Method and system for constructing inheriting path of folk song of Yi nationality

Abstract

The invention belongs to the field of culture inheritance modeling, and provides a Yi nationality folk song inheritance path construction method and system, and the method comprises the steps: obtaining multi-source culture element data of Yi nationality folk songs, and analyzing the culture element data into a plurality of semantic element nodes; constructing a semantic map of the Yi nationality folk song based on the semantic element nodes; performing clustering processing on the semantic map to generate a plurality of semantic association clusters, and defining each inheritance path as a structural unit comprising the various nodes; according to the learner portrait information, constructing an adaptability function to evaluate the matching degree between each inheritance path and the learner, selecting the inheritance path with the maximum matching degree, and generating a personalized Yi nationality folk song inheritance path; collecting behavior data of the learner at each node, and feeding back and updating the function value to the path structure; and archiving the finally constructed inheritance path and the learner behavior result, and generating a path resume and inheritance track map.

Description

Technical Field

[0001] The present invention belongs to the field of cultural inheritance modeling, and particularly relates to a method and system for constructing an inheritance path of Yi ethnic group folk songs. Background Art

[0002] Currently, for the protection and inheritance of Yi ethnic group folk songs, the following several methods are mainly adopted: First, static archiving is carried out in the form of audio and video recordings to form a digital resource library; second, phased exhibitions are carried out relying on folk activities or intangible cultural heritage exhibitions; third, courses are taught and vocal music training is carried out by universities or cultural institutions. However, the above methods generally have the following problems: First of all, the existing inheritance methods mainly focus on content preservation, lacking a path structure design oriented to the characteristics of learners, resulting in a lack of hierarchy and adaptability in the learning process, and it is difficult to effectively support the individualized and continuous inheritance goals.

[0003] Secondly, although Yi ethnic group folk song teaching modules are introduced in some university courses, they mostly exist in the form of elective courses or special courses. The teaching paths are not systematic and lack in-depth integration with the original body culture, singing context and semantic structure of the folk songs, making it difficult to achieve the synchronous development of culture and ability.

[0004] Thirdly, there is a lack of dynamic evaluation mechanism and behavior feedback mechanism in the existing teaching and inheritance process. The performance of learners in different links cannot be captured and archived in time, resulting in the disconnection of teaching strategies and path recommendations, and it is difficult to form a closed-loop inheritance system.

[0005] In addition, the content of Yi ethnic group folk songs involves many special elements such as Yi ethnic group dialects, cultural etiquette and multi-voice singing methods, and the learning difficulty is relatively high. Without targeted adaptation strategies and path adjustment mechanisms, it is very easy for learners to give up halfway, affecting the continuity of inheritance. Summary of the Invention

[0006] In order to solve the problems in the prior art, the present invention provides a method for constructing an inheritance path of Yi ethnic group folk songs, including the following steps: Step S10, obtaining multi-source cultural element data of Yi ethnic group folk songs, where the cultural element data includes audio data, lyric texts, singing contexts, festival backgrounds, singer information and Yi ethnic group language translation content, and parsing the cultural element data into multiple semantic element nodes; Step S20, constructing a semantic graph of Yi ethnic group folk songs based on the semantic element nodes, where each node in the graph represents a folk song knowledge unit, and the connection edges between the nodes represent logical dependency relationships; Step S30, performing clustering processing on the semantic graph to generate multiple semantic association clusters, and constructing an inheritance node cluster including native nodes, education nodes, practice nodes and evaluation nodes according to the semantic association clusters, and defining each inheritance path as a structural unit including the above various types of nodes; Step S40: Based on the learner profile information, construct an adaptability function to evaluate the matching degree between each inheritance path and the learner, select the inheritance path with the highest matching degree, and generate a personalized Yi ethnic folk song inheritance path. Step S50: During the execution of the inheritance path, collect the learner's behavior data at each node, including melody restoration degree, semantic understanding degree, and practice completion degree indicators, calculate the evaluation function value of each node, and feedback and update the function value to the path structure. Step S60: Archive the finally constructed inheritance path and the learner's behavior results to generate a path resume and an inheritance trajectory map.

[0007] Further, the step S10 includes: Step S101: Collect the original singing data of Yi ethnic folk songs, including audio data and video materials of multiple voices, and label the corresponding singer, region, and pedigree information. Step S102: Use a speech recognition system to transcribe the audio data into lyric texts, and have bilingual personnel perform Yi language semantic annotation and translation. Step S103: Based on field investigations, record the singing context of folk songs and classify and code them using multi-level tags. Step S104: Extract the festival background by combining the Yi calendar and local chronicles, and set festival types and cultural embedding level tags for each folk song. Step S105: Construct a singer information file, collecting their name, gender, age, and teacher-student pedigree. Step S106: Establish a nested structure of Yi language paragraphs, syllable pinyin, and corresponding Chinese interpretations in the lyrics. Step S107: Structurally represent the collected data as a set of semantic element nodes, and record the context attributes and label information.

[0008] Further, the step S20 includes: Step S201: Define a semantic graph model using a heterogeneous graph structure, including various types such as lyric nodes, melody nodes, festival nodes, and singer nodes. Step S202: Number the semantic element nodes and establish a node attribute table to record their types, semantic labels, and cultural level information. Step S203: Invoke an association rule mining algorithm to calculate the semantic association degree between nodes based on the node occurrence frequency and co-occurrence frequency, and construct a connection edge when the association degree is greater than the set threshold. Step S204: Perform knowledge transfer and manual completion operations on isolated nodes. Step S205: Save the semantic graph in a graph database or RDF triple format.

[0009] Furthermore, the step S30 includes: Step S301: Extract the attributes of all nodes in the semantic graph, and construct node vectors including word embeddings, structural encodings, function labels, and context levels; Step S302: Cluster the node vectors using a graph clustering algorithm; Step S303: Divide the nodes in the clustering result into native nodes, educational nodes, practice nodes, and evaluation nodes; Step S304: Set a standard inheritance path structure unit, including at least one native node, one educational node, one practice node, and one evaluation node; Step S305: Combine the path structures based on the node connection relationships, generate path numbers, and store them in the path library.

[0010] Furthermore, the step S40 includes: Step S401: Collect the ethnic identity, language ability, vocal music foundation, preference labels, and historical learning behaviors of the learner, and construct a multi-dimensional learner portrait vector; Step S402: Extract the content attributes, language types, rhythm features, and teaching difficulty information in the path structure, and construct a path attribute vector; Step S403: Calculate the fitness scores between the learner portrait and each path using a weighted cosine similarity function; Step S404: Select the path with the highest fitness score as the personalized inheritance path for the current learner; Step S405: Store the binding relationship between the learner and the path, and synchronize it to the system database.

[0011] Furthermore, the step S50 includes: Step S501: Configure a node behavior collection module before path execution, and collect melody imitation singing, answer understanding, practice record, and scoring feedback behavior data according to the node type; Step S502: Calculate the standardized index values for the collected data, and the indexes include melody restoration degree, semantic understanding degree, and practice completion degree; Step S503: Set a weighted evaluation function, and integrate multiple indexes to calculate the comprehensive score of the node; Step S504: Bind the evaluation score to the path structure, and adjust the node recommendation weight or insert supplementary nodes according to the score.

[0012] Furthermore, the step S60 includes: Step S601: Create a path resume data structure to record the learner identifier, path number, node execution order, completion status, scoring value, and timestamp; Step S602, archive the path resume in a structured format into the path database; Step S603, construct an inheritance trajectory map, with the nodes being the inheritance nodes in the resume and the edges being the execution order relationships between the nodes, and attaching execution time and status attributes to the edges; Step S604, visualize the learning path based on the trajectory map, identify learning bottlenecks, and feedback the trajectory data to the path recommendation module.

[0013] On the other hand, the present invention also provides a system for constructing an inheritance path of Yi ethnic group folk songs, including the following modules: A multi-source cultural element collection module, which is used to obtain multi-source cultural element data of Yi ethnic group folk songs. The cultural element data includes audio data, lyric texts, singing contexts, festival backgrounds, singer information, and Yi language translation content, and parse the cultural elements into multiple semantic element nodes; A semantic map construction module, which is used to construct a semantic map of Yi ethnic group folk songs based on the semantic element nodes. Each node in the semantic map represents a folk song knowledge unit, and the connection edges between the nodes represent logical dependency relationships; A node clustering and path construction module, which is used to perform clustering processing on the semantic map to generate multiple semantic association clusters, and construct an inheritance node cluster including native nodes, education nodes, practice nodes, and evaluation nodes according to the semantic association clusters, and define each inheritance path as a structural unit including the above various types of nodes; A path matching and recommendation module, which is used to construct an adaptability function according to the learner portrait information to evaluate the matching degree between each inheritance path and the learner, select the inheritance path with the largest matching degree, and generate a personalized inheritance path of Yi ethnic group folk songs; A behavior collection and evaluation module, which is used to collect the behavior data of learners at each node during the execution of the inheritance path, including melody restoration degree, semantic understanding degree, and practice completion degree indicators, calculate the evaluation function values of each node, and feedback and update the function values to the path structure; A path archiving and trajectory map module, which is used to archive the finally constructed inheritance path and the learner behavior results to generate a path resume and an inheritance trajectory map.

[0014] Further, the multi-source cultural element collection module includes: An audio collection unit, which is used to collect audio and video materials of Yi ethnic group folk songs including multiple voices, and label singer, region, and pedigree information; A speech transcription and translation unit, which is used to transcribe the audio into lyric texts by using a speech recognition system, and perform Yi language annotation and translation by bilingual personnel; A context annotation unit, which is used to record the singing context based on the field survey results and perform multi-level label encoding; A festival annotation unit for extracting festival backgrounds and establishing cultural level tags by combining the Yi calendar and local literature; A singer file - building unit for constructing a singer information file and recording their individual attributes and pedigree; A Yi language nested - structure unit for establishing the pinyin and interpretation structure of Yi language paragraphs in lyrics; A node generation unit for structuring data into a set of semantic element nodes and generating attribute tags.

[0015] Furthermore, the semantic graph construction module includes: A graph - structure definition unit for establishing a graph model containing node types of lyrics, melody, festival, and singer in a heterogeneous graph manner; A node - attribute encoding unit for numbering each node and establishing an attribute table of its type, label, and cultural level; A semantic - edge generation unit for calling an association - rule mining algorithm, calculating the semantic association degree between nodes according to the occurrence frequency and co - occurrence frequency, and generating connection edges when the association degree exceeds a set threshold; An isolated - node processing unit for performing knowledge transfer or manual completion on unconnected isolated nodes; A graph - storage unit for saving the constructed semantic graph in the form of a graph database or RDF triples.

[0016] Furthermore, the node clustering and path construction module includes: A node - vector generation unit for extracting the attributes of all nodes in the graph and constructing a node vector containing word embedding, structure encoding, function label, and context level; A clustering - analysis unit for clustering node vectors using a graph - clustering algorithm; A node - classification unit for dividing the clustering results into native nodes, educational nodes, practice nodes, and evaluation nodes; A path - template setting unit for setting a standard inheritance - path structure template, including at least one native node, one educational node, one practice node, and one evaluation node; A path - combination and numbering unit for generating a path structure by combining nodes based on connection relationships and assigning path numbers to be stored in a path library.

[0017] Furthermore, the path matching and recommendation module includes: A portrait - modeling unit for collecting learner's ethnic group, language ability, vocal music foundation, learning preferences, and historical behavior data to construct a learner portrait vector; A path - attribute modeling unit for extracting content features, language types, rhythm styles, and teaching difficulties in the path to construct a path - attribute vector; An adaptation degree calculation unit, which is used to calculate the matching scores between the learner and each path by using a weighted cosine similarity function; A path selection unit, which is used to select the path with the highest adaptation degree score as the personalized recommendation path; A path binding unit, which is used to bind the learner to the selected path and write the binding relationship into the system database.

[0018] Further, the behavior collection and evaluation module includes: A behavior collection unit, which is used to configure a node behavior collection mechanism before path execution and collect vocal imitation audio, answer texts, performance records and scoring data according to node types; An index calculation unit, which is used to calculate standardized index values for the collected behavior data, including melody restoration degree, semantic understanding degree and practice completion degree; An evaluation function unit, which is used to set a weighted evaluation function and calculate the comprehensive score of the node according to multiple indexes; A path adjustment unit, which is used to feedback the evaluation result to the path structure and adjust the node recommendation weight or add supplementary nodes according to the score.

[0019] Further, the path archiving and trajectory map module includes: A resume structure construction unit, which is used to create a path resume data structure including learner identification, path number, node execution order, completion status, scoring result and time stamp; A resume archiving unit, which is used to save the path resume in a structured format to the path database; A map generation unit, which is used to construct an inheritance trajectory map according to the resume data. The nodes in the map represent path nodes, and the edges represent execution order and status attributes; A map visualization unit, which is used to display the trajectory map as an interactive graph, identify path bottlenecks and feedback them to the path recommendation module.

[0020] The present invention realizes the organic integration among the teaching content of Yi ethnic group folk songs, cultural semantics, practical application and learning evaluation by constructing an inheritance path structure including native nodes, educational nodes, practical nodes and evaluation nodes, overcomes the problems of fragmented teaching content and lack of systematicness in the traditional inheritance mode, and improves the integrity and logic of the inheritance process.

[0021] The present invention introduces a learner portrait and an adaptation function, and dynamically generates a personalized inheritance path according to the learner's language ability, vocal music foundation and learning preferences, which can significantly improve the learning efficiency and content acceptance degree, and enhance the pertinence and intelligent level of path recommendation.

[0022] By collecting behavior data during the path execution process and conducting evaluation and feedback, the present invention realizes the dynamic adjustment and resume archiving of the path structure, while constructing a visualized inheritance trajectory map, providing an operable quantitative basis for teaching management, system optimization and cultural research, and having good adaptability, scalability and promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0024] Figure 1 is a flowchart of the method of the present invention; Figure 2 is a system block diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] The following will make a preferred description of the invention in combination with the drawings and the specific embodiments.

[0026] The present embodiment solves the above problems through the following steps: In one embodiment, referring to Figure 1 , the present invention provides a method for constructing an inheritance path of Yi ethnic group folk songs, which is used to systematically establish a personalized inheritance path covering multiple dimensions such as original resources, educational content, practice scenarios and evaluation feedback in the context of the digitization of intangible culture, so as to realize the dynamic, accurate and continuous inheritance of Yi ethnic group folk songs in college teaching and regional communities. The method specifically includes the following steps: Step S10, obtaining multi-source cultural element data of Yi ethnic group folk songs, where the cultural element data includes audio data, lyric texts, singing contexts, festival backgrounds, singer information and Yi language translation content, and parsing the cultural element data into multiple semantic element nodes.

[0027] In the process of constructing the inheritance path of Yi ethnic group folk songs, in order to ensure that the generated path can accurately reflect the multi-dimensional cultural characteristics of the folk songs, it is necessary to first comprehensively and systematically analyze the original cultural resources and transform them into knowledge nodes with semantic expression capabilities. Since Yi ethnic group folk songs not only have musical characteristics, but also are nested with complex contexts such as language habits, cultural etiquette, social functions and festival customs, it is necessary to establish a data structure that integrates audio, text, semantics and background information to support the subsequent semantic map construction and path modeling processes.

[0028] The multi-source cultural element data refers to a data set centered on Yi ethnic group folk songs, originating from different media and cultural levels. The element content included is used to construct node information in the knowledge graph and is the basic material for the construction of the entire inheritance path.

[0029] Among the cultural element data: Audio data refers to digital audio files recording the actual performances of Yi ethnic group folk songs, including acoustic features such as melody, rhythm, and singing intonation.

[0030] Lyric texts refer to the Chinese and Yi ethnic group language lyrics corresponding to Yi ethnic group folk songs, including poetic structures and language styles.

[0031] Singing context refers to the social and cultural situations in which the folk songs are used, including singing locations, uses (such as weddings, etc.).

[0032] Festival background refers to whether the folk song is associated with specific festivals or rituals.

[0033] Singer information refers to the names, ethnic groups, genders, ages, inheritance lineages, etc. of traditional singers.

[0034] Yi ethnic group language translation content refers to the translation of lyrics or semantic content from the Yi ethnic group language into a common language for non-Yi ethnic group language users to understand and learn.

[0035] In a specific implementation of step S10, step S10 specifically includes the following sub-steps: Step S101: Collect the original performance data of Yi ethnic group folk songs, obtain audio files including complete tracks and corresponding performance video materials. The collection of the audio files should include multiple vocal part versions. If there are performance versions of multiple inheritance lineages, the source ethnic group, region, and inheritor information should be marked separately.

[0036] Step S102: Extract lyric text data, transcribe the audio data into text content through an automatic speech recognition system. If the system's automatic recognition is inaccurate, the Yi ethnic group language lyrics should be dictated manually, and semantic annotation and translation should be carried out by researchers with bilingual abilities to form a lyric text structure in a comparative format.

[0037] Step S103: Record singing context information, obtain the usage situations of the folk songs through field investigations or folk literature, and classify them into types including but not limited to weddings, funerals, farming, blessings, drinking songs, etc., and encode them using a standardized tagging system.

[0038] Step S104: Obtain festival background information, combine the Yi ethnic group calendar, local chronicles, and cultural department materials to determine the ethnic festivals, traditional activities, or cultural ceremonies associated with the folk song, and set multi-level tags to identify their importance, spread, and degree of ritual embedding.

[0039] Step S105: Build a singer information file, identify the singers of the collected folk song materials and archive the data, collect basic information such as their life resumes, affiliated villages, and master-disciple relationships, and form a "singer-song" mapping table.

[0040] Step S106: Organize the Yi language translation content. For the Yi language paragraphs in the lyric text, mark their corresponding syllable pinyin and free translation content, and record them in a nested structure form for subsequent machine processing and language teaching system calls.

[0041] Step S107: Uniformly organize the above-collected and organized information into a structured data format, construct a set of semantic element nodes, and each node is attached with corresponding tags, context identifiers, and semantic dimension information, which are input as the basic elements for graph construction into the subsequent steps.

[0042] Through the implementation of the above steps, not only the systematic collection of Yi folk songs in multiple cultural dimensions is realized, but also the accuracy of subsequent semantic modeling and the cultural depth of path construction are ensured. By converting the original unstructured data into structured and semantic knowledge nodes, the system's parsability and inheritability of traditional folk songs are greatly improved, and at the same time, it provides complete data support for constructing paths for personalized learning.

[0043] In an exemplary specific implementation, collect the wedding folk song "A Shi Qie" widely sung during the "Torch Festival" in Liangshan area. Through audio transcription and text comparison, the following node structure is formed: "A Shi Qie (Node 1)" is associated with "wedding scene (Node 2)", "Torch Festival (Node 3)", "Singer A (Node 4)", "Yi sentence pattern B (Node 5)", etc., forming multi-sided connections in the graph, providing complete cultural semantic support for subsequent path selection and learning recommendation.

[0044] Step S20: Build a semantic graph of Yi folk songs based on the semantic element nodes. Each node in the graph represents a folk song knowledge unit, and the connection edges between nodes represent logical dependency relationships.

[0045] In order to realize the structured, personalized, and dynamic recommendation of the inheritance path of Yi folk songs, it is necessary to further organize the semantic element nodes parsed in Step S10 into a knowledge graph structure with associated semantics and context logic. Since there are non-linear and multiple-dependent semantic relationships among various cultural elements in Yi folk songs, a single data level cannot support path modeling and recommendation engines. Therefore, it is necessary to build a semantic graph to organically connect discrete nodes in a logical dependency manner to realize the functions of knowledge networking, association visualization, and semantic reasoning, providing a topological structure foundation and semantic connectivity support for subsequent path construction.

[0046] In a specific implementation of step S20, step S20 specifically includes the following sub-steps: Step S201: Define the structural model of the semantic graph. An heterogeneous graph structure is adopted, which includes multiple types of nodes and edges. The node types include, but are not limited to, lyric nodes, melody nodes, festival nodes, scene nodes, singer nodes, and language nodes. The edge types include causal relationships, homologous relationships, similarity relationships, and temporal relationships.

[0047] Step S202: Map the semantic element nodes in step S10 to the node set of the graph structure according to their attributes. Assign an ID number to each node using a unique identifier, establish a node attribute table, and record the node type, language category, information source, semantic label, and cultural level.

[0048] Step S203: Construct the logical dependency relationship between nodes. For each pair of node pairs that may have a semantic relationship, call the association rule mining algorithm to calculate their association degree index. The association degree calculation can adopt the following form:

[0049] where, represents the semantic association degree between node and node , and respectively represent the occurrence frequencies of node and node , represents the frequency of their co-occurrence.

[0050] When exceeds the set threshold , a directed edge pointing from node to node is added to the semantic graph, and the type and weight of the edge are marked.

[0051] Step S204: Perform graph structure optimization processing. Complement the semantics of isolated nodes. If a node is not effectively connected to any other node, add an inference connection to it through a knowledge transfer mechanism or expert annotation method to improve the connectivity of the overall graph.

[0052] Step S205: Complete the storage and call configuration of the graph structure data. Save the constructed semantic graph structure in the format of a graph database, support subsequent query, path deduction, and visualization display. Optional implementation solutions include: using the Neo4j graph database to achieve persistent storage of nodes and edges, or using triple knowledge representation based on the RDF format to achieve cross-platform compatibility.

[0053] Through the above steps, a knowledge graph system covering the semantic relationships throughout the life cycle of Yi folk songs can be constructed, realizing the visualization, networking, and structuring of the complex semantic structures between the original data nodes, thereby providing a basic topological structure, context semantic support, and dynamic expansion ability for path generation. At the same time, the graph structure can be continuously updated, with adaptability and evolution ability, providing a semantic reasoning basis for personalized path recommendation.

[0054] In a specific example, in the collection example of the folk song "A Shi Qie", its lyric nodes can be connected to the "wedding scene" node and the "Torch Festival" node through edges, the melody nodes are connected to the "female solo" node and the "flat-tone style" node, the singer nodes are connected to the "inheritance pedigree - branch X" node, and the language nodes are connected to the "southern Yi dialect" node. Through the construction of these nodes and edges, a complete multi-dimensional semantic association sub-graph can be formed in the semantic graph for subsequent path construction and evaluation.

[0055] Step S30: Cluster the semantic graph to generate multiple semantic association clusters, and construct an inheritance node cluster containing native nodes, education nodes, practice nodes, and evaluation nodes according to the semantic association clusters. Define each inheritance path as a structural unit containing the above types of nodes.

[0056] In the process of constructing the inheritance path of Yi folk songs, to ensure that the inheritance path has complete cultural expression, teaching adaptation, practice scenario mapping, and learning evaluation and feedback capabilities, it is necessary to conduct a structural analysis of the constructed semantic graph of Yi folk songs, identify groups of nodes with adjacent semantics through clustering, and then abstract them into functional nodes in the path, and combine them into inheritance path units with stable structure, clear goals, and derivable behaviors. Since there are differences in the expression content, usage scenarios, and cultural functions of each node in the semantic graph, clustering methods are used to uniformly classify at the semantic level to ensure that each path contains a continuous cognitive chain from "cultural starting point" to "ability verification".

[0057] In this step: The native node represents a node of the folk song's ontological content, such as lyric segments, melody fragments, singing styles, etc.

[0058] The education node is a node associated with teaching activities, such as course modules, teaching objectives, textbook locations, etc.

[0059] The practice node represents a node of the folk song application scenarios such as actual performances, festival performances, and folk activities.

[0060] The evaluation node is a node used to carry the results of learning behaviors, such as semantic understanding accuracy rate, pitch matching degree, context reproduction index, etc.

[0061] In a specific implementation, step S30 includes the following sub-steps: Step S301: Node vector feature construction

[0062] Extract the attributes of all nodes in the semantic graph and construct a vectorized representation of each node. The vector includes: Word embedding encoding, reflecting the semantic distribution of the node name; Adjacency structure encoding, indicating the connection situation of the node in the graph; Function label vector, indicating its category (native, educational, etc.); Context level identifier, encoding the context level of the node in the inheritance logic.

[0063] The implementation method is to splice the above elements into a fixed-dimensional vector

[0064] Where: is the node number; is the vector dimension, for example, set to 128 dimensions; Word embedding uses a pre-trained model such as Word2Vec or BERT; Adjacency encoding adopts a combination of node degree and PageRank value; Function label is One-hot encoding; Level identifier is represented by a hierarchical number.

[0065] Step S302: Semantic clustering analysis

[0066] Input all node vectors into a graph clustering algorithm for clustering. Optional implementation schemes include: K-means clustering; Hierarchical clustering; Louvain community partitioning algorithm; HDBSCAN density clustering algorithm.

[0067] Taking K-means as an example: Set the number of clusters ; Initialize cluster centers; Iterative update: Assignment: Assign each vector to the nearest cluster center; Update: Recalculate each cluster center as the mean of the vectors it belongs to; Terminate when the cluster assignment no longer changes or reaches the maximum number of iterations.

[0068] The output is One semantic association cluster 。

[0069] Step S303: Functional node classification For each node within each semantic cluster, conduct a functional determination. Combine the original label of the node and the context attributes of the subgraph it belongs to to judge its functional type. The classification criteria are as follows: If a node is directly related to audio, lyrics, or rhythm, label it as a "native node"; If a node links to a course number, textbook chapter, or learning objective, label it as an "educational node"; If a node is associated with a festival, activity venue, or community performance, label it as a "practical node"; If a node participates in a scoring mechanism, behavior metrics, or learning feedback, label it as an "evaluation node".

[0070] Each set of nodes of each type is respectively denoted as: Native node set ; Educational node set ; Practical node set ; Evaluation node set 。

[0071] Step S304: Definition of inheritance path structure

[0072] Construct a standard inheritance path structure unit and set the path as a quadruple:

[0073] Where: , representing a native node; , representing an educational node; , representing a practical node; , representing an evaluation node.

[0074] If a certain path does not contain any of the above four types of nodes, it is not included in the set of valid paths.

[0075] Step S305: Path combination and storage

[0076] Extract combinations of the above four types of nodes with connection relationships from the graph as specific path units, and record their node numbers, connection order, and path IDs. The path information is stored in a structured path library to support subsequent path evaluation and personalized matching.

[0077] Through the implementation of the above steps, it is possible to extract an inheritance path structure with cultural integrity, teaching systematicity, and practical feasibility from the complex Yi ethnic folk song knowledge graph, effectively avoiding the problems of "isolated content, disjointed teaching, and lack of feedback" in the inheritance of traditional folk songs. This method can be adapted to different teaching groups and supports the dynamic expansion and version update of the path, with high plasticity and promotion value.

[0078] In a specific example, still taking the folk song "Ashique" as an example, the following nodes are extracted from the semantic graph: The lyric melody node "Ashique Lyric Segment 1" is classified as a native node ; The textbook chapter node "Chapter 3 of Vocal Music Basic Course" is classified as an educational node ; The practical task node "Campus Performance during the Torch Festival in 2024" is classified as a practical node ; The scoring node "Pitch Reduction Rate Evaluation Standard" is classified as an evaluation node .

[0079] Combine the above four types of nodes to generate a path quadruple:

[0080] The system stores this path in the path number library, marks it as "Path P-1003", and uses it as a candidate for subsequent learner path matching. If the subsequent learner profile matches well with the function of this path, this path will be preferentially recommended as an inheritance plan.

[0081] Step S40: According to the learner profile information, construct an adaptability function to evaluate the matching degree between each inheritance path and the learner, select the inheritance path with the largest matching degree, and generate a personalized Yi ethnic folk song inheritance path.

[0082] To achieve personalized recommendation and adaptive matching of the Yi ethnic folk song inheritance path, it is necessary to establish an individual characteristic model based on the learner's knowledge background, cultural cognitive level, language ability, and learning preferences, and combine the previously constructed inheritance path structure to evaluate the adaptability degree between each path and the learner. Since different learners have significant differences in aspects such as language foundation, vocal music experience, and cultural familiarity, using a fixed path will cause problems such as poor teaching effects and cultural understanding deviations. Therefore, it is necessary to construct a quantifiable adaptability function, and by aligning the characteristics between the learner profile and the path attributes, select the most matching path plan, so as to generate a precise, efficient, and dynamically adjustable personalized Yi ethnic folk song inheritance path.

[0083] In this step: Learner profile information refers to a multi-dimensional feature vector generated based on information such as the learner's basic attributes, historical behaviors, and preference tags, and is used to describe the learner's ability status and learning needs.

[0084] The adaptability function refers to the functional form used to calculate the matching degree between the learner profile and the path structure, and is usually expressed in the form of a similarity metric or a scoring function.

[0085] The inheritance path refers to a complete structural unit composed of a native node, an educational node, a practice node, and an evaluation node, and is used to describe the implementation process of a specific folk song in teaching.

[0086] The personalized path refers to the path that best matches a specific learner profile selected based on the adaptability function, and is the only output result of the "person-path" mapping relationship.

[0087] In a specific implementation, step S40 specifically includes the following sub-steps: Step S401: Construct a learner profile feature model

[0088] Extract the learner profile from the registration information, course records, and interaction behaviors to form a feature vector containing the following elements: Ethnic identity code, used to judge its affinity for the Yi language and culture; Language ability level, divided into three levels: high, medium, and low based on the language proficiency test scores; Vocal music foundation grading, archived as primary, intermediate, or advanced based on parameters such as pitch accuracy, sense of rhythm, and vocal range; Learning preference tags, including preference information such as liking strong rhythms, popular lyrics, and female vocal performances; Historical performance indicators, including behavioral indicators such as completion rate, engagement level, and duration.

[0089] Encode the above features into vector form and represent them as:

[0090] Among them, is the learner profile vector, represents the feature value of the th dimension, is the total number of features.

[0091] Step S402: Vectorized modeling of path attributes For each path constructed in step S30 , extract the attribute information such as the content label, language feature, rhythm style, teaching difficulty, and practice method of the nodes it contains, and construct a path attribute vector:

[0092] Among them, represents the feature vector of the th path, represents the attribute value of this path in the th dimension, corresponding to the th feature in the learner profile.

[0093] Step S403: Adaptability function calculation

[0094] Calculate the matching degree between the path and the learner profile using the vector similarity function:

[0095] Among them: is the adaptability score between the learner and the path ; is the weight coefficient of the th dimension; and are the numerical values of the learner and the path in the th dimension respectively.

[0096] Step S404: Select the optimal path Calculate the matching scores for the entire set of candidate paths and obtain the path number with the highest score:

[0097] Take the path pointed to by as the personalized Yi folk song inheritance path for the current learner.

[0098] Step S405: Path binding and system recording

[0099] Bind the path to the learner one by one, record its path number, execution time and path version, and synchronize them to the learning platform database for subsequent behavior feedback and dynamic adjustment.

[0100] By implementing the above steps, it is possible to finely profile learners with different backgrounds, abilities and preferences, ensure that the generated inheritance paths are highly consistent with their cognitive structures and learning goals, thereby significantly improving learning efficiency, cultural cognition accuracy and the stability of inheritance quality. This mechanism not only has real-time and scalability, but also has the ability to evaluate and feedback on paths, and can re-optimize paths as the behavior of learners changes.

[0101] In a specific example: For a learner L, the system extracts the following portrait information: Ethnic identity: Han; Language ability: Completely unfamiliar with Yi language, scored 1; Vocal music foundation: Intermediate, scored 3; Preference tags: Strong sense of rhythm, female singing style; Historical performance: High completion but weak cultural understanding.

[0102] Construct the learner vector:

[0103] Extract the path from the path candidate library (such as the example of "A Shi Qie" mentioned above), and its attribute vector is:

[0104] Set the weight , then the similarity calculation result is: , according to the principle of the largest score, the system selects the path number P-1032 as the personalized inheritance path for the learner L, and pushes the corresponding courses, singing videos and practice plans to achieve a complete closed-loop from portrait recognition to path construction.

[0105] Step S50, during the execution of the inheritance path, collect the learner's behavior data at each node, including indicators such as melody restoration degree, semantic understanding degree and practice completion degree, calculate the evaluation function value of each node, and feedback and update the function value to the path structure.

[0106] After completing the learner path recommendation and entering the path execution stage, in order to realize the dynamic monitoring of the learning process and the adaptive adjustment of the path structure, it is necessary to collect the learner's behavior performance data in real time during the execution of the path node. Since the learning of Yi folk songs not only involves melody imitation and language memory, but also involves cultural understanding and practice performance, a single-dimensional evaluation is not sufficient to reflect the learning state. Therefore, representative behavior indicators should be collected respectively during the execution of the native node, education node, practice node and evaluation node, a multi-dimensional evaluation function should be constructed, and the evaluation results should be fed back to the path model in real time for subsequent path adjustment and learning plan optimization.

[0107] In this step: Melody restoration degree refers to the degree of restoration of the standard pitch, rhythm and musical figure by the learner when singing, and is an index to evaluate the quality of folk song imitation singing.

[0108] Semantic understanding degree refers to the depth of the learner's understanding of the lyrics content, cultural background and semantic structure, which can be realized through answering questions or voice retelling.

[0109] The practice completion degree refers to whether the learner has completed the practice tasks, such as performance participation, recording submission, scenario demonstration, etc. The completion situation is used to evaluate the learning effect.

[0110] In a specific implementation, step S50 includes the following sub-steps: Step S501: Initialize the behavior data acquisition mechanism

[0111] Before each node of the path is executed, bind the behavior acquisition module. The module sets the corresponding acquisition scheme according to the node type, specifically including: For native nodes, collect audio input and perform melody comparison analysis; For educational nodes, collect Q&A data, semantic annotations or oral summaries; For practice nodes, record the activity participation status and achievement upload records; For evaluation nodes, integrate the system score and the tutor's score.

[0112] Optional implementation solutions include: using speech recognition algorithms to compare pitch deviations, using natural language processing to score text understanding, and using a task management system to quantify the practice completion status.

[0113] Step S502: Index calculation and standardization processing

[0114] For each type of behavior data, define a normalization index , representing the original performance score of the th type of index. After standardization processing, it falls into the interval [0,1] and is used for subsequent comprehensive scoring. The indexes include but are not limited to: melody restoration degree, semantic understanding degree, practice completion degree. Optionally, pitch matching degree can be used for scoring.

[0115] Step S503: Construction and calculation of the evaluation function

[0116] Set a multi-index weighted evaluation function , in the following form:

[0117] Where: is the comprehensive evaluation score; is the number of index types; is the weight of the th index, satisfying ; is the standardized score of the th behavior index.

[0118] The system dynamically sets according to different path types Weights. For example, for the cultural understanding path, the weight of semantic understanding can be increased, while for the performance-oriented path, the weights of melody restoration and practice completion can be increased.

[0119] Step S504: Execute the path feedback mechanism

[0120] Bind the node evaluation function value to the path structure. The system determines whether to fine-tune the path according to the scores of each node. The adjustment contents include: Reduce the recommendation priority of nodes with lower scores; Add supplementary nodes for weak links; Rearrange the execution order of nodes; Mark as "path to be strengthened" or "path adaptation successful".

[0121] Update the path results and synchronously write them into the database for recommendation optimization in the subsequent learning stage or the construction of individual evolution paths.

[0122] Through the implementation of this step, a behavior feedback loop can be established during the operation of the Yi folk song inheritance path, which can not only accurately reflect the real performance of learners at different stages, but also provide a quantitative basis for the dynamic reconstruction and personalized iteration of the path. The evaluation process comprehensively covers three dimensions of music skills, language understanding, and practice completion, with the advantages of definable indicators, observable behaviors, and adjustable paths, significantly improving the intelligence, adaptability, and stability of the inheritance path.

[0123] Specifically, taking the inheritance path P-1032 of the aforementioned "A Shi Qie" as an example, after the learner L executes the original node "melody imitation", the system collects his recording and analyzes the melody deviation to obtain the restoration score ; in the education node "lyric understanding answering questions", 4 out of 6 questions are answered correctly, and the semantic understanding degree is calculated ; in the practice node "Torch Festival performance", the submitted recording video is not completed, and the practice completion degree . Assuming the weight w = [0.4, 0.4, 0.2], the evaluation function value is:

[0124] The system identifies it as a "partially completed path", marks this path as "recommended to strengthen the practice link", and inserts a "simple performance task" node into the path database for strengthening the practice task in the subsequent learning stage to ensure the complete closure of the inheritance chain.

[0125] Step S60: Archive the finally constructed inheritance path and the learner's behavior results to generate a path resume and an inheritance trajectory map.

[0126] After completing the execution of the inheritance path and behavior evaluation, to ensure the traceability of the learning process and the sustainable optimization of the path system, it is necessary to archive the path structure executed by each learner and their behavior evaluation results, and generate an inheritance path resume record and a trajectory map for analysis and display based on this data. Through the archiving mechanism, the process record and phased analysis of the individual inheritance process can be realized, and the construction of the trajectory map helps to understand the node distribution, link bottlenecks and dynamic evolution trends of the inheritance of Yi folk songs from the overall perspective of the path space. This archiving and mapping mechanism not only provides a basis for subsequent teaching decision-making and inheritance intervention, but also provides data support for knowledge graph update, curriculum system transformation and policy planning.

[0127] In this step: The path resume refers to recording the inheritance path completed by the learner as structured data in chronological order, including node numbers, completion status, timestamps and evaluation scores.

[0128] The inheritance trajectory map refers to visually displaying the learning trajectory, path jumps and phased status of a learner or group in the entire inheritance network in the form of a graph structure.

[0129] In a specific implementation, the step S60 includes the following sub-steps: Step S601: Initialization of the archiving data structure

[0130] After the path execution ends, create a structured archiving object and define the following fields: Learner identification (UID); Path number (PID); Path version (VersionID); Node list (NodeList), including the number, type and execution order of each node; Node completion status (StatusList), recording the completion status of each node, with values including "completed", "not completed", "partially completed"; Node evaluation value (EvalList), corresponding to the behavior evaluation score of the node in step S50; Execution time (TimeStampList), the entry and exit times of each node; Overall evaluation score (PathScore), the overall path performance score obtained according to the comprehensive weight.

[0131] This structure is organized and stored in JSON or relational table structure. Optional implementation options include: using MongoDB database for document storage, or using PostgreSQL to record in a standardized form.

[0132] Step S602: Generation of path resume records

[0133] Assemble the above fields into a learner-path resume entry and store it in the path resume table to form a complete process record. For example: { "UID": "L-102", "PID": "P-1032", "NodeList": ["A1", "E3", "P2", "V5"], "StatusList": ["Completed", "Completed", "Not Completed", "Partially Completed"], "EvalList": [0.92, 0.67, 0.00, 0.55], "TimeStampList": ["2025-02-10T08:30", ..., "2025-02-10T10:15"], "PathScore": 0.636 }

[0134] Furthermore, if there are multiple path versions, the version history can be recorded and subsequent comparative analysis can be supported.

[0135] Step S603: Generation of Trajectory Map Nodes and Edges Construct a set of map nodes, where the nodes are the path nodes executed each time, and the edges are the logical order and chronological relationship between the nodes. The map structure is represented as:

[0136] Where: is the trajectory map of the user ; is the set of path nodes; is the chronological edge between the nodes; Each edge is attached with the execution time and completion status attributes.

[0137] If the same learner executes multiple paths, a multi-path trajectory superposition graph can be constructed.

[0138] Step S604: Integration of Trajectory Map Visualization and System Feedback

[0139] Use a visualization graph engine (such as D3.js, ECharts) to display the trajectory map in the form of a dynamic interactive graph, supporting the following operations: Display the learner's path execution chain; Highlight low-score nodes; Show the overlapping areas of multi-user paths; Identify the areas with high-frequency node failures.

[0140] Write the graph data back to the backend of the inheritance system to support subsequent path recommendation optimization and the application of graph evolution algorithms.

[0141] See Figure 2 , in another embodiment, the present invention also provides a system for constructing the inheritance path of Yi ethnic group folk songs, including: A multi-source cultural element collection module for obtaining multi-source cultural element data of Yi ethnic group folk songs. The cultural element data includes audio data, lyric texts, singing contexts, festival backgrounds, singer information, and Yi language translation content, and parsing the cultural elements into multiple semantic element nodes; A semantic graph construction module for constructing a semantic graph of Yi ethnic group folk songs based on the semantic element nodes. Each node in the semantic graph represents a folk song knowledge unit, and the connection edges between nodes represent logical dependency relationships; A node clustering and path construction module for clustering the semantic graph to generate multiple semantic association clusters, and constructing an inheritance node cluster including native nodes, education nodes, practice nodes, and evaluation nodes according to the semantic association clusters, and defining each inheritance path as a structural unit including the above types of nodes; A path matching and recommendation module for constructing an adaptability function according to the learner profile information to evaluate the matching degree between each inheritance path and the learner, selecting the inheritance path with the largest matching degree, and generating a personalized inheritance path of Yi ethnic group folk songs; A behavior collection and evaluation module for collecting the behavior data of learners at each node during the execution of the inheritance path, including melody restoration degree, semantic understanding degree, and practice completion degree indicators, calculating the evaluation function values of each node, and feeding back and updating the function values to the path structure; A path archiving and trajectory graph module for archiving the finally constructed inheritance path and the learner behavior results, and generating a path resume and an inheritance trajectory graph.

[0142] In a further implementation, the multi-source cultural element collection module includes: An audio collection unit for collecting audio and video materials of Yi ethnic group folk songs including multiple voices, and annotating singer, region, and pedigree information; A speech transcription and translation unit for using a speech recognition system to transcribe the audio into lyric texts, and performing Yi language annotation and translation by bilingual personnel; A context annotation unit for recording the singing context based on the field survey results and performing multi-level label encoding; A festival annotation unit for extracting festival backgrounds and establishing cultural level tags by combining the Yi calendar and local literature; A singer file - building unit for constructing a singer information file and recording their individual attributes and pedigree; A Yi language nested - structure unit for establishing the pinyin and interpretation structure of Yi language paragraphs in lyrics; A node generation unit for structuring data into a set of semantic element nodes and generating attribute tags.

[0143] In a further implementation, the semantic graph construction module includes: A graph - structure definition unit for establishing a graph model containing node types of lyrics, melody, festivals, and singers in a heterogeneous - graph manner; A node - attribute encoding unit for numbering each node and establishing an attribute table of its type, label, and cultural level; A semantic - edge generation unit for calling an association - rule mining algorithm, calculating the semantic association degree between nodes according to the occurrence frequency and co - occurrence frequency, and generating a connection edge when the association degree exceeds a set threshold; An isolated - node processing unit for performing knowledge migration or manual completion on unconnected isolated nodes; A graph - storage unit for saving the constructed semantic graph in the form of a graph database or RDF triples.

[0144] In a further implementation, the node clustering and path - construction module includes: A node - vector generation unit for extracting the attributes of all nodes in the graph and constructing a node vector containing word embedding, structure encoding, function label, and context level; A clustering - analysis unit for clustering node vectors using a graph - clustering algorithm; A node - classification unit for dividing the clustering results into primitive nodes, educational nodes, practice nodes, and evaluation nodes; A path - template setting unit for setting a standard inheritance - path structure template, including at least one primitive node, one educational node, one practice node, and one evaluation node; A path - combination and numbering unit for generating a path structure by combining nodes based on connection relationships and assigning a path number to be stored in the path library.

[0145] In a further implementation, the path matching and recommendation module includes: A portrait - modeling unit for collecting learner's ethnicity, language ability, vocal music foundation, learning preferences, and historical behavior data to construct a learner portrait vector; A path - attribute modeling unit for extracting content features, language types, rhythm styles, and teaching difficulties in the path to construct a path - attribute vector; An adaptation degree calculation unit, which is used to calculate the matching scores between the learner and each path by using a weighted cosine similarity function; A path selection unit, which is used to select the path with the highest adaptation degree score as the personalized recommendation path; A path binding unit, which is used to bind the learner to the selected path and write the binding relationship into the system database.

[0146] In a further implementation, the behavior collection and evaluation module includes: A behavior collection unit, which is used to configure a node behavior collection mechanism before path execution and collect vocal imitation audio, answering text, performance records and scoring data according to node types; An index calculation unit, which is used to calculate standardized index values for the collected behavior data, including melody restoration degree, semantic understanding degree and practice completion degree; An evaluation function unit, which is used to set a weighted evaluation function and calculate the comprehensive score of nodes according to multiple indexes; A path adjustment unit, which is used to feedback the evaluation result to the path structure and adjust the node recommendation weight or add supplementary nodes according to the score.

[0147] In a further implementation, the path archiving and trajectory graph module includes: A resume structure construction unit, which is used to create a path resume data structure including learner identification, path number, node execution order, completion status, scoring result and time stamp; A resume archiving unit, which is used to save the path resume in a structured format to the path database; A graph generation unit, which is used to construct an inheritance trajectory graph according to the resume data. Nodes in the graph represent path nodes, and edges represent execution order and status attributes; A graph visualization unit, which is used to display the trajectory graph as an interactive graph, identify path bottlenecks and feedback them to the path recommendation module.

[0148] It should be noted that the explanatory descriptions of the foregoing embodiments of the Yi ethnic folk song inheritance path construction method also apply to the devices in the embodiments of the present application, and will not be elaborated here.

[0149] Those of ordinary skill in the art can realize that the units and algorithm steps described in the embodiments disclosed herein can be implemented by a combination of electronic hardware, computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraint conditions of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0150] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0151] In several embodiments provided by the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (hereinafter referred to as ROM), random access memory (hereinafter referred to as RAM), magnetic disks, or optical discs that can store program codes.

[0152] The above is only the specific implementation manner of the present application. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all should be covered by the protection scope of the present application. For the parts of the module structure that are not specifically defined in the present invention, the content recorded in the prior art shall prevail. The prior art mentioned in the foregoing background art part and the specific embodiment part of the present invention can be used as a part of the present invention to understand the meaning of some technical features or parameters.

Claims

1. A method for constructing the inheritance path of Yi ethnic group folk songs, characterized in that The method includes the following steps: Step S10, obtaining multi-source cultural element data of Yi ethnic group folk songs, where the cultural element data includes audio data, lyric texts, singing contexts, festival backgrounds, singer information, and Yi language translation content, and parsing the cultural element data into multiple semantic element nodes; Step S20, constructing a semantic graph of Yi ethnic group folk songs based on the semantic element nodes, where each node in the graph represents a folk song knowledge unit, and the connecting edges between nodes represent logical dependency relationships; Step S30, performing clustering processing on the semantic graph to generate multiple semantic association clusters, and constructing an inheritance node cluster including native nodes, education nodes, practice nodes, and evaluation nodes according to the semantic association clusters, and defining each inheritance path as a structural unit including the above types of nodes; Step S40, according to the learner portrait information, constructing an adaptability function to evaluate the matching degree between each inheritance path and the learner, selecting the inheritance path with the largest matching degree, and generating a personalized Yi ethnic group folk song inheritance path; Step S50, during the execution of the inheritance path, collecting the behavior data of the learner at each node, including melody restoration degree, semantic understanding degree, and practice completion degree indicators, calculating the evaluation function value of each node, and feeding back and updating the function value to the path structure; Step S60, archiving the finally constructed inheritance path and the learner behavior result to generate a path resume and an inheritance trajectory graph.

2. The method for constructing the inheritance path of Yi ethnic group folk songs according to claim 1, wherein, The step S10 includes: Step S101, collecting the original singing data of Yi ethnic group folk songs, including audio data and video materials of multiple voices, and annotating the corresponding singer, region, and pedigree information; Step S102, using a speech recognition system to transcribe the audio data into lyric texts, and having bilingual personnel perform Yi language semantic annotation and translation; Step S103, recording the singing context of the folk song based on field surveys and classifying and coding it using multi-level tags; Step S104, extracting the festival background by combining the Yi calendar and local chronicles, and setting festival types and cultural embedding level tags for each folk song; Step S105, constructing a singer information file and collecting their name, gender, age, and teacher-student pedigree; Step S106, establishing a nested structure of Yi language paragraphs, syllable pinyin, and corresponding Chinese interpretations in the lyrics; Step S107, structurally representing the collected data as a set of semantic element nodes, and recording the context attributes and tag information.

3. The method for constructing the inheritance path of Yi ethnic group folk songs according to claim 1, characterized in that, The step S20 includes: Step S201, defining a semantic graph model using a heterogeneous graph structure, including various types such as lyric nodes, melody nodes, festival nodes, and singer nodes; Step S202, numbering the semantic element nodes and establishing a node attribute table to record their types, semantic tags, and cultural level information; Step S203, calling an association rule mining algorithm to calculate the semantic association degree between nodes based on the node occurrence frequency and co-occurrence frequency, and constructing a connecting edge when the association degree is greater than a set threshold; Step S204, performing knowledge migration and manual complementation operations on isolated nodes; Step S205, saving the semantic graph in the form of a graph database or RDF triples.

4. The method for constructing the inheritance path of Yi ethnic group folk songs according to claim 1, wherein The step S30 includes: Step S301: Extract the attributes of all nodes in the semantic graph, and construct node vectors that include word embeddings, structural encodings, function labels, and context levels. Step S302: Use a graph clustering algorithm to cluster the node vectors. Step S303: Divide the nodes in the clustering results into native nodes, educational nodes, practice nodes, and evaluation nodes. Step S304: Set a standard inheritance path structure unit, which includes at least one native node, one educational node, one practice node, and one evaluation node. Step S305: Combine path structures based on node connection relationships, generate path numbers, and store them in the path library.

5. The method for constructing the inheritance path of Yi ethnic group folk songs according to claim 1, characterized in that The said Step S40 includes: Step S401: Collect the ethnic identity, language ability, vocal music foundation, preference labels, and historical learning behaviors of learners, and construct a multi-dimensional learner portrait vector. Step S402: Extract the content attributes, language types, rhythm characteristics, and teaching difficulty information in the path structure, and construct a path attribute vector. Step S403: Use a weighted cosine similarity function to calculate the fitness scores between the learner portrait and each path. Step S404: Select the path with the highest fitness score as the personalized inheritance path for the current learner. Step S405: Store the binding relationship between the learner and the path, and synchronize it to the system database.

6. The method for constructing the inheritance path of Yi ethnic group folk songs according to claim 1, characterized in that The said Step S50 includes: Step S501: Configure a node behavior collection module before path execution, and collect behavior data such as melody imitation singing, answer understanding, practice records, and scoring feedback according to node types. Step S502: Calculate standardized index values for the collected data. The indexes include melody restoration degree, semantic understanding degree, and practice completion degree. Step S503: Set a weighted evaluation function to integrate multiple indexes and calculate the comprehensive score of the node. Step S504: Bind the evaluation score to the path structure, and adjust the node recommendation weight or insert supplementary nodes according to the score.

7. The method for constructing the inheritance path of Yi ethnic group folk songs according to claim 1, wherein The said Step S60 includes: Step S601: Create a path resume data structure to record learner identification, path number, node execution order, completion status, scoring value, and timestamp. Step S602: Archive the path resume in a structured format into the path database. Step S603: Construct an inheritance trajectory graph, where the nodes are the inheritance nodes in the resume, the edges are the execution order relationships between the nodes, and execution time and status attributes are attached to the edges. Step S604: Visualize the learning path based on the trajectory graph, identify learning bottlenecks, and feedback the trajectory data to the path recommendation module.

8. A system for constructing an inheritance path of Yi ethnic group folk songs, characterized in that, The said system includes: A multi-source cultural element collection module, which is used to obtain multi-source cultural element data of Yi ethnic group folk songs. The cultural element data includes audio data, lyric texts, singing contexts, festival backgrounds, singer information, and Yi language translation content, and parse the cultural elements into multiple semantic element nodes. A semantic graph construction module, which is used to construct a semantic graph of Yi ethnic group folk songs based on the semantic element nodes. Each node in the semantic graph represents a folk song knowledge unit, and the connection edges between the nodes represent logical dependency relationships. The node clustering and path construction module is used to perform clustering processing on the semantic graph, generate multiple semantically associated clusters, and construct an inheritance node cluster containing native nodes, educational nodes, practice nodes, and evaluation nodes according to the semantically associated clusters. Each inheritance path is defined as a structural unit containing the above various types of nodes; The path matching and recommendation module is used to construct an adaptability function based on the learner portrait information to evaluate the matching degree between each inheritance path and the learner, select the inheritance path with the highest matching degree, and generate a personalized Yi ethnic folk song inheritance path; The behavior collection and evaluation module is used to collect the behavior data of the learner at each node during the execution of the inheritance path, including indicators such as melody restoration degree, semantic understanding degree, and practice completion degree, calculate the evaluation function value of each node, and feedback and update the function value to the path structure; The path archiving and trajectory graph module is used to archive the finally constructed inheritance path and the learner behavior results, and generate a path resume and an inheritance trajectory graph.

9. The Yi folk song inheritance path construction system according to claim 8, characterized in that The multi-source cultural element collection module includes: The audio collection unit is used to collect Yi ethnic folk song audio and video materials including multiple voices, and annotate the singer, region, and genealogy information; The speech transcription and translation unit is used to transcribe the audio into lyric text using a speech recognition system, and perform Yi language annotation and translation by bilingual personnel; The context annotation unit is used to record the singing context based on the field survey results and perform multi-level label encoding; The festival annotation unit is used to extract the festival background by combining the Yi calendar and local literature and establish cultural level labels; The singer file creation unit is used to construct a singer information file and record their individual attributes and genealogy; The Yi language nested structure unit is used to establish the pinyin and interpretation structure of the Yi language paragraphs in the lyrics; The node generation unit is used to structure the data into a semantic element node set and generate attribute labels.

10. The Yi folk song inheritance path construction system according to claim 8, characterized in that, The semantic graph construction module includes: The graph structure definition unit is used to establish a graph model containing node types such as lyrics, melody, festival, and singer in a heterogeneous graph manner; The node attribute encoding unit is used to number each node and establish an attribute table of its type, label, and cultural level; The semantic edge generation unit is used to call the association rule mining algorithm, calculate the semantic association degree between nodes according to the occurrence frequency and co-occurrence frequency, and generate a connection edge when the association degree exceeds the set threshold; The isolated node processing unit is used to perform knowledge migration or manual completion on the isolated nodes that have not established connections; The graph storage unit is used to save the constructed semantic graph in the form of a graph database or RDF triples.

Images