nodes during the calculation process, thereby outputting a corresponding matching score. This matching score quantifies the teaching suitability of the professional matching sub-node under the current professional group conditions. The model output is a set of deterministic score values ​​for each professional matching sub-node under each core demonstration node.

[0072] After obtaining the matching scores of each professional adaptation sub-node from the graph neural network model output, all candidate professional adaptation sub-nodes under the same core demonstration node are sorted. The sorting rule is to arrange them according to the matching score from high to low, and select the professional adaptation sub-node with the highest score as the final adaptation node for that professional group area under the current core demonstration node. When multiple professional adaptation sub-nodes have similar scores, the system makes a decision according to a preset priority rule. This priority rule is explicitly set in implementation as follows: cross-professional adaptation sub-nodes take precedence over single-professional adaptation sub-nodes, and sub-nodes with a wider coverage of course system tags take precedence over sub-nodes with a narrower coverage. The priority rule is uniformly configured by teaching administrators during the system deployment phase, for example, setting the priority of cross-nodes to one level higher than that of single-professional nodes. After determining the final professional adaptation sub-node, the teaching system reads the material index list associated with the sub-node and establishes a one-to-one mapping relationship between the index list and the professional adaptation demonstration layer auxiliary display terminal of the corresponding professional group area, generating a stable control channel identifier. The control channel is used to accurately send the corresponding material content to the designated auxiliary display terminal during the subsequent teaching timeline scheduling process, thereby ensuring that the content displayed in the professional adaptation demonstration layer is consistent with the matching result of the graph neural network.

[0073] In the collaborative alignment module, the teaching timeline is unified with the presentation sequence of the main presentation layer as the main line, and a content synchronization relationship is established between the main presentation layer and the professional adaptation presentation layers, generating a collaborative control instruction set that includes time encoding and content mapping relationship.

[0074] The presentation content in the main presentation layer uses the teacher's actual lecture sequence as the sole reference, linearly unfolding the entire classroom presentation process. Before being presented in the classroom, each teaching case is broken down into several core presentation nodes. Each core node corresponds to a continuous and complete segment of content in the main presentation layer, which may include textual descriptions, image displays, or video playback. Based on this predetermined presentation sequence, the entire teaching process is mapped onto a continuously advancing timeline, which is then divided into multiple interconnected content segments. Each content segment corresponds to the actual presentation interval of a core presentation node. To ensure clear time references for subsequent control and synchronization operations, a unique time code is defined at the beginning of each content segment. This time code is generated sequentially, for example, using the start of the class as the initial reference point, the first content segment is marked with the first time code, the second with the second time code, and so on, until the end of the classroom presentation. This time code does not involve specific physical time calculations but serves as a logical identifier for content switching and synchronization scheduling, indicating the current position of the classroom presentation process.

[0075] After constructing the timeline and corresponding time codes for the main presentation layer, the synchronization method between the professional adaptation presentation layer and the main presentation layer is systematically set based on the clearly defined hierarchical relationships in the case structure tree. Specifically, each core presentation node is associated with several professional adaptation child nodes in the case structure tree. These child nodes represent different versions of the same teaching topic presented in different professional contexts. For this hierarchical relationship, the position of each professional adaptation child node in terms of pedagogical understanding and its relationship with the main presentation content are analyzed one by one, and a clear synchronization strategy is configured for the professional adaptation presentation layer accordingly. When the professional adaptation content and the main presentation content are completely consistent at the cognitive level and suitable for simultaneous presentation with the main presentation content, an immediate synchronization strategy is set, so that the professional adaptation presentation layer starts displaying synchronously with the main presentation layer when the corresponding time code is triggered. When the professional adaptation content needs to appear as supplementary explanation after the main presentation, a delayed synchronization strategy is set, and the delay method is clearly defined for this delayed synchronization, such as delaying the display by one content segment. When professionally adapted content is used to guide students to establish background knowledge or understanding framework in advance, it is set to an advance synchronization strategy so that it is presented in advance before the main presentation layer enters the corresponding core presentation node.

[0076] For each professionally adapted demonstration layer corresponding to a professional group area, a playback schedule matching the timeline of the main presentation layer is generated. In practice, the time-encoded sequence of the main presentation layer is used as the baseline input, and the trigger time for each adapted content within that demonstration layer is calculated one by one, based on the synchronization strategy corresponding to that professionally adapted demonstration layer. For adapted content using an immediate synchronization strategy, the trigger time code in its playback schedule remains consistent with the time code of the core demonstration node corresponding to the main presentation layer; for adapted content using a delayed synchronization strategy, the trigger time code is shifted to a pre-set subsequent time code based on the corresponding time code; for adapted content using an early synchronization strategy, its trigger time code is moved forward to a specified time code position before the target core demonstration node. After generating the complete playback schedule, each playback record is further expanded using the material index list associated with that professionally adapted demonstration layer, forming instruction entries containing the target auxiliary display terminal identifier, trigger time code, and specific material index information. Finally, all instruction entries are summarized and encapsulated into a unified collaborative control instruction set. This instruction set is arranged in time-coded order and can be parsed and executed sequentially during classroom demonstrations. This allows each auxiliary display terminal to accurately present the corresponding professional and adapted content at the correct time, ensuring the consistency of time and content in the multi-layered demonstration process.

[0077] In the demonstration module, based on the collaborative control instruction set, the main presentation layer display terminal is controlled to present the core demonstration node content, and the corresponding professional adaptation presentation layer display terminals are simultaneously controlled to present the corresponding adaptation content sequence.

[0078] The collaborative control instruction set is stored in structured data format. Each instruction entry explicitly contains three essential fields: trigger time, target display terminal identifier, and content index identifier. The trigger time is represented using a relative time encoding method under a unified teaching timeline, for example, using the start of teaching as zero point and encoding in increments of minutes or seconds. The target display terminal identifier uniquely identifies the main presentation layer display terminal or the auxiliary display terminal corresponding to a specific major's adaptation presentation layer in the classroom. For example, it can be distinguished by numbering the main display terminal from the auxiliary display terminals corresponding to each major group. The content index identifier points to a specific teaching material entry in the material library with an established index relationship. During implementation, the collaborative control instruction set is first parsed line by line. According to the target display terminal identifier recorded in the instruction entry, the instruction entries are categorized and assigned to the corresponding display terminal's set of instructions to be executed. After categorization, the instruction sets corresponding to each display terminal are sorted according to the execution time order based on the trigger time field, forming a dedicated rendering task queue for that display terminal. The rendering task queue is arranged in chronological order, and each queue element is bound to a specific trigger time and content index information, ensuring that the instructions are executed strictly according to the teaching rhythm during the teaching demonstration. The trigger time is set based on the presentation progress of the main presentation layer. For example, when the main presentation layer enters the display stage of a certain core presentation node, the corresponding time code is written into the instruction entry, and the instruction entries of the auxiliary display terminal also use this time code as the scheduling basis. In this way, the originally discrete collaborative control instruction set is transformed into an ordered and schedulable rendering task queue within each display terminal.

[0079] When the trigger time for an instruction entry in a rendering task queue arrives, the corresponding teaching material resource is precisely located from the material library based on the content index identifier recorded in the instruction entry. All material resources in the material library have been uniformly managed and indexed in the early stages. Different types of materials, including text documents, image resources, and video clips, all use the content index as the sole retrieval entry point. During scheduling, the corresponding material is directly read based on the content index and packaged according to the display terminal's presentation requirements to form a data package to be rendered. This data package contains the material's core data and necessary presentation control information, such as media type identifiers, playback duration information, and display area parameters. The display area parameters are used to limit the presentation area position of the material on the target display terminal. After packaging, the data package to be rendered is sent to the display terminal indicated by the target display terminal identifier through the teaching demonstration control module. For the main presentation layer display terminal, the data packets to be rendered correspond to the core presentation node content, ensuring that the main screen content remains consistent with the teaching mainline during the teacher's presentation. For each major-specific adaptation presentation layer display terminal, the data packets to be rendered correspond to the adapted content sequence that has passed the major-specific adaptation screening, enabling students in different major groups to synchronously receive teaching materials matching their major backgrounds at the same teaching time. The execution method of triggering time adopts a combination of time polling and event triggering. For example, the current teaching timeline position is detected with a fixed time step, and the sending operation is immediately executed when it is detected to be consistent with the trigger time of the instruction entry, thereby ensuring the consistency and stability of the presentation actions across multiple terminals in the time dimension.

[0080] In the switching monitoring module, the adaptation content sequence and collaborative control instruction set of each professional adaptation demonstration layer are regenerated.

[0081] The main presentation layer display terminal is responsible for controlling the unified teaching pace during the teaching process. When the teacher needs to switch the current teaching case based on the actual situation of the classroom lecture, a manual case switching operation is triggered through the main presentation layer display terminal. This operation is sent to the teaching system control unit in the form of a clear control command. The control command includes at least one case identifier field to uniquely identify the target teaching case, and a time identifier field to indicate the time when the switching action occurs. After receiving the manual case switching command, the system immediately performs an integrity check on the command content, confirms that the command originates from a registered main presentation layer display terminal, and parses the command format to accurately extract the target case identifier information. The case identifier corresponds to a case record in the teaching case library and is used to clearly indicate the target ideological and political teaching case that the current teaching process should switch to. During the parsing process, the system also records the progress position of the unified teaching timeline when the case switching command is triggered, so as to maintain the continuity of the teaching pace when the content is rearranged in the future. In this way, the system can respond to teachers' real-time teaching decisions without changing the classroom logic display structure. This ensures that case switching behavior has a clear control entry point, a traceable trigger source, and a resolvable target, providing accurate input conditions for the subsequent recalculation of content in the demonstration layer for each major.

[0082] After parsing the target case identifier, the system retrieves the corresponding case structure tree data from the teaching case database based on the identifier. For each professional group region, the system, combined with the corresponding student profile data, re-executes the adaptation content selection process, traversing and analyzing the professional adaptation sub-nodes under each core demonstration node in the target case structure tree. Specifically, the system uses the professional tags and curriculum system tags associated with each professional group region as input conditions, calls the trained graph neural network model, calculates the matching score for the professional adaptation sub-nodes in the target case structure tree that meet the candidate conditions, and regenerates the adaptation content sequence corresponding to that professional group region based on the score results. The adaptation content sequence explicitly specifies the professional adaptation sub-nodes and their associated material index list that should be selected for each core demonstration node under the target case. This recalculation process is performed independently for each professional group region, ensuring that different professional groups can still obtain teaching demonstration content that matches their own curriculum system and teaching focus after case switching.

[0083] After recalculating the adapted content sequences for each major's demonstration layer, the teaching system further updates the collaborative control instruction set in conjunction with the real-time progress of the current unified teaching timeline. The system first obtains the timeline progress information recorded when the case switching instruction is triggered, determines the specific position of the current teaching content segment, and accordingly judges that the display of the target case should begin from the corresponding core demonstration node. Subsequently, based on the case structure tree of the target case and the newly generated adapted content sequence, the system regenerates the playback schedule corresponding to the main lecture demonstration layer and each major's adapted demonstration layer, reconfiguring the starting point, duration, and synchronization method of each demonstration layer on the timeline. During this process, the system maintains the controlling position of the main lecture demonstration layer as the main timeline thread, ensuring that the playback order of the main lecture content after case switching aligns with the teacher's teaching intent. Figure 1 Furthermore, the system precisely aligns the presentation times of each professional adaptation demonstration layer or sets necessary time offsets. Through this update mechanism, the system achieves a smooth content transition after case switching without interrupting classroom teaching, ensuring consistency and coordination in time and content between the main lecturer's demonstration layer and the professional adaptation demonstration layers, thereby maintaining the overall continuity and organization of classroom teaching.

[0084] The central teaching controller integrates the display logic module, case analysis module, filtering module, collaborative alignment module, demonstration module, and switching monitoring module, and connects to the independent display terminals of each demonstration layer through a data transmission interface.

[0085] The controller is an integrated industrial computing control device, comprising a processor component, a storage component, a bus interface component, and a multi-channel communication interface board. The processor component uses a multi-core general-purpose processor or an embedded processor to support the operating environment of the display logic module, case analysis module, filtering module, collaborative alignment module, demonstration module, and switching monitoring module. The storage component includes high-speed RAM and non-volatile storage media to store teaching case material indexes, case structure tree data, professional group mapping relationships, and collaborative control instruction sets. The central teaching controller physically integrates the various functional modules through an internal bus and establishes a stable data connection with external independent display terminals through communication interface boards. This controller is typically deployed in a rack-mount or embedded form within a classroom control cabinet or lectern console, possessing continuous operation capabilities and unified timing scheduling capabilities, serving as the physical hub for multi-demonstration layer collaborative control in the classroom.

[0086] The data transmission interface is used to transmit the adapted content sequence output by the demonstration module in the central teaching controller to the independent display terminal.

[0087] The data transmission interface, serving as the data communication carrier between the central teaching controller and each independent display terminal, is composed of wired or wireless communication modules. Its hardware includes a network interface chip, a physical layer transceiver module, and interface connection ports. The data transmission interface can employ an Ethernet interface, fiber optic interface, or high-speed wireless communication module, responsible for transmitting the adapted content sequence, material index identifiers, and collaborative control commands output by the demonstration module in the central teaching controller to the corresponding display terminal in the form of a data stream. The interface integrates a data buffer and transmission scheduling unit to ensure data integrity and timing consistency during concurrent transmission from multiple terminals. Each independent display terminal establishes a fixed or dynamic connection with the central teaching controller through its corresponding data transmission interface, enabling different professional adaptation demonstration layers to receive their respective content commands under unified control. This interface, as a physical communication channel, undertakes the data distribution task for synchronous multi-screen demonstrations in the classroom.

[0088] The interactive input trigger is used to receive operation instructions from the main lecturer during the teaching process and is connected to the switching monitoring module of the central teaching controller.

[0089] The interactive input trigger serves as the physical input carrier for the instructor's teaching operations. Its hardware form includes touch input devices, physical button controllers, or wireless command input terminals. The interactive input trigger integrates an input acquisition circuit, a command encoding unit, and a communication device to convert the teacher's operational behaviors during the teaching process into standardized control commands. The trigger can be installed on the lectern control panel, a handheld control terminal, or a desktop touch device, establishing a communication connection with the switching monitoring module of the central teaching controller via wired or wireless means. When the teacher performs operations such as switching cases, advancing demonstrations, or adjusting presentations, the interactive input trigger acquires the corresponding input signals in real time and sends the parsed control commands to the central teaching controller. This hardware carrier allows human operations during the teaching process to directly participate in the classroom demonstration control flow, constituting the physical entry point for human-computer interaction.

[0090] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0091] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0092] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0093] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0094] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0095] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0096] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0097] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0098] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A personalized ideological and political education teaching system based on AI intelligent recommendation, characterized in that, include: The display logic module is used to construct a classroom logical display structure based on the professional attributes in the objective profile data of students in the class. It includes a lecturer demonstration layer and a professional adaptation demonstration layer, with each demonstration layer associated with an independent display terminal. The case analysis module is used to analyze and extract core teaching knowledge points from ideological and political teaching cases, map each core teaching knowledge point to a core demonstration node, and generate corresponding professional-adaptive sub-nodes for each core demonstration node to construct a case structure tree. The filtering module is used to calculate the matching score between each major adaptation sub-node and the major attributes in the student's objective profile through a pre-set graph neural network model, and generate an adaptation content sequence for each major adaptation demonstration layer based on the matching score. The collaborative alignment module is used to unify the teaching timeline with the presentation sequence of the main presentation layer as the main line, establish content synchronization relationship between the main presentation layer and the professional adaptation presentation layers, and generate a collaborative control instruction set containing time encoding and content mapping relationship. The demonstration module is used to control the main presentation layer display terminal to present the core presentation node content based on the collaborative control instruction set, and to simultaneously control the various professional adaptation presentation layer display terminals to present the corresponding adaptation content sequence. The switching monitoring module is used to regenerate the appropriate content sequence and collaborative control instruction set for each major's appropriate demonstration layer based on the updated case structure tree if the main lecturer's demonstration layer triggers a case switching instruction during the teaching process. The display logic module constructs a classroom logic display structure based on the professional attributes in the objective student profile data, including a lecturer demonstration layer and a professional adaptation demonstration layer. Each demonstration layer is associated with an independent display terminal, specifically including: The objective profile data of students in the class includes the students' major attributes and each student's unique number; Based on the professional attributes in the objective student profile data, all students are divided into multiple non-overlapping professional grouping areas. The professional attributes include professional tags and curriculum system tags. Create a corresponding professional adaptation demonstration layer for each professional group area, and create a lecturer demonstration layer for the lecturer. Integrate the lecturer demonstration layer with all professional adaptation demonstration layers into a classroom logical display structure. Based on the classroom's logical display structure, the main presentation layer is mapped to the main display terminal in the classroom, and the professional adaptation presentation layers are mapped to the designated auxiliary display terminals in the classroom. In the filtering module, the preset graph neural network model structure adopts a heterogeneous graph structure composed of professional groups and professional matching sub-nodes, and performs training iterations in an offline environment. The model inputs a case structure tree and outputs the matching score of each professional matching sub-node.

2. The personalized ideological and political education system based on AI intelligent recommendation according to claim 1, characterized in that, In the case analysis module, core teaching knowledge points are extracted from ideological and political education cases. Each core teaching knowledge point is mapped to a core demonstration node, and a corresponding professionally adapted sub-node is generated for each core demonstration node. The construction of the case structure tree specifically includes: The ideological and political education teaching cases are taken from the original material library of ideological and political education teaching cases. The materials include text, image resources or video clips. The correspondence between the materials and the corresponding course system tags is established and information is extracted to identify and mark the core teaching knowledge points of the teaching cases. Using the ideological and political education teaching case identifier as the case root node, each core teaching knowledge point is mapped to a core demonstration node under the case root node. Based on the coverage relationship of all professional attribute keywords entered in the teaching system, a corresponding professional adaptation sub-node is generated for each core demonstration node to construct a case structure tree. The case structure tree is expanded based on the course system tags corresponding to the professional adaptation sub-nodes, and the corresponding material index list is matched.

3. The personalized ideological and political education system based on AI intelligent recommendation according to claim 2, characterized in that, The expansion of the case structure tree based on the course system tags corresponding to the professional adaptation sub-nodes specifically includes: Summarize the set of course system tags associated with each professional adaptation sub-node under each core demonstration node and evaluate the overlap. When the overlap meets the set conditions, activate the corresponding cross-professional adaptation sub-node. The cross-major adaptation sub-node is associated with two or more professional attributes at the same time, and selects materials from the original material library that are consistent with the overlapping course system tags as a cross-material index list and associates them with the cross-major adaptation sub-node.

4. The personalized ideological and political education system based on AI intelligent recommendation as described in claim 1, characterized in that, The filtering module calculates the matching score between each major-adaptation sub-node and the major attributes in the student's objective profile, and generates an adaptation content sequence for each major-adaptation demonstration layer based on the matching score. Specifically, this includes: The professional adaptation sub-nodes under each core demonstration node in the case structure tree are used as candidate nodes of the graph neural network, and professional labels and curriculum system labels are used as node features. A graph structure is constructed based on the relationship between the professional attributes of students in different professional grouping areas and the professional matching sub-node labels. The matching score of each professional matching sub-node is calculated by graph neural network. Select the professional adaptation sub-node with the highest score and establish the control channel of the professional adaptation demonstration layer auxiliary display terminal corresponding to the professional group area.

5. The personalized ideological and political education system based on AI intelligent recommendation according to claim 1, characterized in that, In the collaborative alignment module, the teaching timeline is unified based on the presentation sequence of the main presentation layer. Content synchronization relationships are established between the main presentation layer and the professional adaptation presentation layers. A collaborative control instruction set containing time encoding and content mapping relationships is generated, specifically including: Using the presentation sequence of the main presentation layer as the main thread, the timeline is divided into a sequence containing multiple ordered content segments, and a time code is defined for the starting point of each content segment. Based on the hierarchical relationship between the core demonstration node and the professional adaptation sub-node in the case structure tree, a synchronization strategy is set between the professional adaptation demonstration layer and the main presentation layer. The synchronization strategy includes immediate synchronization, delayed synchronization, and early synchronization. Based on the time encoding and synchronization strategy, a playback schedule associated with the timeline of the main presentation layer is generated for the professional adaptation presentation layer corresponding to each professional group area; Based on the playback schedule and the material index associated with the professionally adapted presentation layer, a collaborative control instruction set is synthesized, which includes the target auxiliary display terminal identifier, trigger time, and content index.

6. The personalized ideological and political education system based on AI intelligent recommendation according to claim 1, characterized in that, In the demonstration module, based on the collaborative control instruction set, the main presentation layer display terminal is controlled to display the core demonstration node content, and simultaneously, the corresponding professional adaptation presentation layer display terminals are controlled to display the corresponding adaptation content sequence. Specifically, this includes: The collaborative control instruction set is parsed, and the trigger time, target display terminal identifier, and associated content index contained therein are extracted. Based on the trigger time, the instruction entries in the collaborative control instruction set are inserted into the rendering task queue of the corresponding display terminal. Based on the content index, the corresponding material resources are scheduled from the material library and encapsulated into a data package to be rendered. When the trigger time arrives, the data package to be rendered is sent to the display terminal pointed to by the target display terminal identifier, driving the main presentation layer display terminal to present the core presentation node content, and simultaneously driving each professional adaptation presentation layer display terminal to present the corresponding adaptation content sequence.

7. The personalized ideological and political education system based on AI intelligent recommendation according to claim 1, characterized in that, The switching monitoring module specifically includes regenerating the adaptation content sequence and collaborative control instruction set for each professional adaptation demonstration layer, including: Receive a manual case switching command sent from the main presentation layer display terminal, and parse the target case identifier from the command; Based on the target case identifier, recalculate the adaptation content sequence of each professional adaptation demonstration layer, and update the collaborative control instruction set in conjunction with the progress of the current unified teaching timeline.

8. The personalized ideological and political education system based on AI intelligent recommendation according to claim 1, characterized in that, This includes a central teaching controller, data transmission interface, and interactive input triggers: The central teaching controller integrates the display logic module, case analysis module, filtering module, collaborative alignment module, demonstration module, and switching monitoring module, and connects to the independent display terminals of each demonstration layer through a data transmission interface; The data transmission interface is used to transmit the sequence of adapted content output by the demonstration module in the central teaching controller to the independent display terminal. The interactive input trigger is used to receive operation instructions from the main lecturer during the teaching process and is connected to the switching monitoring module of the central teaching controller.

Images