Intelligent beauty course resource management method and device, equipment and medium

Abstract

The invention relates to the technical field of resource management. The intelligent beauty course resource management method and device, the equipment and the medium are provided, and the method comprises the following steps: carrying out sub-atlas loading processing on a preset culture knowledge base based on a context identifier of a target teaching scene, and generating an aesthetic rule sub-atlas; performing activation processing on a preset agent role configuration library to generate a multi-role agent cluster, and performing collaborative analysis reasoning processing on the digital resource object according to the aesthetic rule sub-atlas to generate a consensus art analysis result; based on the consensus art analysis result, structured feedback generation processing is carried out, and structured art evaluation feedback is generated; structured art evaluation feedback is used as enhanced metadata to be bound with original resources, and an aesthetic course resource library capable of being retrieved in a cultural dimension is constructed, so that the technical effects of improving cross-cultural semantic analysis precision, enhancing teaching scene adaptive ability and optimizing multi-dimensional evaluation collaboration are achieved.

Description

Technical Field

[0001] This invention relates to the field of resource management technology, and in particular to intelligent methods, devices, equipment and media for managing resources in aesthetic education courses. Background Technology

[0002] With the development of globalized education, the need for cross-cultural adaptation of aesthetic education curriculum resources is becoming increasingly prominent.

[0003] Traditional technologies rely on manual annotation or single-cultural rule bases to extract deep cultural features from multimodal art resources. However, this approach suffers from insufficient cross-cultural semantic parsing capabilities, making it difficult to accurately identify cross-cultural symbols and generate comprehensive and objective digital resource objects. When facing the problem of knowledge adaptation in different teaching scenarios, traditional technologies use a static loading model of unified knowledge graphs, but fail to adapt them differently based on the depth of teaching, thus negatively impacting teaching effectiveness. Furthermore, traditional technologies often neglect the issues of synergy in multi-dimensional analysis of art evaluation and the construction of conflict arbitration mechanisms, resulting in a lack of completeness and authority in evaluation conclusions. Summary of the Invention

[0004] Therefore, it is necessary to provide intelligent methods, devices, equipment, and media for managing aesthetic education curriculum resources to address the aforementioned technical issues, so as to improve the accuracy of cross-cultural semantic analysis, enhance the adaptability of teaching scenarios, and optimize the synergy of multi-dimensional evaluation.

[0005] Firstly, this application provides an intelligent method for managing aesthetic education curriculum resources, which includes:

[0006] Cross-cultural semantic feature extraction is performed on the input multimodal aesthetic education resources to generate digital resource objects;

[0007] Based on the contextual identifiers of the target teaching scenario, a sub-graph loading process is performed on the pre-set cultural knowledge base to generate an aesthetic rule sub-graph.

[0008] The preset agent role configuration library is activated to generate a multi-role agent cluster. Based on the aesthetic rule sub-graph, the digital resource objects are subjected to collaborative analysis and reasoning to generate consensus-based art analysis results.

[0009] Structured feedback is generated based on the consensus-based art analysis results to produce structured art evaluation feedback.

[0010] By binding structured art evaluation feedback with the original resources as enhanced metadata, an art education curriculum resource library that can be searched from a cultural perspective is constructed.

[0011] In one embodiment, collaborative analysis and reasoning are performed on digital resource objects based on aesthetic rule sub-graphs to generate consensus-based art analysis results, including:

[0012] Control the graph analysis agent in the multi-role agent cluster to perform graph rule matching processing on digital resource objects and generate graph conformity analysis results;

[0013] Control the color semantic agent in the multi-role agent cluster to perform color symbol mapping on digital resource objects and generate cultural color analysis results;

[0014] Control the art history association agent in the multi-role agent cluster to perform historical school comparison processing on digital resource objects and generate artistic innovation analysis results;

[0015] The cultural context advisor agent in the multi-role agent cluster is controlled to perform aesthetic paradigm verification on digital resource objects and generate cultural compatibility analysis results.

[0016] The results of composition conformity analysis, cultural color analysis, artistic innovation analysis, and cultural adaptability analysis are cross-validated through a preset interactive protocol to generate a set of validation conclusions.

[0017] When there are conflicts in the verification conclusion set, arbitration is carried out according to the priority rules in the aesthetic rule subgraph to generate a consensus art analysis result.

[0018] When there are no conflicts in the verification conclusion set, the verification conclusion set is integrated to generate a conflict-free consensus set; based on the result fusion rules in the aesthetic rule subgraph, the conflict-free consensus set is restructured to generate a consensus-based art analysis result.

[0019] In one embodiment, when conflicts exist in the verification conclusion set, arbitration is performed according to the priority rules in the aesthetic rule subgraph to generate a consensus-based art analysis result, including:

[0020] Extract conflicting pairs of conclusions from the set of verification conclusions;

[0021] Based on the cultural authority weight rules in the aesthetic rule subgraph, the conflicting conclusions are weighted and processed to generate a preliminary arbitration result.

[0022] When the preliminary arbitration result does not meet the preset confidence threshold, the historical case library in the aesthetic rule sub-graph is called to perform similarity matching and generate supporting evidence for the case.

[0023] The following formula is used to dynamically revise the preliminary arbitration result based on supporting case evidence, generating the final arbitration conclusion:

[0024]

[0025] in, This indicates the final arbitration conclusion. Represents the set of conflicting conclusions. Rules indicating the weight of cultural authority This represents the weight calculation function. This represents the confidence level indicator function. This indicates that historical cases support the evidence. Indicates the case correction operator, This represents a dynamic correction function. Indicates the conclusion fusion operator, Represents the set of non-conflicting conclusions;

[0026] The final arbitration conclusion is merged with the non-conflicting conclusions in the set of verification conclusions to generate a consensus-based art analysis result.

[0027] In one embodiment, the cultural context advisor agent in the multi-role agent cluster is controlled to perform aesthetic paradigm verification on digital resource objects and generate cultural compatibility analysis results, including:

[0028] Perform cultural symbol recognition processing on digital resource objects to generate a set of cultural symbols;

[0029] Based on the cultural paradigm rule base in the aesthetic rule subgraph, the cultural symbol set is subjected to paradigm conformity verification processing to generate a basic paradigm score.

[0030] When cross-cultural symbols are detected in the cultural symbol set, a preset cross-cultural symbol conflict database is called to perform symbol conflict verification and generate symbol conflict identifiers; based on the symbol conflict identifiers, the basic paradigm score is dynamically adjusted in terms of symbol weights to generate an adjusted paradigm score.

[0031] When it is detected that there are no cross-cultural symbols in the cultural symbol set, the basic paradigm score is used as the adjusted paradigm score; cultural context restoration processing is performed on the digital resource objects to generate a context restoration analysis report; the adjusted paradigm score and the context restoration analysis report are comprehensively evaluated to generate cultural adaptability analysis results.

[0032] In one embodiment, structured art evaluation feedback is bound to the original resources as enhanced metadata to construct an art education curriculum resource library that can be searched along cultural dimensions, including:

[0033] Cultural dimension metadata is extracted and processed from structured art evaluation feedback to generate a cultural semantic metadata dataset;

[0034] Based on the cultural semantic meta-dataset, a cultural label hierarchy is constructed to generate a cultural dimension label tree, which includes branches of cultural schools, techniques, and aesthetic paradigms.

[0035] The original resources are enhanced with cross-cultural features, and the cultural semantic metadata is injected into the metadata layer of the original resources to generate enhanced aesthetic education resources.

[0036] Using the following formula, we can dynamically index and associate enhanced aesthetic education resources with cultural dimension tag trees to construct an aesthetic education course resource library that can be searched along the cultural dimension:

[0037]

[0038] in, This indicates a culturally searchable art education curriculum resource library. Represents a cultural dimension label tree. This indicates an enhanced collection of aesthetic education resources. This represents the index aggregation operator. This indicates the dimension index construction function. This represents a resource-tag association function. This represents a specific node in the cultural dimension label tree. This indicates enhanced resources for aesthetic education.

[0039] In one embodiment, based on the contextual identifiers of the target teaching scenario, a sub-graph loading process is performed on a pre-defined cultural knowledge base to generate an aesthetic rule sub-graph, including:

[0040] The contextual markers of the target teaching scenario are deconstructed to generate a set of core cultural elements;

[0041] Cross-cultural association analysis is performed based on the core cultural element set to generate a cultural association strength matrix.

[0042] The graph partitioning rule library in the preset cultural knowledge base is called to dynamically partition the cultural association strength matrix and generate a sub-graph loading scheme.

[0043] Perform hybrid graph construction processing based on the sub-graph loading scheme:

[0044] When the core cultural elements set is a single cultural type, load the corresponding cultural foundation sub-graph and generate an aesthetic rule sub-graph.

[0045] When the core cultural element set includes cross-cultural elements, a graph fusion process is performed based on the cultural association strength matrix to generate an aesthetic rule sub-graph.

[0046] In one embodiment, the cultural association strength matrix is ​​obtained using the following formula, including:

[0047]

[0048] in, Represents the cultural association strength matrix. Indicates the collection of core cultural elements. This indicates a pre-defined cross-cultural association rule base. Representation of element-rule interaction operators, This represents the matrix processing function.

[0049] Secondly, this application also provides an intelligent aesthetic education curriculum resource management device, which includes:

[0050] The feature extraction module is used to perform cross-cultural semantic feature extraction processing on the input multimodal aesthetic education resources to generate digital resource objects;

[0051] The graph loading module is used to load subgraphs from a pre-set cultural knowledge base based on contextual identifiers of the target teaching scenario, and generate aesthetic rule subgraphs.

[0052] The collaborative reasoning module is used to activate the preset agent role configuration library to generate a multi-role agent cluster, and to perform collaborative analysis and reasoning on digital resource objects based on the aesthetic rule sub-graph to generate consensus-based art analysis results.

[0053] The feedback generation module is used to generate structured feedback based on consensus-based art analysis results, producing structured art evaluation feedback.

[0054] The resource library construction module is used to bind structured art evaluation feedback as enhanced metadata with the original resources to build an art education curriculum resource library that can be searched from a cultural dimension.

[0055] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods in the first aspect of this application.

[0056] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the methods in the first aspect of this application.

[0057] The intelligent aesthetic education curriculum resource management method, device, equipment, and medium provided in this application include the following methods: cross-cultural semantic feature extraction processing of input multimodal aesthetic education resources; generating structured digital resource objects by mining the deep cultural symbols and semantic information contained in the resources, providing a standardized data foundation for subsequent analysis; and loading sub-graphs into a pre-set cultural knowledge base based on the contextual identifiers of the target teaching scenario, selecting suitable aesthetic rule knowledge in conjunction with the core cultural elements of the scenario, and constructing an aesthetic rule sub-graph that highly matches the teaching scenario, breaking the limitations of traditional static loading of unified knowledge graphs.

[0058] The pre-defined agent role configuration library is activated to form a multi-role agent cluster. Based on an aesthetic rule sub-graph, this cluster collaboratively analyzes and reasons about digital resource objects from dimensions such as composition, color, art historical relevance, and cultural context. Through cross-validation and conflict arbitration mechanisms, it generates unified and authoritative consensus-based art analysis results. Based on these consensus-based art analysis results, structured feedback generation is performed. Following standardized evaluation dimensions and expression formats, the analysis results are transformed into structured art evaluation feedback containing evaluation indicators, conclusions, and suggestions. This structured art evaluation feedback is used as enhanced metadata and bound to the original multimodal aesthetic education resources. By constructing a cultural dimension tagging system and a dynamic indexing mechanism, an aesthetic education curriculum resource library supporting cultural dimension retrieval is formed, achieving intelligent management and efficient application of aesthetic education resources. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 A flowchart of an intelligent aesthetic education curriculum resource management method according to one embodiment of the present invention;

[0061] Figure 2 The flowchart in one embodiment of the present invention describes how a cultural context advisor agent in a multi-role agent cluster performs aesthetic paradigm verification on digital resource objects and generates cultural compatibility analysis results.

[0062] Figure 3 This is a structural diagram of an intelligent aesthetic education curriculum resource management device according to one embodiment of the present invention. Detailed Implementation

[0063] To make the above-mentioned objects, features, and advantages of this application more apparent and understandable, the specific embodiments of this application will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0064] First, the application scenarios of the embodiments of this application are described. In the embodiments of this application, intelligent methods, devices, equipment, and media for managing aesthetic education curriculum resources are provided, applicable to, but not limited to, the digital construction of aesthetic education curriculum resources in primary and secondary schools, the management of cross-cultural aesthetic education teaching resources for art majors in universities, and the construction of aesthetic education resource databases for off-campus art training institutions.

[0065] In illustrative purposes, the intelligent aesthetic education curriculum resource management method, device, equipment and medium provided in the embodiments of this application can also be applied to other application scenarios such as intelligent management of art popular science resources in the cultural and tourism industry, resource retrieval and adaptation of online aesthetic education platforms, and development of digital aesthetic education resources of art college collections. These are only examples and do not limit the specific application scenarios.

[0066] like Figure 1 As shown, this application provides an intelligent method for managing aesthetic education curriculum resources, which includes:

[0067] S101: Extract cross-cultural semantic features from the input multimodal aesthetic education resources to generate digital resource objects.

[0068] For example, before extracting cross-cultural semantic features, multimodal data preprocessing is performed on the input multimodal aesthetic education resources. Format standardization and content parsing are performed on aesthetic education resources of different modalities, such as visual, auditory, and textual, to remove irrelevant redundant information and retain the core content carriers with artistic aesthetic value and cultural connotation. After preprocessing, cultural symbol recognition is performed on the processed multimodal aesthetic education resources. The system locates various symbolic elements in the resources, such as visual symbols, color symbols, and textual semantic symbols, that carry cross-cultural semantics, clarifying the presentation forms and distribution characteristics of each type of symbol in the multimodal aesthetic education resources.

[0069] Based on cross-cultural semantic parsing rules, deep semantic mining is performed on the identified cultural symbols to extract core information such as cultural connotations, aesthetic paradigm associations, and cross-cultural semantic relationships. Simultaneously, the extracted cross-cultural semantic features are normalized and structured to generate a cross-cultural semantic feature set with hierarchical and relational attributes. This structured cross-cultural semantic feature set is then integrated and encapsulated with the basic digital information of multimodal aesthetic education resources. Following pre-defined resource object data specifications, data hierarchy and relationships are constructed to generate digital resource objects.

[0070] S102: Based on the contextual identifiers of the target teaching scenario, perform sub-graph loading processing on the preset cultural knowledge base to generate an aesthetic rule sub-graph.

[0071] For example, the contextual identifiers of the target teaching scenario are deconstructed, breaking down the core content contained within the identifiers, such as cultural type elements, teaching level elements, and aesthetic guidance elements, to extract a set of core cultural elements that can represent the cultural needs of the target teaching scenario. Based on the set of core cultural elements, the graph partitioning rule library in the preset cultural knowledge base is invoked. Through element-rule matching analysis, the graph partitioning range in the preset cultural knowledge base that matches the set of core cultural elements is determined, and the boundaries of the aesthetic rule knowledge that needs to be loaded are clarified.

[0072] Based on the defined knowledge graph partitioning range, nodes and relationships are filtered and extracted from the complete knowledge graph in the preset cultural knowledge base to obtain key graph content such as cultural aesthetic rule nodes, art paradigm nodes, and cross-cultural relationships related to the core cultural element set. If the core cultural element set contains cross-cultural elements, the multiple cultural sub-graphs extracted need to be merged to integrate the aesthetic rule association information under different cultural backgrounds and eliminate logical conflicts between graphs.

[0073] If the core cultural elements set is a single cultural type, the content of the single-cultural aesthetic rule map extracted through screening is directly retained. The screened or integrated map content undergoes structural and integrity verification, checking the relevance of map nodes, the adaptability of rule content, and the rationality of information hierarchy. After optimizing and adjusting the map content, an aesthetic rule sub-map is generated.

[0074] The aesthetic rule subgraph is obtained using the following formula:

[0075]

[0076] In the formula, Representing the aesthetic rules subgraph Indicates the first A cultural sub-map Indicates the fusion weight coefficient. This indicates a cross-cultural indicator function. Represents a single cultural foundation subgraph. Indicates the number of fused subgraphs;

[0077] S103: Activate the preset agent role configuration library to generate a multi-role agent cluster, and perform collaborative analysis and reasoning on digital resource objects based on the aesthetic rule sub-graph to generate consensus-based art analysis results.

[0078] For example, the system parses the configuration information of various agent roles in the preset agent role configuration library, filters the agent role types that are suitable for the digital resource object analysis task, completes the instantiation operation according to the role activation protocol, assigns data analysis permissions and aesthetic rule sub-graph access range to each role, and integrates the activated agent roles to generate a multi-role agent cluster.

[0079] Each agent role in the multi-role agent cluster retrieves the rule content of the corresponding analysis dimension in the aesthetic rule sub-graph, conducts special analysis and reasoning for digital resource objects and generates analysis results for each dimension, collects all special analysis results to form a preliminary analysis result set, and cross-validates the preliminary analysis result set according to a preset interaction protocol.

[0080] If conflicting conclusions are detected in the preliminary analysis result set, weights are evaluated according to the priority rules in the aesthetic rule subgraph, and the conflicting conclusions are corrected and arbitrated by combining the historical case database in the aesthetic rule subgraph. If no conflicting conclusions are detected, the preliminary analysis result set is directly integrated for consistency. According to the result fusion rules in the aesthetic rule subgraph, the processed analysis results are restructured and information fused to generate consensus-based art analysis results.

[0081] S104: Based on the consensus-based art analysis results, perform structured feedback generation processing to generate structured art evaluation feedback.

[0082] For example, the consensus-based art analysis results are decomposed into dimensions to separate the core analytical content of each dimension, such as the composition conformity analysis conclusion, the cultural color analysis conclusion, the artistic innovation analysis conclusion, and the cultural adaptability analysis conclusion, thus clarifying the art evaluation direction and judgment basis corresponding to the conclusions of each dimension.

[0083] Based on the evaluation standards for aesthetic education curriculum resources, art evaluation indicators are extracted from the core analysis content of each dimension, the hierarchical classification and expression paradigm of the evaluation indicators are determined, and key information that can be transformed into evaluation suggestions from the conclusions of each dimension is sorted out to generate a set of evaluation indicators and a set of suggestions.

[0084] The preset structured feedback template is invoked, and the evaluation indicator set and suggestion information set are filled into the corresponding content areas of the template according to the evaluation dimension arrangement rules, indicator presentation format and suggestion organization method set in the template, thus completing the initial structured organization of the feedback content.

[0085] The initial structured feedback content undergoes format verification and logical consistency checks to correct issues such as disorganized content layout and contradictory expressions. The hierarchical relationship and readability of the evaluation information are optimized, and the final adjustments to the content are used to generate structured art evaluation feedback.

[0086] S105: Structured art evaluation feedback is used as enhanced metadata and bound to the original resources to build an art education curriculum resource library that can be searched from a cultural dimension.

[0087] For example, the structured art evaluation feedback is subjected to cultural dimension metadata extraction operations to select the core information that represents cultural schools, color symbolism, aesthetic paradigms, and cultural adaptability. The selected information is then formatted and semantically annotated according to metadata specifications to generate a cultural semantic metadata dataset with a unified structure.

[0088] Based on the cultural semantic metadata dataset, a hierarchical cultural dimension label is constructed. According to the classification logic of cultural attributes, core branches such as cultural school branches, technique type branches, and aesthetic paradigm branches are established. Specific label nodes are refined under each branch to form a hierarchical cultural dimension label tree.

[0089] The original resources are extended by the metadata layer. The cultural semantic metadata dataset is fully injected into the metadata layer of the original resources to supplement the cultural dimension feature information of the original resources. After the cross-cultural feature enhancement operation is completed, enhanced aesthetic education resources are generated.

[0090] A dynamic indexing method is adopted to establish a mapping between enhanced aesthetic education resources and each tag node in the cultural dimension tag tree. The index structure is optimized based on the association strength between resources and tags, and all enhanced aesthetic education resources and their corresponding index information are integrated to build an aesthetic education course resource library that can be searched by cultural dimension.

[0091] An embodiment of this application provides an intelligent aesthetic education curriculum resource management method, comprising: performing cross-cultural semantic feature extraction processing on input multimodal aesthetic education resources; generating structured digital resource objects by mining the deep cultural symbols and semantic information contained in the resources, providing a standardized data foundation for subsequent analysis; loading a sub-graph of a preset cultural knowledge base based on the contextual identifiers of the target teaching scenario; selecting suitable aesthetic rule knowledge in conjunction with the core cultural elements of the scenario; and constructing an aesthetic rule sub-graph that highly matches the teaching scenario, breaking the limitations of traditional static loading of unified knowledge graphs.

[0092] The pre-defined agent role configuration library is activated to form a multi-role agent cluster. Based on an aesthetic rule sub-graph, this cluster collaboratively analyzes and reasons about digital resource objects from dimensions such as composition, color, art historical relevance, and cultural context. Through cross-validation and conflict arbitration mechanisms, it generates unified and authoritative consensus-based art analysis results. Based on these consensus-based art analysis results, structured feedback generation is performed. Following standardized evaluation dimensions and expression formats, the analysis results are transformed into structured art evaluation feedback containing evaluation indicators, conclusions, and suggestions. This structured art evaluation feedback is used as enhanced metadata and bound to the original multimodal aesthetic education resources. By constructing a cultural dimension tagging system and a dynamic indexing mechanism, an aesthetic education curriculum resource library supporting cultural dimension retrieval is formed, achieving intelligent management and efficient application of aesthetic education resources.

[0093] In one embodiment, collaborative analysis and reasoning are performed on digital resource objects based on aesthetic rule sub-graphs to generate consensus-based art analysis results, including:

[0094] (1) Control the graph analysis agent in the multi-role agent cluster to perform graph rule matching processing on digital resource objects and generate graph conformity analysis results.

[0095] For example, the composition analysis agent accesses the aesthetic rule sub-graph, locates and reads the stored composition rule system, and clarifies the composition standards under different cultural backgrounds. It deconstructs the visual elements of digital resource objects, extracting composition-related features such as the position of the core subject, the distribution of auxiliary elements, the proportion of image segmentation, and the percentage of white space. The extracted composition features are compared one by one with the standards in the composition rule system, marking conforming and non-conforming items, and recording the specific basis for each comparison. Based on the number of conforming items and the severity of non-conforming items, combined with preset fit calculation rules, a composition fit value is obtained. According to the level classification standard corresponding to the fit value, the composition fit level is determined, and the comparison basis, fit value, and level judgment result are integrated to generate a structured composition fit analysis result.

[0096] The composition rules system includes the golden ratio standard under different cultural backgrounds, the principles of symmetry and balance, the norms for visual focus positioning, and the requirements for the use of white space; the composition conformity analysis results include composition feature extraction records, rule comparison basis, conformity value, and conformity level judgment; the conformity calculation rules include the weight of the number of conforming items and the deduction standard for the severity of non-conforming items; the level classification standard includes the level ranges of excellent, good, qualified, and unqualified corresponding to the conformity value.

[0097] (2) Control the color semantic agent in the multi-role agent cluster to perform color symbol mapping processing on digital resource objects and generate cultural color analysis results.

[0098] For example, the color semantic agent retrieves the color symbol rule library from the aesthetic rule sub-map and filters out color symbol entries that match the cultural context of the target teaching scenario. Color sampling and recognition are performed on digital resource objects, distinguishing and recording the color value ranges and distribution areas of the main color, secondary color, and accent color in the image. Each identified color is matched one by one with entries in the color symbol rule library to determine the emotional orientation, cultural connotation, and symbolic meaning of each color in the target cultural context. The semantic superposition effect produced by different color combinations is analyzed to determine whether there are color symbolic conflicts or inconsistencies. Color recognition records, single-color symbol mapping results, and color combination analysis conclusions are integrated to form a cultural color analysis result containing color information, symbolic meaning, and a preliminary judgment of cultural suitability.

[0099] The color symbolism rule base includes the emotional connotations, cultural implications, and symbolic meanings of various colors in different cultural contexts; the target teaching scenario cultural context includes the regional culture, ethnic culture, and contemporary culture attributes corresponding to the teaching scenario; color sampling and recognition includes color sampling point selection, color value detection, and color classification; color combination analysis conclusions include the harmony of color matching and the consistency of symbolic semantics; and cultural color analysis results include color sampling and recognition records, details of single-color symbolic mapping, color combination semantic analysis, and preliminary judgment of cultural adaptation.

[0100] (3) Control the art history association agent in the multi-role agent cluster to perform historical school comparison processing on digital resource objects and generate art innovation analysis results.

[0101] For example, the art history association agent enters the aesthetic rules sub-graph, retrieves and extracts the art history school feature database, and clarifies the core characteristics of each historical period and art school. The digital resource object is then deconstructed for artistic features, extracting core features such as creative techniques, expressive styles, thematic presentation methods, and visual texture. The extracted core features are compared with the features of each school in the art history school feature database to calculate the matching degree of each type of feature, identifying the art school with the highest correlation to the digital resource object. From the perspectives of technical innovation, stylistic integration, and thematic expansion, the unique aspects that distinguish the digital resource object from the associated schools are analyzed, recording the specific manifestations and value of innovative points. By integrating the school feature comparison records, the associated school determination results, and the details of the innovation point analysis, combined with the innovation level assessment criteria, the innovation level is determined, generating the art innovation analysis results.

[0102] The art history school feature database includes the core creative techniques, typical styles, common theme types, representative works, and background of different art schools in various historical periods; the art feature decomposition includes the identification of creative techniques, the classification of styles, the extraction of theme content, and analysis of image texture; the similarity comparison includes the calculation of single feature matching degree and the weighted summation of comprehensive matching degree; the innovation point analysis includes the improvement of techniques, the style integration method, and new thematic perspectives; the innovation level evaluation criteria include the number of innovation points and the level division corresponding to the degree of innovation; the art innovation analysis results include details of art feature extraction, school feature comparison records, determination of related schools, explanation of innovation points, and innovation level.

[0103] (4) Control the cultural context advisor agent in the multi-role agent cluster to perform aesthetic paradigm verification on digital resource objects and generate cultural adaptability analysis results.

[0104] For example, the cultural context consultant agent obtains cultural paradigm verification standards matching the target teaching scenario from the aesthetic rule sub-map, clarifying aesthetic preferences, artistic expression taboos, and norms for the use of cultural symbols within the target cultural context. Cultural elements are identified in digital resource objects, extracting elements with cultural orientation such as graphic symbols, color symbols, pattern styles, and thematic content. Based on the cultural paradigm verification standards, each extracted cultural element is verified to ensure it conforms to the aesthetic requirements of the target cultural context, determining whether there are issues such as misuse of cultural symbols, conflict between expression forms and cultural context, or inconsistencies in value orientation. The creative background and cultural origins of the digital resource objects are traced and analyzed to reconstruct their original cultural context, and compared with the target cultural context to analyze the impact of contextual differences on suitability. The cultural element verification records, problem judgment results, contextual source analysis, and suitability level assessment are integrated, along with improvement suggestions, to generate cultural suitability analysis results.

[0105] The cultural paradigm verification criteria include aesthetic value orientation, artistic expression norms, cultural symbol connotation definition, and artistic expression taboos within the target cultural context; cultural element identification includes graphic symbol extraction, color symbol interpretation, pattern style classification, and thematic content cultural orientation analysis; problem judgment results include types of cultural symbol misuse, points of conflict in expression, and explanations of value orientation deviations; contextual source analysis includes restoration of the creative era background, extraction of original cultural context characteristics, and comparison of differences between the target and original contexts; suitability level assessment includes the level standards corresponding to the degree of suitability; and cultural suitability analysis results include cultural element extraction records, verification basis, suitability level, problem prompts, contextual source report, and improvement suggestions.

[0106] (5) Cross-validate the results of composition conformity analysis, cultural color analysis, artistic innovation analysis and cultural adaptability analysis through a preset interaction protocol to generate a set of validation conclusions.

[0107] For example, the cross-validation rules defined in the preset interaction protocol are invoked to clarify the logical connections, mutual verification standards, and contradiction judgment thresholds among the various analysis results. The composition level and rule fit are extracted from the composition conformity analysis results and logically checked against the cultural element layout adaptability and aesthetic paradigm conformity in the cultural suitability analysis results to determine whether the compositional expression conflicts with cultural suitability requirements. The symbolic meaning and color combination effect of colors are extracted from the cultural color analysis results and cross-verified against the color symbol adaptability in the cultural suitability analysis results to check whether the color semantics are consistent with the target cultural context.

[0108] Extracting relevant artistic styles and innovative point types from the results of the artistic innovation analysis, and verifying their correlation with color style from the cultural color analysis results and technical expression from the composition conformity analysis results, this process determines whether the innovative points are logically consistent with the results of other dimensions of analysis. All consistent and contradictory items in the verification process are recorded, with the specific conflicting content and involved dimensions noted for contradictory items. Finally, the original analysis results, verification process records, and consistent or contradictory markers are integrated to generate a verification conclusion set containing complete verification information.

[0109] The pre-defined interaction protocol includes cross-validation rules, data interaction format, and result recording specifications. The cross-validation rules include the logical correlation dimensions between the analysis results, the judgment criteria for mutual verification, and the quantitative threshold for contradiction judgment. The logical consistency check includes the consistency of conclusion statements, the consistency of judgment criteria, and the correlation of influencing factors. Mutual verification includes semantic correspondence, feature matching degree, and contextual adaptability. The correlation check includes the correlation between innovation points and color style, and the coordination between innovation points and technical performance. The verification conclusion set includes the original text of each analysis result, the cross-validation process record, the list of consistent items, the list of contradictory items, and conflict details.

[0110] (6) When there is a conflict in the set of verification conclusions, arbitration is carried out according to the priority rules in the aesthetic rule subgraph to generate a consensus art analysis result.

[0111] For example, conflict identification is performed on the verification conclusion set to screen out all logically contradictory conclusion pairs, clarifying the analytical dimensions and specific conflict content of each pair. Priority rules are retrieved from the aesthetic rule sub-graph to determine the weight ranking of conclusions across different analytical dimensions and the criteria for judging cultural authority. According to the priority rules, the weights of each conclusion in the conflict pair are calculated, and the conclusion with the higher weight is initially identified as a candidate conclusion. If the weight advantage of a candidate conclusion does not reach a pre-set confidence threshold, cases similar to the current conflict scenario are retrieved from the historical arbitration case database of the aesthetic rule sub-graph, and the arbitration basis, handling methods, and result feedback from these cases are extracted as supporting evidence. The candidate conclusions are revised based on the supporting evidence from these cases. If the supporting evidence from these cases is consistent with the candidate conclusions, the credibility of the conclusions is strengthened; if there are differences, the conclusion statements are adjusted or additional limiting conditions are added. The revised final arbitration conclusions are integrated with the unconflicting conclusions in the verification conclusion set, unifying the expression format, supplementing logical connections, and generating a consensus-based art analysis result.

[0112] The conflict identification process includes screening of contradictory content, extraction of conflict conclusion pairs, and labeling of conflict dimensions. Prioritization rules include weighting of conclusions from different analytical dimensions, criteria for determining cultural authority, procedural steps for conflict arbitration, and standards for setting confidence thresholds. Weight calculation includes basic dimension weights, conclusion credibility coefficients, and comprehensive weight calculations. The historical arbitration case database includes arbitration cases from past conflict scenarios, with each case containing conflict details, arbitration basis, handling results, and feedback on the application of results. Supporting evidence for cases includes the arbitration logic and basis for conclusions from similar cases. The final arbitration conclusion includes the revised conflict conclusion and an explanation of the arbitration basis. The consensus-building analysis results include arbitration process records, the final arbitration conclusion, integrated content of conflict-free conclusions, and an overall logic verification report.

[0113] (7) When there is no conflict in the verification conclusion set, the verification conclusion set is integrated to generate a conflict-free consensus set; based on the result fusion rules in the aesthetic rule subgraph, the conflict-free consensus set is restructured to generate a consensus art analysis result.

[0114] For example, the results of compositional conformity analysis, cultural color analysis, artistic innovation analysis, and cultural adaptability analysis in the verification conclusion set are categorized and organized according to dimensions. Information within the same analytical dimension is grouped together, and duplicate statements are removed. Logical connecting statements between the conclusions of each dimension are added to explain the relationship between compositional expression and cultural adaptability, and the correspondence between color semantics and artistic innovation, thus forming a coherent and conflict-free consensus set.

[0115] The results fusion rules are derived from the aesthetic rule subgraph, clarifying the structural reorganization framework, dimension title specifications, and conclusion presentation format requirements for the conflict-free consensus set. Following the reorganization framework, the content of each dimension in the conflict-free consensus set is filled into the corresponding sections, with standardized font styles, hierarchical identifiers, and data presentation methods. The reorganized content undergoes a completeness check to confirm no information omissions, and logical verification is performed to ensure no implicit contradictions exist between the conclusions of each dimension, resulting in a structurally sound, comprehensive, and logically coherent consensus-based art analysis result.

[0116] The consistency integration process includes dimensional classification, removal of duplicate content, supplementation of logical connections, and verification of content coherence; the conflict-free consensus set includes the analysis conclusions of each dimension after classification, logical connection statements, and content integrity explanations; the result fusion rules include the reorganization framework of the conflict-free consensus set, dimension sorting standards, title naming conventions, conclusion expression format, and integrity and logical verification requirements; the structural reorganization framework includes the dimension arrangement order, section division, and content filling rules; integrity verification includes information item verification and key content confirmation; logical verification includes checking the causal relationship between conclusions and identifying hidden contradictions; and the consensus art analysis results include the conclusions of each dimension after structural reorganization, integrity verification reports, and logical verification results.

[0117] In one embodiment, when conflicts exist in the verification conclusion set, arbitration is performed according to the priority rules in the aesthetic rule subgraph to generate a consensus-based art analysis result, including:

[0118] (1) Extract conflicting pairs of conclusions from the set of verification conclusions.

[0119] For example, a full traversal of the verification conclusion set is performed, examining the logical expression and judgment tendency of each analysis result one by one, identifying conclusions with contradictory expressions and mutually contradictory judgments. The analytical dimensions corresponding to each pair of contradictory conclusions are clarified, and the core judgment basis for each conclusion is recorded. All contradictory conclusions are then organized into pairs to generate a set of conflicting conclusion pairs.

[0120] The validation conclusion set includes the conclusions of each analysis dimension, cross-validation records, and consistency and contradiction markers; the conflict conclusion pair set includes contradictory conclusions, corresponding analysis dimensions, and core judgment criteria.

[0121] (2) Based on the cultural authority weight rules in the aesthetic rule sub-graph, the conflict conclusions are weighted and processed to generate preliminary arbitration results.

[0122] For example, the aesthetic rules subgraph is accessed, and the stored cultural authority weight rules are retrieved. Based on these rules, the cultural authority weight factor corresponding to each conclusion in the conflicting conclusion pair is determined. The weight calculation function is called, substituting the weight factors of each conclusion into the calculation to obtain the weight value of each conclusion. Based on the weight values, the conclusion with higher priority is determined, and a preliminary arbitration result is generated.

[0123] Among them, the cultural authority weighting rules in the aesthetic rules sub-map include the authority levels of different cultural systems and the weighting standards for the conclusions of each analytical dimension; the weight calculation function includes the operation logic of the weight factors and the calculation method of the weight values; the preliminary arbitration results include the conclusions with higher priority and the record of the weight calculation process.

[0124] (3) When the preliminary arbitration result does not meet the preset confidence threshold, the historical case library in the aesthetic rule sub-graph is called to perform similarity matching to generate case supporting evidence.

[0125] For example, the confidence level of the preliminary arbitration result is determined to confirm whether it reaches a preset confidence threshold. If the preliminary arbitration result does not meet the threshold, the aesthetic rules sub-graph is accessed to retrieve the historical case database. The core features of the conflict conclusion pair are compared with the case features in the historical case database to select cases with a high degree of matching with the current conflict scenario. The arbitration basis, handling method, and result feedback from the matching cases are extracted and integrated to form supporting evidence for the cases.

[0126] Among them, the pre-set confidence threshold includes the standard for judging the reliability of the preliminary arbitration result; the historical case library in the aesthetic rules sub-graph includes the case content, arbitration process, and result information of past conflict arbitration scenarios; the similarity matching processing includes the comparison logic of conflict features and case features, and the method of judging the matching degree; the case supporting evidence includes the arbitration basis, handling method, and result feedback of the matched case.

[0127] (4) Using the following formula, the preliminary arbitration result is dynamically revised based on the supporting evidence from the case, and the final arbitration conclusion is generated:

[0128]

[0129] in, This indicates the final arbitration conclusion. Represents the set of conflicting conclusions. Rules indicating the weight of cultural authority This represents the weight calculation function. This represents the confidence level indicator function. This indicates that historical cases support the evidence. Indicates the case correction operator, This represents a dynamic correction function. Indicates the conclusion fusion operator, This represents the set of non-conflicting conclusions.

[0130] For example, a dynamic correction function is invoked, combining the results of the weight calculation function, supporting evidence from historical cases, and a confidence indicator function to dynamically correct the preliminary arbitration result. Then, a conclusion fusion operator is used to integrate the non-conflicting conclusion set to generate the final arbitration conclusion. The dynamic correction function is invoked, taking the results of the weight calculation function for the conflicting conclusion pair set and the cultural authority weighting rule as input. Supporting evidence from historical cases is incorporated, and the preliminary arbitration result is adjusted using a case correction operator. Simultaneously, the confidence indicator function is used to confirm the rationality of the correction direction. Finally, the conclusion fusion operator is invoked to integrate the corrected conclusion with the non-conflicting conclusion set to generate the final arbitration conclusion.

[0131] The dynamic correction function includes the adjustment logic of the preliminary arbitration result and the method of integrating case information; the case correction operator includes the combination rules of case supporting evidence and the preliminary arbitration result; the confidence level indicator function includes the determination logic of the confidence level of the preliminary arbitration result; the conclusion fusion operator includes the integration method of the revised conclusion and the set of non-conflicting conclusions; and the final arbitration conclusion includes the revised conflicting conclusion and the integrated non-conflicting conclusion content.

[0132] (5) The final arbitration conclusion and the non-conflicting conclusions in the set of verification conclusions are merged to generate a consensus-based art analysis result.

[0133] For example, the content of the final arbitration conclusion is organized, while non-conflicting conclusions are extracted from the verification conclusion set. The expression formats of the two types of conclusions are standardized, and logical connections between the conclusions are supplemented to ensure their relevance and coherence. The integrated content undergoes logical consistency verification, correcting any contradictory or disjointed parts. The integrated and verified content is then structured to generate a consensus-based art analysis result.

[0134] Among them, the non-conflicting conclusions in the verification conclusion set include logically consistent and non-contradictory analysis results; the fusion processing includes the unification of conclusion format, the supplementation of logical connections, and the integration of content; the consensus-based art analysis results include the integrated arbitration conclusions and non-conflicting conclusions, logically connected content, and structured expression.

[0135] like Figure 2 As shown, the cultural context advisor agent in the multi-role agent cluster performs aesthetic paradigm verification on digital resource objects, generating cultural compatibility analysis results, including:

[0136] S201: Perform cultural symbol recognition processing on digital resource objects to generate a set of cultural symbols.

[0137] For example, a comprehensive content analysis is performed on digital resource objects to extract culturally relevant content such as graphic elements, color elements, pattern styles, and thematic content. The extracted content is then assessed for its cultural attributes, and elements carrying specific cultural connotations are selected. These elements are then categorized and organized according to cultural type and symbol type to generate a structured set of cultural symbols.

[0138] Among them, digital resource objects include multimodal forms of aesthetic education resource content and corresponding basic information; cultural symbol recognition and processing include cultural content extraction, cultural attribute determination, and symbol classification and organization; cultural symbol sets include elements that carry cultural connotations, the cultural types corresponding to the elements, and symbol types.

[0139] S202: Based on the cultural paradigm rule base in the aesthetic rule subgraph, perform paradigm conformity verification on the cultural symbol set and generate a basic paradigm score.

[0140] For example, the aesthetic rules subgraph is accessed, and the cultural paradigm rule base stored within it is retrieved. Each element in the cultural symbol set is compared one by one with the normative requirements in the cultural paradigm rule base, and elements that conform to the norms are marked as those that do not. The percentage of elements that conform to the norms is calculated, and combined with the degree of deviation of the non-conforming elements, a corresponding basic paradigm score is generated.

[0141] Among them, the cultural paradigm rule base in the aesthetic rules sub-map includes the norms for the use of symbols in different cultural contexts and the requirements for aesthetic expression; the paradigm conformity verification process includes the comparison of symbols and norms, the marking of conformity status, and the statistics of the proportion of conformity; the basic paradigm score includes the proportion of elements that conform to the norms and the score corresponding to the degree of deviation.

[0142] S203: When cross-cultural symbols are detected in the cultural symbol set, the preset cross-cultural symbol conflict library is called to perform symbol conflict verification processing and generate symbol conflict identifiers; based on the symbol conflict identifiers, the symbol weights of the basic paradigm score are dynamically adjusted to generate the adjusted paradigm score.

[0143] For example, a cross-cultural attribute detection is performed on the set of cultural symbols to determine whether cross-cultural symbols exist. If cross-cultural symbols exist, a pre-defined cross-cultural symbol conflict database is invoked to retrieve conflict information of the symbol in the target cultural context. The conflict type and degree of the symbol are determined, and a symbol conflict identifier is generated. Based on the weight adjustment rules corresponding to the symbol conflict identifier, the score of the basic paradigm score is dynamically adjusted to generate an adjusted paradigm score.

[0144] Among them, cross-cultural symbols include symbolic elements that have different connotations in different cultural contexts; the pre-set cross-cultural symbol conflict database includes the conflict type, conflict degree, and corresponding impact description of cross-cultural symbols; symbol conflict verification and processing includes conflict information retrieval and conflict type determination; symbol conflict identification includes the conflict type and conflict degree of cross-cultural symbols; symbol weight dynamic adjustment processing includes the application of weight adjustment rules and score adjustment; and the adjusted paradigm score includes the adjusted score and the basis for adjustment.

[0145] S204: When it is detected that there are no cross-cultural symbols in the cultural symbol set, the basic paradigm score is used as the adjusted paradigm score; cultural context restoration processing is performed on the digital resource objects to generate a context restoration analysis report; the adjusted paradigm score and the context restoration analysis report are comprehensively evaluated to generate cultural adaptability analysis results.

[0146] For example, a cross-cultural attribute test is performed on the set of cultural symbols. If no cross-cultural symbols are found, the basic paradigm score is directly determined as the adjusted paradigm score. The creative background and cultural origin of the digital resource object are traced to restore its original cultural context, and a context restoration analysis report is compiled. Combining the score of the adjusted paradigm and the cultural matching results in the context restoration analysis report, a comprehensive evaluation is conducted to clarify the degree of fit between the digital resource object and the target cultural context, generating a cultural suitability analysis result.

[0147] The cultural context restoration process includes tracing the creative background, extracting the original cultural context, and organizing contextual information; the context restoration analysis report includes the creative background, original cultural context, and matching status with the target context of the digital resource object; the comprehensive evaluation process includes the combined evaluation of scoring and report, and the determination of the degree of fit; the cultural fit analysis results include the degree of fit level, evaluation basis, and relevant explanations.

[0148] In one embodiment, structured art evaluation feedback is bound to the original resources as enhanced metadata to construct an art education curriculum resource library that can be searched along cultural dimensions, including:

[0149] (1) Extract and process the cultural dimension metadata of the structured art evaluation feedback to generate a cultural semantic metadata dataset.

[0150] For example, structured art evaluation feedback is decomposed to identify information related to cultural attributes, including descriptions of cultural schools, technical types, and aesthetic paradigms. This cultural attribute information is then converted into metadata formats to clarify the metadata fields corresponding to each piece of information. The converted metadata is then categorized and organized according to cultural dimensions to generate a structured cultural semantic metadata dataset.

[0151] The structured art evaluation feedback includes evaluation conclusions for each analysis dimension, descriptions of cultural attributes, and suitability determination; the extraction and processing of cultural dimension metadata includes locating cultural attribute information, converting metadata formats, and classifying and organizing it; the cultural semantic metadata dataset includes metadata fields corresponding to cultural dimensions and classified metadata content.

[0152] (2) Based on the cultural semantic metadata dataset, cultural label hierarchy is constructed and processed to generate a cultural dimension label tree, which includes cultural school branches, technique type branches and aesthetic paradigm branches.

[0153] For example, the metadata categories in the cultural semantic metadata set are sorted out to identify core cultural dimension branches, including cultural school branches, technique type branches, and aesthetic paradigm branches. The metadata content under each branch is hierarchically decomposed, dividing it into sub-levels such as first-level tags and second-level tags. The tags at each level are organized according to their subordinate relationships to generate a cultural dimension tag tree with a hierarchical structure.

[0154] The cultural semantic metadata dataset includes classified cultural dimension metadata; the cultural label hierarchy construction process includes dimension branch determination, label hierarchy decomposition, and hierarchy relationship organization; the cultural dimension label tree includes cultural school branches, technique type branches, aesthetic paradigm branches, and hierarchical labels under each branch.

[0155] (3) Perform cross-cultural feature enhancement processing on the original resources, inject the cultural semantic meta dataset into the original resource metadata layer, and generate enhanced aesthetic education resources.

[0156] For example, the original resource is obtained, and its metadata layer is located. The original resource metadata layer is then extended, reserving storage fields for cultural dimension metadata. The content from the cultural semantic metadata set is then filled into the reserved fields one by one, completing the injection of the cultural semantic metadata dataset. The injected metadata layer is then subjected to integrity verification to ensure the validity of the association between the metadata and the original resource, generating enhanced aesthetic education resources.

[0157] The original resources include multimodal aesthetic education resource content and basic metadata; cross-cultural feature enhancement processing includes metadata layer expansion, metadata injection, and integrity verification; the original resource metadata layer includes the basic attribute information of the original resources; and the enhanced aesthetic education resources include the original resource content and the metadata layer after injecting cultural semantic metadata.

[0158] (4) Using the following formula, the enhanced aesthetic education resources are dynamically indexed and associated with the cultural dimension tag tree to construct an aesthetic education curriculum resource library that can be searched by cultural dimension:

[0159]

[0160] in, This indicates a culturally searchable art education curriculum resource library. Represents a cultural dimension label tree. This indicates an enhanced collection of aesthetic education resources. This represents the index aggregation operator. This indicates the dimension index construction function. This represents a resource-tag association function. This represents a specific node in the cultural dimension label tree. This indicates enhanced resources for aesthetic education.

[0161] For example, the resource-tag association function is invoked to associate and match each enhanced aesthetic education resource in the enhanced aesthetic education resource set with a specific node in the cultural dimension tag tree. The dimension index construction function is invoked to build a corresponding resource index for each cultural dimension tag tree node based on the matching results. The index aggregation operator is invoked to integrate the resource indexes corresponding to all nodes. The integrated index is then validated for search effectiveness to ensure that resources can be retrieved via cultural dimension tags, thus constructing an aesthetic education course resource library that can be searched by cultural dimension.

[0162] The enhanced aesthetic education resource set includes multiple enhanced aesthetic education resources; the cultural dimension tag tree includes tag nodes at each level; the dynamic index association processing includes resource-tag matching, dimension index construction, and index aggregation; the aesthetic education course resource library that can be searched by cultural dimension includes the integrated resource index, enhanced aesthetic education resources, and cultural dimension tag tree association relationships.

[0163] In one embodiment, based on the contextual identifiers of the target teaching scenario, a sub-graph loading process is performed on a pre-defined cultural knowledge base to generate an aesthetic rule sub-graph, including:

[0164] (1) Deconstruct the contextual elements of the target teaching scenario to generate a set of core cultural elements.

[0165] For example, the contextual identifiers of the target teaching scenario are obtained, and the context-related content, such as cultural type information, teaching level information, and aesthetic guidance information, is extracted. The extracted content is then filtered for core elements, retaining cultural elements closely related to aesthetic rules. The selected elements are then categorized and organized by attribute to generate a core cultural element set.

[0166] Among them, the contextual identifiers of the target teaching scenario include cultural type information, teaching level information, and aesthetic guidance information corresponding to the target teaching scenario; the deconstruction and processing of contextual elements includes the decomposition of contextual content, the selection of core elements, and the classification and organization of elements; the core cultural element set includes cultural elements closely related to aesthetic rules and the attribute classification of the elements.

[0167] (2) Based on the core cultural elements set, cross-cultural association analysis is performed to generate a cultural association strength matrix.

[0168] For example, each cultural element in the core cultural element set is identified, and the corresponding cultural system is clarified. The types and degrees of cross-cultural associations between different cultural elements are analyzed. The associations of each element are organized in matrix form, and the association attributes between elements are recorded to generate a cultural association strength matrix.

[0169] The core cultural elements set includes cultural elements closely related to aesthetic rules and the corresponding attribute classifications of the elements; the cross-cultural association analysis and processing includes sorting out cultural elements, analyzing the types and degrees of association, and organizing the association status into a matrix; the cultural association strength matrix includes the association types, the degree of association, and the records of association attributes between cultural elements.

[0170] (3) Call the graph partitioning rule library in the preset cultural knowledge base to perform dynamic partitioning of the cultural association strength matrix and generate a sub-graph loading scheme.

[0171] For example, a preset cultural knowledge base is accessed, and the stored graph partitioning rule library is retrieved. Based on the partitioning criteria in the graph partitioning rule library and the element associations presented by the cultural association strength matrix, graph regions matching the core cultural element set are delineated. The sub-graph content and loading priority corresponding to each partition are defined, and a sub-graph loading scheme is generated.

[0172] The pre-defined cultural knowledge base includes a graph partitioning rule library, which includes graph partitioning standards and the correspondence between partitions and subgraphs; dynamic partitioning processing includes dividing graph regions based on association and determining subgraph content; and subgraph loading schemes include matching subgraph content and subgraph loading priority.

[0173] (4) Perform hybrid map construction processing according to the sub-map loading scheme:

[0174] When the core cultural elements set is a single cultural type, load the corresponding cultural foundation sub-graph and generate an aesthetic rule sub-graph.

[0175] When the core cultural element set includes cross-cultural elements, a graph fusion process is performed based on the cultural association strength matrix to generate an aesthetic rule sub-graph.

[0176] For example, a hybrid graph construction process is performed according to the subgraph loading scheme: when the core cultural element set is a single cultural type, the corresponding cultural foundation subgraph is loaded to generate an aesthetic rule subgraph. The type of the core cultural element set is determined, confirming that it is a single cultural type. According to the subgraph loading scheme, the cultural foundation subgraph corresponding to this single cultural type is loaded from the preset cultural knowledge base. The loaded cultural foundation subgraph is subjected to content integrity verification to ensure that it covers the aesthetic rules corresponding to the core cultural element set. The verified subgraph is then structurally optimized to generate an aesthetic rule subgraph.

[0177] The hybrid graph construction process includes selecting the sub-graph loading method based on the type of the core cultural element set and verifying the sub-graph content; the core cultural element set is a single cultural type, which includes elements that contain only one cultural system; the corresponding cultural foundation sub-graph includes aesthetic rule knowledge and cultural paradigm content under a single cultural type; the aesthetic rule sub-graph includes verified and optimized aesthetic rule knowledge and cultural paradigm content for a single cultural type.

[0178] When the core cultural element set includes cross-cultural elements, graph fusion processing is performed based on the cultural association strength matrix to generate an aesthetic rule sub-graph. The type of the core cultural element set is determined, confirming that it contains cross-cultural elements. Based on the degree of element association presented by the cultural association strength matrix, multiple corresponding cultural sub-graphs are retrieved from a pre-set cultural knowledge base. Content conflict checks are performed on the retrieved cultural sub-graphs, resolving rule inconsistencies between different sub-graphs. The content of the checked sub-graphs is integrated according to association strength to generate a structurally unified aesthetic rule sub-graph.

[0179] Among them, the core cultural element set includes cross-cultural elements, which contain elements from at least two cultural systems; the cultural association strength matrix includes the association type and degree of association between cultural elements; the graph fusion processing includes retrieving multiple cultural sub-graphs, content conflict verification, and sub-graph content integration; and the aesthetic rule sub-graph includes integrated cross-cultural aesthetic rule knowledge and cultural paradigm content.

[0180] In one embodiment, the cultural association strength matrix is ​​obtained using the following formula, including:

[0181]

[0182] in, Represents the cultural association strength matrix. Indicates the collection of core cultural elements. This indicates a pre-defined cross-cultural association rule base. Representation of element-rule interaction operators, This represents the matrix processing function.

[0183] For example, the core cultural element set and a preset cross-cultural association rule base are obtained. The core cultural element set corresponding to the target teaching scenario is determined, and the content of each cultural element contained therein is clarified. A preset resource storage area is accessed, and the preset cross-cultural association rule base is retrieved to confirm the cross-cultural association rule content covered by the rule base. The element-rule interaction operator is invoked to perform interactive operations on the core cultural element set and the preset cross-cultural association rule base. The element-rule interaction operator is started, with the core cultural element set as the first part of the operation input and the preset cross-cultural association rule base as the second part. According to the operation logic corresponding to the element-rule interaction operator, each element in the core cultural element set is matched with the corresponding rule in the preset cross-cultural association rule base to obtain the interaction operation result between the elements and the rules.

[0184] The matrix processing function is invoked to perform matrix processing on the result of the element-rule interaction operation, generating a cultural association strength matrix. The matrix processing function is started by inputting the result of the element-rule interaction operation. Following the organizational logic corresponding to the matrix processing function, the association information in the interaction operation result is arranged in a matrix structure. The matrix construction is then complete, yielding the cultural association strength matrix.

[0185] The core cultural element set includes cultural elements corresponding to the target teaching scenario; the pre-set cross-cultural association rule base includes association rules and attribute descriptions between different cultural elements; and the pre-set resource storage area includes space for storing the pre-set cross-cultural association rule base. The element-rule interaction operator includes the matching logic and operation process between core cultural elements and cross-cultural association rules; the interaction operation results include the matching relationship between core cultural elements and corresponding rules, and association feature information. The matrix processing function includes the matrix arrangement logic of the interaction operation results and the organization rules of the matrix structure; the cultural association strength matrix includes association information between core cultural elements and association features presented in matrix form.

[0186] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0187] In one embodiment, such as Figure 3As shown, this application also provides an intelligent aesthetic education curriculum resource management device 300, which includes:

[0188] Feature extraction module 301 is used to perform cross-cultural semantic feature extraction processing on the input multimodal aesthetic education resources to generate digital resource objects;

[0189] The graph loading module 302 is used to load subgraphs from a pre-set cultural knowledge base based on contextual identifiers of the target teaching scenario, and generate aesthetic rule subgraphs.

[0190] The collaborative reasoning module 303 is used to activate the preset agent role configuration library to generate a multi-role agent cluster, and to perform collaborative analysis and reasoning on digital resource objects based on the aesthetic rule sub-graph to generate consensus art analysis results.

[0191] Feedback generation module 304 is used to generate structured feedback based on consensus-based art analysis results, and generate structured art evaluation feedback.

[0192] The resource library construction module 305 is used to bind structured art evaluation feedback as enhanced metadata with the original resources to build an art education curriculum resource library that can be searched from a cultural dimension.

[0193] Specifically, the feature extraction module 301 performs multimodal content parsing on the input multimodal aesthetic education resources, removing redundant information and retaining the core content with cultural connotations. It locates symbolic elements in the resources that carry cross-cultural semantics and mines the cultural connotations and related features corresponding to these elements. The extracted cross-cultural semantic features are then structured and organized, fused and encapsulated with the basic information of the multimodal aesthetic education resources to generate a digital resource object.

[0194] Among them, multimodal aesthetic education resources include aesthetic education content in the form of visual, auditory, and textual forms; cross-cultural semantic feature extraction and processing includes multimodal content parsing, cultural symbol mining, and feature structured organization; digital resource objects include structured cross-cultural semantic features and basic information of multimodal aesthetic education resources.

[0195] The graph loading module 302 deconstructs the contextual markers of the target teaching scenario to extract a core cultural element set. Based on this core cultural element set, it searches a preset cultural knowledge base and selects suitable graph partitions. Depending on the type of the core cultural element set, it loads the corresponding cultural foundation sub-graph or merges multiple cultural sub-graphs. After content verification and structural optimization, it generates an aesthetic rule sub-graph.

[0196] The contextual identifiers for the target teaching scenario include information such as the cultural type and teaching orientation of the target teaching scenario. The pre-set cultural knowledge base includes aesthetic rule maps from different cultural backgrounds. The sub-map loading process includes element deconstruction, map partitioning and filtering, and sub-map loading and fusion. The aesthetic rule sub-map includes aesthetic rules and cultural paradigms adapted to the target teaching scenario.

[0197] Collaborative reasoning module 303 parses the agent role information in the preset agent role configuration library, selects suitable agent roles, instantiates and activates them, and integrates them to form a multi-role agent cluster. Each agent role retrieves the rule content of the corresponding dimension in the aesthetic rule sub-graph to conduct specialized analysis and reasoning on digital resource objects. The analysis results of each dimension are collected for cross-validation, conflicting conclusions are handled, and the results are integrated to generate a consensus-based art analysis result.

[0198] The system includes a pre-defined proxy role configuration library containing proxy role configuration information for types such as composition analysis and color semantics; a multi-role proxy cluster containing various instantiated and activated analysis proxy roles; collaborative analysis and reasoning processing including specialized analysis, cross-validation, and conflict arbitration; and consensus-based art analysis results including analysis conclusions from various dimensions and a unified integrated result.

[0199] The feedback generation module 304 breaks down the conclusions of each dimension in the consensus-based art analysis results, extracting the corresponding art evaluation indicators and suggestions. It then calls a pre-defined structured feedback template, filling in the evaluation indicators and suggestions according to the template specifications. The filled content undergoes format validation and logical optimization to generate structured art evaluation feedback.

[0200] The consensus-based art analysis results include art analysis conclusions from various dimensions; the structured feedback generation and processing includes evaluation index extraction, template filling, and content optimization; and the structured art evaluation feedback includes standardized evaluation indicators, conclusions, and recommendations.

[0201] Module 305, the resource repository construction module, extracts cultural dimension metadata from structured art evaluation feedback to form a cultural semantic metadata dataset. It then constructs a cultural dimension tag tree and injects the cultural semantic metadata dataset into the metadata layer of the original resources to generate enhanced aesthetic education resources. Finally, it establishes a dynamic index link between the enhanced aesthetic education resources and the cultural dimension tag tree, integrating the index with resource content to construct an aesthetic education curriculum resource repository that can be searched based on cultural dimensions.

[0202] The structured art evaluation feedback includes standardized art evaluation content; enhanced metadata includes extracted cultural dimension metadata; binding processing includes metadata injection and index association; and the cultural dimension searchable art education curriculum resource library includes enhanced art education resources, cultural dimension tag trees, and dynamic indexes.

[0203] Collaborative reasoning module 303 is also used for:

[0204] Control the graph analysis agent in the multi-role agent cluster to perform graph rule matching processing on digital resource objects and generate graph conformity analysis results;

[0205] Control the color semantic agent in the multi-role agent cluster to perform color symbol mapping on digital resource objects and generate cultural color analysis results;

[0206] Control the art history association agent in the multi-role agent cluster to perform historical school comparison processing on digital resource objects and generate artistic innovation analysis results;

[0207] The cultural context advisor agent in the multi-role agent cluster is controlled to perform aesthetic paradigm verification on digital resource objects and generate cultural compatibility analysis results.

[0208] The results of composition conformity analysis, cultural color analysis, artistic innovation analysis, and cultural adaptability analysis are cross-validated through a preset interactive protocol to generate a set of validation conclusions.

[0209] When there are conflicts in the verification conclusion set, arbitration is carried out according to the priority rules in the aesthetic rule subgraph to generate a consensus art analysis result.

[0210] When there are no conflicts in the verification conclusion set, the verification conclusion set is integrated to generate a conflict-free consensus set; based on the result fusion rules in the aesthetic rule subgraph, the conflict-free consensus set is restructured to generate a consensus-based art analysis result.

[0211] Collaborative reasoning module 303 is also used for:

[0212] Extract conflicting pairs of conclusions from the set of verification conclusions;

[0213] Based on the cultural authority weight rules in the aesthetic rule subgraph, the conflicting conclusions are weighted and processed to generate a preliminary arbitration result.

[0214] When the preliminary arbitration result does not meet the preset confidence threshold, the historical case library in the aesthetic rule sub-graph is called to perform similarity matching and generate supporting evidence for the case.

[0215] The following formula is used to dynamically revise the preliminary arbitration result based on supporting case evidence, generating the final arbitration conclusion:

[0216]

[0217] in, This indicates the final arbitration conclusion. Represents the set of conflicting conclusions. Rules indicating the weight of cultural authority This represents the weight calculation function. This represents the confidence level indicator function. This indicates that historical cases support the evidence. Indicates the case correction operator, This represents a dynamic correction function. Indicates the conclusion fusion operator, Represents the set of non-conflicting conclusions;

[0218] The final arbitration conclusion is merged with the non-conflicting conclusions in the set of verification conclusions to generate a consensus-based art analysis result.

[0219] Collaborative reasoning module 303 is also used for:

[0220] Perform cultural symbol recognition processing on digital resource objects to generate a set of cultural symbols;

[0221] Based on the cultural paradigm rule base in the aesthetic rule subgraph, the cultural symbol set is subjected to paradigm conformity verification processing to generate a basic paradigm score.

[0222] When cross-cultural symbols are detected in the cultural symbol set, a preset cross-cultural symbol conflict database is called to perform symbol conflict verification and generate symbol conflict identifiers; based on the symbol conflict identifiers, the basic paradigm score is dynamically adjusted in terms of symbol weights to generate an adjusted paradigm score.

[0223] When it is detected that there are no cross-cultural symbols in the cultural symbol set, the basic paradigm score is used as the adjusted paradigm score; cultural context restoration processing is performed on the digital resource objects to generate a context restoration analysis report; the adjusted paradigm score and the context restoration analysis report are comprehensively evaluated to generate cultural adaptability analysis results.

[0224] The repository building module 305 is also used for:

[0225] Cultural dimension metadata is extracted and processed from structured art evaluation feedback to generate a cultural semantic metadata dataset;

[0226] Based on the cultural semantic meta-dataset, a cultural label hierarchy is constructed to generate a cultural dimension label tree, which includes branches of cultural schools, techniques, and aesthetic paradigms.

[0227] The original resources are enhanced with cross-cultural features, and the cultural semantic metadata is injected into the metadata layer of the original resources to generate enhanced aesthetic education resources.

[0228] Using the following formula, we can dynamically index and associate enhanced aesthetic education resources with cultural dimension tag trees to construct an aesthetic education course resource library that can be searched along the cultural dimension:

[0229]

[0230] in, This indicates a culturally searchable art education curriculum resource library. Represents a cultural dimension label tree. This indicates an enhanced collection of aesthetic education resources. This represents the index aggregation operator. This indicates the dimension index construction function. This represents a resource-tag association function. This represents a specific node in the cultural dimension label tree. This indicates enhanced resources for aesthetic education.

[0231] In one embodiment, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0232] In one embodiment, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the above-described method embodiments.

[0233] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0234] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. An intelligent method for managing aesthetic education curriculum resources, characterized in that: The method includes: Cross-cultural semantic feature extraction is performed on the input multimodal aesthetic education resources to generate digital resource objects; Based on the contextual identifiers of the target teaching scenario, a sub-graph loading process is performed on the pre-set cultural knowledge base to generate an aesthetic rule sub-graph. The preset agent role configuration library is activated to generate a multi-role agent cluster. Based on the aesthetic rule sub-graph, the digital resource object is subjected to collaborative analysis and reasoning to generate a consensus-based art analysis result. Based on the consensus-based art analysis results, structured feedback generation processing is performed to generate structured art evaluation feedback; The structured art evaluation feedback is used as enhanced metadata and bound to the original resources to construct an art education curriculum resource library that can be searched from a cultural dimension.

2. The intelligent aesthetic education curriculum resource management method according to claim 1, characterized in that, The process of performing collaborative analysis and reasoning on the digital resource objects based on the aesthetic rule sub-graphemum to generate consensus-based art analysis results includes: The graph analysis agent in the multi-role agent cluster is controlled to perform graph rule matching processing on the digital resource object and generate graph conformity analysis results. The color semantic agent in the multi-role agent cluster is controlled to perform color symbol mapping processing on the digital resource object to generate cultural color analysis results. The art history association agent in the multi-role agent cluster is controlled to perform historical genre comparison processing on the digital resource object and generate artistic innovation analysis results. The cultural context advisor agent in the multi-role agent cluster is controlled to perform aesthetic paradigm verification on the digital resource object and generate cultural compatibility analysis results. The results of the composition conformity analysis, the cultural color analysis, the artistic innovation analysis, and the cultural adaptability analysis are cross-validated using a preset interaction protocol to generate a set of verification conclusions. When there are conflicts in the set of verification conclusions, arbitration is carried out according to the priority rules in the aesthetic rule subgraph to generate the consensus art analysis result. When there are no conflicts in the set of verification conclusions, the set of verification conclusions is integrated to generate a conflict-free consensus set; based on the result fusion rules in the aesthetic rule subgraph, the conflict-free consensus set is restructured to generate the consensus art analysis result.

3. The intelligent aesthetic education curriculum resource management method according to claim 2, characterized in that, When conflicts exist in the verification conclusion set, arbitration is performed according to the priority rules in the aesthetic rule subgraph to generate the consensus art analysis result, including: Extract mutually contradictory conflicting conclusion pairs from the set of verification conclusions; Based on the cultural authority weight rules in the aesthetic rule sub-graph, the conflicting conclusions are weighted and processed to generate a preliminary arbitration result. When the preliminary arbitration result does not meet the preset confidence threshold, the historical case library in the aesthetic rule sub-graph is invoked for similarity matching to generate supporting evidence for the case. Using the following formula, the preliminary arbitration result is dynamically revised based on the supporting evidence in the case to generate the final arbitration conclusion: in, This indicates the final arbitration conclusion. Represents the set of conflicting conclusions. Rules indicating the weight of cultural authority This represents the weight calculation function. This represents the confidence level indicator function. This indicates that historical cases support the evidence. Indicates the case correction operator, This represents a dynamic correction function. The conclusion fusion operator is represented. Represents the set of non-conflicting conclusions; The final arbitration conclusion is merged with the non-conflicting conclusions in the set of verification conclusions to generate the consensus art analysis result.

4. The intelligent aesthetic education curriculum resource management method according to claim 2, characterized in that, The cultural context advisor agent in the multi-role agent cluster controls the aesthetic paradigm verification of the digital resource object, generating cultural compatibility analysis results, including: The digital resource objects are subjected to cultural symbol recognition processing to generate a cultural symbol set; Based on the cultural paradigm rule library in the aesthetic rule sub-graph, the cultural symbol set is subjected to paradigm conformity verification processing to generate a basic paradigm score. When cross-cultural symbols are detected in the cultural symbol set, a preset cross-cultural symbol conflict database is invoked to perform symbol conflict verification processing and generate a symbol conflict identifier; based on the symbol conflict identifier, the basic paradigm score is dynamically adjusted in terms of symbol weight to generate an adjusted paradigm score. When it is detected that the cultural symbol set does not contain cross-cultural symbols, the basic paradigm score is used as the adjusted paradigm score; the digital resource object is subjected to cultural context restoration processing to generate a context restoration analysis report; the adjusted paradigm score and the context restoration analysis report are comprehensively evaluated to generate the cultural adaptability analysis result.

5. The intelligent aesthetic education curriculum resource management method according to claim 1, characterized in that, The process of binding the structured art evaluation feedback as enhanced metadata with the original resources to construct an art education curriculum resource library that can be searched from a cultural dimension includes: The structured art evaluation feedback is processed by extracting cultural dimension metadata to generate a cultural semantic metadata dataset. Based on the cultural semantic metadata dataset, a cultural label hierarchy is constructed to generate a cultural dimension label tree, which includes cultural school branches, technique type branches, and aesthetic paradigm branches. The original resources are subjected to cross-cultural feature enhancement processing, and the cultural semantic metadata is injected into the original resource metadata layer to generate enhanced aesthetic education resources. Using the following formula, the enhanced aesthetic education resources are dynamically indexed and associated with the cultural dimension tag tree to construct the aesthetic education course resource library that can be searched by cultural dimension: in, This indicates a culturally searchable art education curriculum resource library. A label tree representing cultural dimensions. This indicates an enhanced collection of aesthetic education resources. This represents the index aggregation operator. This indicates the dimension index construction function. This represents a resource-tag association function. This represents a specific node in the cultural dimension label tree. This indicates enhanced resources for aesthetic education.

6. The intelligent aesthetic education curriculum resource management method according to claim 1, characterized in that, The contextual identifier based on the target teaching scenario performs sub-graph loading processing on the preset cultural knowledge base to generate an aesthetic rule sub-graph, including: The contextual identifiers of the target teaching scenario are deconstructed to generate a set of core cultural elements; Based on the core cultural element set, cross-cultural association analysis is performed to generate a cultural association strength matrix. The graph partitioning rule library in the preset cultural knowledge base is invoked to dynamically partition the cultural association strength matrix and generate a sub-graph loading scheme. Perform hybrid map construction processing according to the sub-map loading scheme: When the core cultural element set is a single cultural type, the corresponding cultural foundation sub-graph is loaded to generate the aesthetic rule sub-graph. When the core cultural element set includes cross-cultural elements, a graph fusion process is performed based on the cultural association strength matrix to generate the aesthetic rule sub-graph.

7. The intelligent aesthetic education curriculum resource management method according to claim 1, characterized in that, The cultural association strength matrix is ​​obtained using the following formula, including: in, Represents the cultural association strength matrix. Indicates the collection of core cultural elements. This indicates a pre-defined cross-cultural association rule base. Representation of element-rule interaction operators, This represents the matrix processing function.

8. An intelligent aesthetic education curriculum resource management device, characterized in that, The device includes: The feature extraction module is used to perform cross-cultural semantic feature extraction processing on the input multimodal aesthetic education resources to generate digital resource objects; The graph loading module is used to load subgraphs from a pre-set cultural knowledge base based on contextual identifiers of the target teaching scenario, and generate aesthetic rule subgraphs. The collaborative reasoning module is used to activate the preset agent role configuration library to generate a multi-role agent cluster, and to perform collaborative analysis and reasoning on the digital resource object according to the aesthetic rule sub-graph to generate a consensus art analysis result. The feedback generation module is used to perform structured feedback generation processing based on the consensus art analysis results, and generate structured art evaluation feedback. The resource library construction module is used to bind the structured art evaluation feedback as enhanced metadata with the original resources to build an art education course resource library that can be searched from a cultural dimension.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the intelligent aesthetic education curriculum resource management method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the intelligent aesthetic education curriculum resource management method according to any one of claims 1 to 7.

Images