Data information storage method and system of online Chinese language learning platform

Abstract

The present application relates to the technical field of data processing, and more particularly to a data information storage method and system for an online Chinese language learning platform. The method comprises the following steps: obtaining platform user interaction data, and performing pinyin input track analysis and Chinese character component structure coding to obtain pinyin aligned input frames; extracting pinyin-Chinese character mapping relationships in the pinyin aligned input frames, and performing Chinese character structure path reconstruction to obtain a user input word structure path set; calling learning content data, and performing multi-modal content node embedding to obtain content linkage segment clusters; constructing a single session learning topology track according to the content linkage segment clusters, and performing bidirectional topology sorting to obtain a learning track topology graph; establishing a multi-dimensional primary key index and a behavior triggering mechanism for the learning track topology graph to obtain a dynamic index distribution block, and transmitting the dynamic index distribution block to the online Chinese language learning platform. The present application significantly improves the stability and expandability of the learning platform.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and particularly relates to a data information storage method and system of an online Chinese language learning platform. BACKGROUND

[0002] The online Chinese language learning platform provides course resources, teaching interaction, homework submission and intelligent evaluation functions through the network, greatly improving the accessibility and flexibility of Chinese language learning, and promoting the transformation and upgrading of Chinese teaching mode. In the actual operation process of such a platform, a large amount of structured and unstructured data such as learners' operation behavior, learning progress, course content, teacher interaction record, answer data and evaluation feedback is continuously generated, forming a huge data information system. In order to support the intelligent recommendation, learning path optimization, personalized learning evaluation and subsequent content update functions of the platform, efficient, controllable and scalable storage and management of the above learning data have become the core demand of the evolution of platform functions and the improvement of service quality.

[0003] The traditional online learning platform mainly uses a relational database system for content management and user data archiving in terms of data storage. Although this method has certain advantages in structured data management, it has obvious bottlenecks when facing multi-dimensional and multi-type learning data. For example, the traditional method is difficult to model and manage unstructured data such as voice interaction data, classroom interaction records, mixed text and video learning progress, resulting in serious data island phenomenon. In addition, when facing large-scale learning user groups with high frequency of reading and writing, the traditional data storage structure often has access conflicts, data redundancy and performance bottlenecks, which seriously affect the stability and response speed of the learning platform. SUMMARY

[0004] Therefore, it is necessary to provide a data information storage method and system of an online Chinese language learning platform to solve at least one of the above technical problems.

[0005] To achieve the above purpose, a data information storage method of an online Chinese language learning platform comprises the following steps:

[0006] Step S1: obtaining platform user interaction data, and performing pinyin input track analysis and Chinese character component structure coding according to the platform user interaction data to obtain pinyin aligned input frames;

[0007] Step S2: extracting pinyin-Chinese character mapping relationship in the pinyin aligned input frame, and using a preset word structure index table to reconstruct the Chinese character structure path of the pinyin-Chinese character mapping relationship to obtain a user input word structure path set;

[0008] Step S3: retrieve learning content data associated with platform user interaction data through an online Chinese language learning platform, and split content segment structure units according to the learning content data; perform multi-modal content node embedding on the user input word structure path set and the content segment structure units to obtain a content linkage segment cluster;

[0009] Step S4: construct a single-session learning topology trajectory according to the content linkage segment cluster, and perform bidirectional topology sorting on the single-session learning topology trajectory to obtain a learning trajectory topology graph;

[0010] Step S5: perform learning state segmentation aggregation on the learning trajectory topology graph to generate a Chinese language state compression frame; establish a multi-dimensional primary key index and a behavior triggering mechanism according to the Chinese language state compression frame to obtain a dynamic index distribution block, and transmit the dynamic index distribution block to the online Chinese language learning platform to deploy mechanism arrangement tasks.

[0011] The present application realizes unified processing and deep mining of multi-modal and multi-level data in Chinese language learning process through detailed analysis and structure coding of platform user interaction data, greatly improving the scientificity and fineness of data management. The analysis of pinyin input trajectory combined with Chinese character component structure coding not only accurately captures the detailed information of user input, but also improves the semantic expression ability of data through structured expression, laying a solid foundation for subsequent mapping relationship and path reconstruction. The use of a preset word structure index table for structure path reconstruction of pinyin-Chinese character mapping effectively realizes the accurate structured representation of user input words, supporting complex semantic association analysis. The splitting of content segment structure units and multi-modal node embedding technology enables the deep integration of user input information and learning content in multiple dimensions such as text, graphics, and audio, greatly enriching the content presentation form and matching accuracy. By constructing a conversation learning topology trajectory and bidirectional topology sorting, not only the knowledge structure and logical order in the learning process are fully revealed, but also the time and semantic coherence of the learning behavior trajectory is enhanced. The state segmentation aggregation and primary key index construction of the learning trajectory topology graph, combined with the behavior triggering mechanism, effectively realize the dynamic compression and rapid indexing of learning state, providing strong data support for intelligent recommendation and personalized learning path planning. The parameter such as the path overlap ratio threshold in topology sorting is set to 0.6 to ensure the rationality of path fusion, preventing information loss caused by excessive merging, and avoiding redundant fine-grained paths to improve the robustness and response speed of the system. Overall, the present scheme effectively solves the problem of managing multi-modal and unstructured data in traditional storage, enhances the processing capability of the platform for large-scale and high-frequency interaction data, and significantly improves the stability, scalability and personalized service level of the learning platform.

[0012] Optionally, step S1 includes:

[0013] Step S11: Obtain platform user interaction data, and perform unified timestamp encoding to obtain an original input behavior record set;

[0014] Step S12: Perform dynamic pinyin segmentation on the original input behavior record set, and label pinyin segmentation boundaries to obtain a pinyin action sequence matrix;

[0015] Step S13: Call a preset local word library index to perform pinyin-Hanzi mapping alignment on the pinyin action sequence to obtain a pinyin-Hanzi alignment table;

[0016] Step S14: Perform Hanzi component structure coding on the pinyin-Hanzi alignment table to generate a component structure vector set;

[0017] Step S15: Associate and fuse the pinyin action sequence matrix and the component structure vector set according to the alignment relationship in the pinyin-Hanzi alignment table to generate a pinyin alignment input frame.

[0018] The application realizes accurate time synchronization and sequence management of input behavior by performing unified timestamp encoding on platform user interaction data, effectively solves the time inconsistency problem in multi-source, multi-time sequence data fusion, and ensures that the data basis for subsequent processing is accurate and reliable. Dynamic pinyin segmentation and pinyin segmentation boundary labeling help capture fine-grained pinyin information of user input, improve the coherence and accuracy of pinyin recognition, and thus more accurately reflect the real input behavior. Reasonable setting of boundary labeling parameters ensures the continuity and integrity of segmentation, avoiding excessive segmentation or omission. Mapping and alignment of pinyin and Hanzi using a local word library index enables pinyin input to be quickly and accurately mapped to corresponding Hanzi, taking into account the coverage rate and response speed of the word library, and the parameter design takes into account the accuracy and computational efficiency of mapping matching. Coding the Hanzi component structure of the pinyin-Hanzi alignment result, by decomposing the Hanzi into basic components and vectorizing the expression, greatly enhances the computability of Hanzi semantic and structural information, which is helpful for subsequent deep semantic analysis and content matching. The pinyin action sequence matrix and the component structure vector set are fused according to the alignment relationship to form a pinyin alignment input frame, realizing the organic integration of multi-dimensional data and improving the expression richness and semantic accuracy of input data.

[0019] Optionally, step S12 comprises:

[0020] Step S121: Extract the character input sequence in the original input behavior record set, and construct a pinyin segment stream;

[0021] Step S122: Perform pinyin substring segmentation on the pinyin segment stream according to a preset initial-medial-final triplet matching logic to obtain a candidate pinyin segment set;

[0022] Step S123: phonetic distribution statistics are performed on the candidate phonetic segment set, and syllable confidence values are calculated according to the phonetic distribution statistics results;

[0023] Step S124: weighting scores are calculated based on the phonetic length, input continuity and corresponding syllable confidence values in the candidate phonetic segment set, the highest-scoring substring segmentation combination is selected as the segmentation boundary rule, and the candidate phonetic segment set is segmented according to the segmentation boundary rule to obtain a phonetic segmentation boundary sequence;

[0024] Step S125: phonetic unit error correction expansion is performed on the phonetic segmentation boundary sequence, thereby constructing a phonetic action item set;

[0025] Step S126: the phonetic action item set is aligned with the corresponding start and end positions, error correction states and phonetic types in the phonetic segment stream to generate a phonetic action sequence matrix.

[0026] The present application realizes the structured representation of input information by constructing a phonetic segment stream for the character sequence in the original input behavior, which facilitates subsequent fine processing and recognition. The phonetic segment stream is segmented in combination with the initial-radical-rime-tone triplet matching logic, which can accurately divide phonetic substrings and avoid recognition errors caused by phonetic fragmentation or overlap, ensuring the integrity and accuracy of the segmentation results. The distribution statistics of candidate phonetic segments and the calculation of syllable confidence values effectively quantify the reliability of phonetic substrings, filter out more reliable phonetic inputs, and improve the overall stability of phonetic recognition. The optimal substring segmentation combination is selected by weighting scores considering the phonetic length, input continuity and confidence, which ensures the scientific and reasonable segmentation boundary rule, effectively balances the fineness and robustness of segmentation, and the parameter setting reflects the consideration of input stream continuity and phonetic accuracy, avoiding the problems of sentence segmentation caused by excessive segmentation and ambiguous definition caused by insufficient segmentation. Error correction expansion is performed on the phonetic segmentation boundary sequence, which helps to correct possible phonetic errors in user input and improves the fault tolerance and recognition accuracy of input data. Finally, the error-corrected phonetic action items are aligned with the start and end positions, error correction states and phonetic types in the phonetic segment stream to construct a phonetic action sequence matrix, forming a unified and accurate input expression structure, which facilitates the subsequent mapping of phonetic and Chinese characters and semantic analysis.

[0027] Optionally, step S125 includes:

[0028] Extracting phonetic substrings in the phonetic segmentation boundary sequence to construct a basic phonetic unit list;

[0029] Loading a built-in phonetic error correction mapping table of the platform, including initial radical replacement rules, rime merging rules and tone tolerance tables;

[0030] According to the pinyin error correction mapping table, each pinyin substring in the basic pinyin unit list is subjected to initial-final error matching expansion, and a pinyin error correction candidate set is obtained;

[0031] The word initial and word final pinyin collocation probability of the left and right context interval ±2 pinyin units in the pinyin segment stream is extracted, the input context misjudgment probability is calculated by combining the self structure error of each candidate in the pinyin error correction candidate set, and the confidence reconstruction value is calculated according to the input context misjudgment probability, and the candidate whose confidence reconstruction value is greater than or equal to 0.6 is retained;

[0032] Each pinyin substring is matched with the candidate with the highest confidence reconstruction value in the corresponding pinyin error correction candidate set, and a standardized pinyin action item set is obtained.

[0033] The pinyin sub-strings in the pinyin segmentation boundary sequence are extracted, and a basic pinyin unit list is constructed, which realizes the structured representation of user input pinyin information, and lays a solid foundation for subsequent error correction and standardization processing. The built-in pinyin error correction mapping table of the loading platform covers initial replacement, final merging and tone tolerance, which can effectively cover the common pronunciation and writing deviations in pinyin input, enhance the comprehensiveness and flexibility of error correction, and enable the system to adapt to diversified user input. Based on the error correction mapping table, the initial-final error matching expansion of the basic pinyin unit is performed, and the pinyin error correction candidate set is constructed, which effectively captures the possible pinyin error range of the user and improves the accuracy and recall rate of pinyin error correction. By extracting the word initial and word final collocation probability of the left and right context interval ±2 pinyin units in the pinyin segment stream, the input context misjudgment probability is calculated by combining the self structure error of the candidate, and the confidence reconstruction value is further calculated, forming a scientific and reasonable confidence evaluation mechanism, which can accurately identify high-confidence pinyin candidates and effectively reduce the risk of misjudgment. The confidence reconstruction value parameter is set to not less than 0.6, which takes into account the error correction sensitivity and robustness, avoids excessive error correction, and ensures the reliability of pinyin standardization. Finally, according to the highest confidence reconstruction value matching standard, the standardization processing of pinyin action items is realized, ensuring the uniformity and standardization of input data, greatly improving the accuracy of subsequent Chinese character mapping and semantic analysis and the overall interactive experience of the system.

[0034] Optionally, the Chinese character structure path reconstruction in step S2 comprises:

[0035] The structure path of each Chinese character in the pinyin-Chinese character mapping relationship is queried and located by using the word structure index table, and a structure record sequence is obtained;

[0036] The structure level is expanded based on the structure record sequence, and a Chinese character structure topology graph sequence is constructed;

[0037] The traversal paths of each graph in the Chinese character structure topology graph sequence from the main component to the secondary component are linearly encoded, and a structure path encoding sequence is obtained.

[0038] The structural path coding sequence is semantically compressed to form a structural path fragment set;

[0039] The structural path fragment set is bound to the user input behavior ID in the pinyin alignment input frame to generate a user input word structural path set.

[0040] The present application realizes the systematic extraction of the complex structure information of Chinese characters by using the word structure index table to query and locate each Chinese character in the pinyin-Hanzi mapping relationship, ensures that each component of the Chinese character and its hierarchical relationship is accurately identified and represented, and solves the problem of incomplete or confused expression of Chinese character structure in traditional input processing. Expanding the hierarchical structure of Chinese characters based on the structure record sequence, constructing a topological graph sequence, helps to fully reflect the internal composition and spatial hierarchy of Chinese characters, supports multi-dimensional analysis of Chinese character structure, and improves the accuracy of subsequent coding and semantic processing. By linearly encoding the traversal path from the main component to the secondary component in the topological graph, the structural path coding sequence is converted, so that the complex graph structure can be expressed simply, which is convenient for computer processing and storage, improves the coding efficiency and the normativity of expression. The structural path coding sequence is processed by semantic compression, which can effectively remove redundant information, extract core structure features, reduce data storage and transmission burden, while maintaining the integrity of structure expression and semantic association, enhancing the response speed and processing performance of the system. The structural path fragment set is bound to the user input behavior ID to form a user input word structural path set, realizing the organic integration of input behavior and Chinese character structure expression, ensuring the time sequence consistency and traceability of input analysis. The overall parameter design takes into account the coding granularity and compression ratio, such as the reasonable value of linear coding length and compression threshold, to ensure that the details of Chinese character structure can be accurately expressed, and the consumption of computing resources and storage space can be effectively controlled, thereby significantly improving the performance and user experience of online Chinese language learning platform in Chinese character input analysis, learning content association and intelligent feedback functions.

[0041] Optionally, the step S3 of splitting the content fragment structure unit comprises:

[0042] According to the user input behavior ID and the course number in the platform user interaction data, the corresponding course content set is obtained through the online Chinese language learning platform, and the course text unit in the corresponding course content set is loaded, so as to obtain learning content data;

[0043] The natural paragraph in the learning content data is semantically segmented to construct a semantic fragment unit;

[0044] According to the semantic fragment unit, word labeling and structure feature labeling are performed to obtain a structured labeling result;

[0045] Perform semantic density analysis on the structured annotation results, mark semantic segments with a semantic density of 0.75 or more as knowledge focus, and obtain a knowledge focus set;

[0046] According to the knowledge focus set, the semantic segment unit is divided into content segment unit, and the timestamp of each content segment unit is relocated to obtain the content segment structure unit.

[0047] The present application realizes efficient association of learning content and user interaction data by accurately obtaining corresponding course content based on user input behavior ID and course number, ensures the pertinence and real-time of content retrieval, and improves the efficiency of platform resource calling. The natural paragraphs in the course text unit are semantically divided to construct semantic segment units, so that the text content can be finely divided for subsequent fine analysis and processing, thereby enhancing the structured expression ability of the content. The word annotation and structural feature marking are performed to effectively reveal the semantic hierarchy and word relationship of the text, enrich the structured information expression of the content, and support subsequent intelligent matching and knowledge point mining. Through semantic density analysis, 0.75 is used as the threshold to accurately distinguish important knowledge focus, ensuring that the marked knowledge points have high semantic relevance and teaching value, improving the accuracy and practicality of the knowledge focus, and facilitating the grasping and effective review of learning content. Based on the knowledge focus set, the content segment unit is divided, and the timestamp is relocated to realize dynamic adjustment of the content structure and accurate mapping of the time dimension, supporting fine tracking and management of learning progress. The overall parameter setting takes into account the fineness and processing efficiency of the division, so that the system can fully capture the core knowledge structure in the course content and efficiently respond to user needs, significantly improving the content service quality and learning experience of the online Chinese language learning platform.

[0048] Optionally, the multi-modal content node embedding in step S3 comprises:

[0049] The user input word structure path set is topologically vectorized to obtain a structure vector encoding set;

[0050] According to the content segment structure unit, text word embedding in the corresponding course content set, supporting Chinese character component graphic texture features in the text, phoneme sequence features in the segment dubbing audio, and text syntax label distribution features are extracted to construct a multi-modal segment feature set;

[0051] Bidirectional matching is performed on the structure vector encoding set and the multi-modal segment feature set to obtain an associated matching matrix;

[0052] From the associated matching matrix, a matching pair mapping pair with a matching degree of 0.7 or more is selected to construct a content node matching set;

[0053] According to the content node matching set, the content segment is grouped and aggregated, and a content linkage segment cluster is generated.

[0054] The present application realizes the accurate digital expression of Chinese character structure information by converting the user input word structure path set into a structure topology vector code, enhances the compatibility and efficiency of input data in computer processing, and effectively supports subsequent multi-modal content fusion. Combined with the text word meaning embedding, Chinese character component graphic texture features, phoneme sequence features of the dubbing audio, and text grammar label distribution features in the course content, a rich multi-modal segment feature set is constructed, the multi-dimensional representation ability of the content in terms of semantics, vision and hearing is improved, and the expression and matching accuracy of the content are enhanced. The bidirectional matching process can realize the comprehensive comparison of input structure coding and multi-modal content features, ensure the bidirectional mutual verification and high consistency of information, avoid the deviation caused by single direction matching, and improve the accuracy and robustness of the matching result. Selecting the matching degree threshold as 0.7 balances the accuracy and coverage of the matching, ensures the high-quality content association, avoids the content omission caused by excessive screening, and improves the overall matching performance of the system and the utilization efficiency of learning resources. Based on the matching result, a content node matching set is constructed and grouped and aggregated to form a content linkage segment cluster, which effectively integrates the association structure between the user input and the learning content, promotes the hierarchical management and dynamic linkage of the content, and improves the intelligent level and user experience of the online Chinese language learning platform in terms of personalized recommendation, learning path planning and knowledge point tracking.

[0055] Optionally, step S4 comprises:

[0056] Step S41: Extracting the segment list in each cluster in the content linkage segment cluster, and time sorting the segment list according to the trigger time of each content segment unit to obtain an initial segment trajectory sequence;

[0057] Step S42: Performing semantic adjacency analysis on the initial segment trajectory sequence to generate a segment transition relationship graph;

[0058] Step S43: Calculating the transition path weight of the edge set in the segment transition relationship graph, and pruning the edges with a transition path weight less than 0.45 to obtain an optimized learning sub-trajectory set;

[0059] Step S44: Aggregating the optimized learning sub-trajectory set in a directed graph structure to obtain a single-session learning topology trajectory graph;

[0060] Step S45: Performing bidirectional topology sorting on the single-session learning topology trajectory graph to obtain a learning trajectory topology graph.

[0061] The application sorts the segment list in the content linkage segment cluster according to the trigger time, accurately restores the time sequence in the user learning process, ensures the time continuity and integrity of the learning track, and improves the timing accuracy of subsequent analysis. The semantic adjacency analysis of the initial segment track sequence effectively reveals the semantic relationship and logical jump between different content segments, promotes the deep understanding and correlation mining of the learning content structure, and enhances the semantic expression ability of the learning track. The edge weight calculation mechanism is adopted, the semantic similarity and knowledge point correlation degree are comprehensively weighted and assigned, the learning transfer strength between segments is quantitatively reflected, and the key nodes and important transfers in the learning path are effectively identified. The weight threshold is set to 0.45 for pruning, reasonably filtering the weak correlation path, avoiding redundancy and noise interference, improving the clarity and accuracy of the learning track, and being beneficial to the aggregation and analysis of subsequent tracks. The optimized sub-track set is aggregated through the directed graph structure, which is helpful to build a multi-path topological structure that conforms to the actual learning logic, fully reflect the diversified learning behavior track of the user, and enhance the expression richness and dynamic adaptability of the track. The bidirectional topological sorting processing ensures that the learning track topological graph is sequentially expanded in both forward and reverse directions, supports multi-angle learning path analysis and evaluation, and improves the integrity and flexibility of track analysis. The overall parameter configuration considers data accuracy and processing efficiency, ensures the stability and accuracy of the learning track construction process, and significantly improves the individualized learning path optimization ability and intelligent teaching support level of the online Chinese language learning platform.

[0062] Optionally, step S5 comprises:

[0063] Step S51: performing semantic stage division on the learning track topological graph to generate a semantic stage segment group;

[0064] Step S52: performing stage aggregation according to the start and end time interval and structure jump degree in the semantic stage segment group to construct a learning state evolution chain;

[0065] Step S53: extracting principal component feature fields based on the learning state evolution chain to construct a Chinese language state compression frame;

[0066] Step S54: extracting the theme density, structure complexity and learning time length segment in the Chinese language state compression frame, converting the theme density, structure complexity and learning time length segment into a multi-primary key combination, and generating a primary key index set;

[0067] Step S55: aggregating the primary keys with the same theme density and structure complexity in the primary key index set into a behavior mode to establish a trigger mapping table;

[0068] Step S56: merging and packaging the primary key index set with the trigger mapping table to obtain a dynamic index distribution block, and transmitting the dynamic index distribution block to the online Chinese language learning platform to deploy the mechanism arrangement task.

[0069] The present application realizes the detailed decomposition of the complex learning process by dividing the learning trajectory topology into semantic stages, so that the learning behavior can be divided into stage segments with clear semantic characteristics, effectively improving the understanding ability of the internal logic and knowledge point level of the learning content. Based on the aggregation of the start and end time interval and the structure jump degree of the stage segment, the learning content with continuous time and strong structural correlation can be reasonably integrated to form a coherent learning state evolution chain, thereby accurately reflecting the dynamic evolution and knowledge mastery process of the learning process. By extracting the principal component feature field in the learning state evolution chain, a Chinese language state compression frame is constructed, which significantly reduces the data dimension and redundant information, realizes efficient expression and storage of the learning state, and improves the computational efficiency of subsequent processing. In the state compression frame, the theme density, structural complexity and learning time segment are extracted and converted into a multi-primary key combination to form a primary key index set, enhancing the multi-dimensional expression ability of the index and supporting fine differentiation and fast retrieval of different learning states. The primary keys with similar theme density and structural complexity are aggregated into behavior patterns, and a trigger mapping table is established to realize the patterned management and automatic triggering mechanism of the learning behavior, improving the response speed and accuracy of the learning state change. Finally, the primary key index set and the trigger mapping table are merged and packaged to generate a dynamic index distribution block, which is transmitted to the learning platform to ensure efficient data transmission and dynamic scheduling of tasks, effectively supporting the intelligent management and personalized teaching services of the platform. The parameters in each step, such as the structural jump degree threshold, the principal component selection ratio and the multi-primary key combination method, are optimized to balance data compression and information retention, ensuring fine characterization of the learning state and improving the processing performance and response speed of the system, ensuring stable operation and efficient service of the online Chinese language learning platform in a large-scale user environment.

[0070] Optionally, the present specification also provides a data information storage system of an online Chinese language learning platform for executing the data information storage method of the online Chinese language learning platform as described above, which comprises:

[0071] A Chinese character structure alignment module is configured to obtain platform user interaction data, and perform pinyin input trajectory analysis and Chinese character component structure coding based on the platform user interaction data to obtain a pinyin alignment input frame.

[0072] A mapping relationship analysis module is configured to extract a pinyin-Chinese character mapping relationship in the pinyin alignment input frame, and reconstruct a Chinese character structure path of the pinyin-Chinese character mapping relationship using a preset word structure index table to obtain a user input word structure path set.

[0073] a content segment embedding module configured to call learning content data associated with platform user interaction data through the online Chinese language learning platform, and split content segment structure units according to the learning content data; perform multi-modal content node embedding on a user input word structure path set and the content segment structure units to obtain a content linkage segment cluster;

[0074] a learning trajectory analysis module configured to construct a single-session learning topology trajectory according to the content linkage segment cluster, and perform bidirectional topology sorting on the single-session learning topology trajectory to obtain a learning trajectory topology graph;

[0075] a learning state aggregation module configured to perform learning state segmentation aggregation on the learning trajectory topology graph to generate a Chinese language state compression frame; establish a multi-dimensional primary key index and a behavior triggering mechanism according to the Chinese language state compression frame to obtain a dynamic index distribution block, and transmit the dynamic index distribution block to the online Chinese language learning platform to deploy mechanism orchestration tasks. BRIEF DESCRIPTION OF DRAWINGS

[0076] Other features, objects, and advantages of the application will become more apparent from the following detailed description when read in connection with the following drawings:

[0077] Figure 1 a schematic diagram of a step flow of a data information storage method of an online Chinese language learning platform of the present application;

[0078] Figure 2 a schematic diagram of a detailed step flow of step S1 in the present application;

[0079] The implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0080] The technical method of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0081] In addition, the accompanying drawings are only schematic illustrations of the present application, and are not necessarily drawn to scale. The same reference numerals in the drawings represent the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities, which do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0082] It should be understood that although terms such as "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0083] To achieve the above object, please refer to Figures 1 to 2 , the present invention provides a method for storing data information of an online Chinese language learning platform, and the method includes the following steps:

[0084] Step S1: Obtain platform user interaction data, and perform pinyin input trajectory parsing and Chinese character component structure encoding according to the platform user interaction data to obtain a pinyin-aligned input frame;

[0085] In this embodiment, during the process of the user practicing real-time pinyin input through the online Chinese language learning platform, interaction data such as the user input sequence, key timestamp, key position start and end interval, and input device type are automatically collected, and are uniformly converted into a time-aligned format with an accuracy of 0.01 seconds. The original input behavior records extracted from this data set contain fields {user_id, key_seq, timestamp, device_type}. Then, the platform uses the built-in pinyin segmentation module to perform three-segment segmentation of the character sequence into initials, finals, and tones. For example, "zhong1guo2" is segmented into {zh-ong-1, guo-2}, forming a pinyin action matrix where n is the number of pinyin units. Subsequently, the locally cached lexicon index (constructed based on GB2312) is called to map the pinyin to standard Chinese characters. For example, "zhong1" is mapped to "中", forming a pinyin-Chinese character alignment table T ph =[(zhong1, 中), (guo2, 国)]. Finally, according to the Chinese character decomposition rules, "中" is parsed into "丨+口", and "国" is parsed into "囗+玉". Each component structure vector uses 5-dimensional one-hot encoding, and the structure encoding matrix Finally, combine P and S and align them with each pinyin in T ph to obtain a pinyin-aligned input frame I 拼 , which is used for structural path reconstruction.

[0086] Step S2: Extract the pinyin-Chinese character mapping relationship in the input frame by pinyin alignment, and use the preset word structure index table to reconstruct the Chinese character structure path of the pinyin-Chinese character mapping relationship, so as to obtain the user input word structure path set;

[0087] In this embodiment, for each record in the pinyin alignment input frame I 拼 in, call the preset word structure index table T struct (in the format of JSON, recording the component path and hierarchical information of each Chinese character) for path reconstruction. For example, query the structure index corresponding to "国" as {"国": ["囗", ["玉", ["丶", "丶"]]]}, and expand it into a linear structure path [囗 → 玉 → 丶 → 丶] through the path parser. Perform a topological expansion process with the main component being prioritized on the structure path to generate a tree diagram structure G, where the nodes are component codes and the edges are structure position offsets (such as up, down, left, right). This diagram structure is traversed and converted into a path vector sequence V = [(囗, 1), (玉, 2), (丶, 3), (丶, 4)], and further adopt the path segment compression rule to only retain the topological main path and branch numbers, obtaining the path compression code C path = ["囗 → 玉", "玉 → 丶 × 2"]. Finally, the system binds the encoded path set to the input ID of the original pinyin input frame to generate the user input word structure path set P 字 = {user_id, [path_1, path_2,...]}, for subsequent multi-modal content node matching.

[0088] Step S3: Retrieve the learning content data associated with the interaction data of the platform users through the online Chinese language learning platform, and split the content fragment structural units according to the learning content data; perform multi-modal content node embedding on the user input word structure path set and the content fragment structural units to obtain the content联动fragment cluster;

[0089] In this embodiment, during the user's learning process, the platform calls the course number associated with the current input behavior ID and retrieves the corresponding course resource package. Each text in the resource package is segmented by paragraph, and each paragraph is subjected to词性标注and semantic syntactic analysis. For example, the paragraph "我们学习汉字的结构" will be decomposed into structural units ["我们", "学习", "汉字", "结构"], and the dependency relationship of each unit is extracted. Subsequently, the platform performs vector matching based on the structure vectors in the user input word structure path set P 字 to extract the following feature vector groups for each content fragment: (1) text semantic vector (generated by the embedding model), (2) graphic texture vector (derived from the convolutional features of Chinese character stroke images), (3) voice phoneme spectrum vector The three types of feature vectors are matched with the user structure vector by cosine similarity, and the matching degrees greater than 0.7 are selected to form a content linkage segment cluster, and each cluster structure is denoted as Finally, the content linkage segment cluster is formed for learning path modeling.

[0090] Step S4: constructing a single-session learning topology trajectory according to the content linkage segment cluster, and performing bidirectional topology sorting on the single-session learning topology trajectory to obtain a learning trajectory topology graph;

[0091] In this embodiment, each cluster in the content linkage segment cluster is sorted in the time dimension to generate an initial learning trajectory sequence S 轨 =[f1, f2,..., fn], where each fi is a segment node. Semantic adjacency scoring is performed between adjacent segments in the sequence to form a segment transition graph G 转 =(V, E), and the weight of E is calculated by the keyword intersection ratio (the weight range is [0, 1]). The edges with a weight less than 0.45 are pruned to retain high coherence paths. The connected subgraphs are then aggregated in graph structure to form a directed graph T 轨 , which records the learning direction. The system then performs forward topology scanning and reverse backtracking, cross-compares the bidirectional paths, calculates the proportion of overlapping nodes, selects path pairs with an intersection ratio greater than or equal to 0.6, and retains them as main path chains M 链 . Redundant paths in non-main path chains are processed by semantic merging and mapped to the nodes of the main path. Finally, the main path and semantic projection form a structured “learning trajectory topology graph” T 拓 , including node hierarchy, transition probability, and semantic annotation, which is used for subsequent state evolution.

[0092] Step S5: aggregating learning state segments to generate Chinese language state compression frames; establishing a multi-dimensional primary key index and a behavior triggering mechanism according to the Chinese language state compression frames to obtain a dynamic index distribution block, and transmitting the dynamic index distribution block to an online Chinese language learning platform to deploy mechanism orchestration tasks.

[0093] In this embodiment, based on the semantic theme, duration, and node structure jump degree between segments in T 拓 , the learning process is divided into multiple semantic stages, each stage has a time interval not less than 60 seconds and a structure jump degree not higher than 2 layers. Principal component features are extracted in each stage, including structure expansion rate (SEI), semantic density (SDI), and node stay average (NTS), to form a state vector matrix K-means compression (k=3) is performed on M 压 to generate Chinese language state compression frames S 帧 . The system then extracts S 帧The main key combination K is composed of the topic density (such as the "culture" topic density = 0.82), the structural complexity (such as the average number of syntactic branches = 4.3), and the learning time length (30-180 seconds per section) 主 ={0.82, 4.3, 120}. The main key combination with the same topic density ± 0.05 and the structural complexity ± 1 is merged into a behavior group, and is aggregated into a behavior pattern set. Finally, a trigger mapping table is constructed, including the main key mapping behavior decision template, and a dynamic index distribution block D 索 ={K 主 , template ID} is output to the platform controller for deployment strategy.

[0094] Optionally, step S1 includes:

[0095] Step S11: Obtain platform user interaction data and perform unified timestamp encoding to obtain an original input behavior record set;

[0096] In this embodiment, when a user performs an input operation through a platform, the platform backend records the complete input behavior sequence of the user, including user ID (user_id), key sequence (key_seq), key timestamp (timestamp), device identification (device_type), and other original input behavior information. The platform sets the timestamp precision to 10 milliseconds, and maps all input records to a standard time axis to avoid cross-device offset problems. For example, the user's input of "zhongguo" is recorded in the following structured format: {"user_id":"U20240601","key_seq":["z","h","o","n","g","g","u","o"],"timestamp":[100,150,180,220,260,300,340,380],"device_type":"mobile"}; linear time series interpolation is used to fill in the missing part of the key interval to obtain an input sequence with uniform time steps. The completed original input behavior record set will serve as the basic data set for subsequent pinyin segmentation and input analysis.

[0097] Step S12: Perform dynamic pinyin segmentation on the original input behavior record set, and label the pinyin section boundaries to obtain a pinyin action sequence matrix;

[0098] In this embodiment, the key sequence in the original input behavior record is extracted and assembled into a pinyin segment stream, such as ["zh", "o", "n", "g", "g", "u", "o"]. According to the built-in initial-final-tone triplet matching table (such as {"zh": S, "ong": V, "1": T}), the system combines the segment stream into a candidate pinyin segment set, such as ["zhong", "guo"]. The structure of each segment is calibrated according to the initial position, syllable boundary, and tone mapping. In this process, the pinyin length threshold is set to 1-6 characters, and the syllable confidence value is calculated by reverse mapping of word frequency, with a minimum threshold of 0.6. Finally, a score function S = 0.4 · L + 0.3 · C + 0.3 · P (L is the length weight, C is the continuity score, and P is the confidence) is used to generate a pinyin segmentation boundary sequence, for example: [{"start": 0, "end": 4, "pinyin": "zhong", "score": 0.83}, {"start": 5, "end": 7, "pinyin": "guo", "score": 0.79}]; the sequence is arranged in matrix form M 拼每行分别表示起始、结束、拼音内容及得分，用于拼音动作生成。

[0099] 步骤S13：调用预设的本地词库索引，对拼音动作序列进行拼音-汉字映射对齐，得到拼音-汉字对齐表；

[0100] 本实施例中，调用本地缓存词库索引(例如基于《通用规范汉字表》构建的索引结构，包含约8105条拼音-汉字映射关系)，对步骤S12中拼音动作序列矩阵M拼进行一对一对齐。例如，"zhong"→"中"，"guo"→"国"。系统内部设定候选汉字置信度列表，例如对"guo”可对应"国(0.85)”,"过(0.70)”,"果(0.58)”等，通过输入上下文进行置信度筛选，保留置信度最高项。对齐结果整理为表格结构：[{"pinyin":"zhong","hanzi":"中","confidence":0.92},{"pi nyin":"guo","hanzi":"国","confidence":0.85}]；该拼音-汉字对齐表Tph用于指导下一步的结构编码绑定操作。

[0101] 步骤S14：对拼音-汉字对齐表进行汉字部件结构编码，生成部件结构向量集；

[0102] 本实施例中，依据Tph中的汉字项，调用结构分解器对每个汉字执行部件级拆解。例如，"中”拆解为"丨”和"口”，"国”拆解为"囗”和"玉”。每个部件采用五维属性向量编码，包括笔画数(strokes)、笔顺类型(stroke_order)、相对位置(position_index)、部件类型(component_type)、部件层级(hierarchy_level)。示例如下：{"中":[{"component":"丨","vector":[1,2,1,1,1]},{"component":"口","vector":[3,1,2,0,1]}],"国":[{"component":"囗","vector":[3,1,1,0,1]},{"component":"玉","vector":[5,2,2,1,2]}]}；将这些部件向量按汉字顺序构建结构向量集其中n为部件总数。该向量集用于与拼音动作序列的融合。

[0103] 步骤S15：将拼音动作序列矩阵与部件结构向量集按照拼音-汉字对齐表中的对齐关系进行关联融合，生成拼音对齐输入帧。

[0104] 本实施例中，将拼音动作序列矩阵M拼与结构向量集Sstruct对应拼音项进行对齐融合。融合方式依据拼音-汉字对齐表中的位置信息，将拼音动作的起止时间及内容分别映射至结构向量集中的对应汉字结构项。融合结果组织为拼音对齐输入帧I拼，结构如下：[{"pinyin":"zhong","time_range":[100,260],"hanzi":"中","components":[{"co mponent":"丨","vector":[1,2,1,1,1]},{"component":"口","vector":[3,1,2,0,1]}]},{"pinyin":"guo","ti me_range":[300,380],"hanzi":"国","components":[{"component":"囗","vector":[3,1,1,0,1]},{"comp onent":"玉","vector":[5,2,2,1,2]}]}]，该结构将拼音行为轨迹与汉字结构表达进行语义级融合，是后续进行结构路径重建及多模态匹配的关键基础帧。

[0105] 可选地，步骤S12包括：

[0106] 步骤S121：提取原始输入行为记录集中的字符输入序列，并构建拼音片段流；

[0107] 本实施例中，从原始输入行为记录集中提取出按键事件序列，以字符形式还原出用户连续拼音输入的字符串，例如从以下原始行为数据：{"user_id":"U20250620","key_seq":["n","i","h","a","o","z","h","o","n","g","g","u","o"],"timestamp":[100,120,150,170,190,250,270,290,310,330,350,370,390]}，拼接字符流为原始输入串"nihaozhongguo"。为便于后续切分处理，该字符串被转化为拼音片段流["n","i","h","a","o","z","h","o","n","g","g","u","o"]，并对每个字符赋予序号索引，用于后续起止位置定位与切分边界跟踪。

[0108] 步骤S122：根据预设的声母-韵母-音调三元组匹配逻辑对拼音片段流进行拼音子串切分，得到候选拼音段集合；

[0109] 本实施例中，加载预设的声母-韵母-音调组合表，该表覆盖常用拼音组合形式(共收录423个常用拼音项)，每项包括声母结构、韵母模式、音调容差范围。例如"zhong"的结构为{"shengm u":"zh","yunmu":"ong","tone":null}。系统扫描拼音片段流，结合最大匹配长度(不超过6个字符)和优先声母匹配的逻辑，从首字符起逐步生成拼音子串候选段，例如：["ni","hao","zhong","guo"]，每个候选拼音段被记录其起止位置、切分组合与结构匹配类型，作为后续置信统计基础。

[0110] 步骤S123：对候选拼音段集合进行拼音分布统计，并根据拼音分布统计结果计算音节置信度值；

[0111] 本实施例中，基于候选拼音段集合进行拼音频次统计，加载平台词频统计表(基于用户行为构建，包含拼音使用频次与错误率分布)，对每个拼音段的音节置信度进行评估。以"zhong"为例，其在近30天用户输入中频次为43212，错误率为0.0089，置信度计算结果为1-误差率×调整系数，设定调整系数为1.2，得到：conf("zhong")＝1-0.0089×1.2≈0.9893；该值被记录于拼音分布统计表中，并附加在对应拼音段结构中供后续加权评分使用。

[0112] 步骤S124：基于候选拼音段集合中的拼音长度、输入连续性及对应的音节置信度值进行加权评分，选取评分最高的子串切分组合作为分节边界规则，并利用分节边界规则对候选拼音段集合进行分节切分，得到拼音切分边界序列；

[0113] 本实施例中，基于每个候选拼音段的长度L、输入连续性C(无中断次数比例)和置信度P，计算其分节评分，评分函数如下：Score＝0.4×L_norm+0.3×C+0.3×P；其中L_norm为长度归一化值，最大设定为6，C值基于输入时序判断(如当前段无延迟即为1)，P为步骤S123所得置信度。以"zhong"为例：L＝5、C＝1、P＝0.9893，得分为：Score＝0.4×(5 / 6)+0.3×1+0.3×0.9893≈0.966；保留所有得分≥0.7的段落组合方案作为可用边界，并以得分最高者构建分节规则表，例如：[{"segment":"ni","start":0,"end":1,"score":0.87},{"segment":"hao","start":2,"end":4,"score":0.91},{"segment":"zhong","start":5,"end":9,"score":0.966},{"segment":"guo","start":10,"end":12,"score":0.924}]。

[0114] 步骤S125：对拼音切分边界序列进行拼音单元纠错扩展，从而构建拼音动作项集合；

[0115] 本实施例中，加载预设拼音纠错映射表，其中包含三类规则：声母替换(如)、韵母归并及音调误差容差(±1级音调调整)。在此基础上，系统对分节拼音段执行结构扩展，例如"zhong”生成纠错候选：["zhong","zong","zheng","zhuang"]。对片段流前后±2个拼音段构建上下文窗口，提取词首词尾拼音对的联合概率，如：{"hao→zhong":0.87,"zhong→guo":0.92}，将上下文置信加权与结构误差指数进行组合评价，设定重构保留阈值为0.6，仅保留："zhong"→{"zong":0.61,"zheng":0.72,"zhong":0.98}，最终保留置信度最高的"zhong”作为拼音动作项。

[0116] 步骤S126：将拼音动作项集合与拼音片段流中对应的起止位置、纠错状态与拼音类型进行对齐，生成拼音动作序列矩阵。

[0117] 本实施例中，将纠错后的拼音动作项与原始拼音片段流进行对应，按起止位置标记拼音段范围，同时标注纠错状态(如是否为原始匹配、是否为结构调整结果)，并对拼音类型(全拼、缩写、混合拼)进行分类。在上述案例中：[{"pinyin":"ni","start":0,"end":1,"corrected":false,"type":"full"},{"pinyin":"hao","start":2,"end":4,"corrected":false,"type":"full"},{"pinyin":"zhong","start":5,"end":9,"corrected":true,"type":"full"},{"pinyin":"guo","start":10,"end":12,"corrected":fals e,"type":"full"}]，将以上结构组织为拼音动作序列矩阵每行包含拼音内容、起止索引、纠错标志及拼音类型，用于驱动后续拼音-汉字映射及语义联动。

[0118] 可选地，步骤S125包括：

[0119] 提取拼音切分边界序列中的拼音子串，构建基础拼音单元列表；

[0120] 本实施例中，从已获得的拼音切分边界序列中，逐一提取所有拼音子串，形成基础拼音单元列表。例如，对输入序列["ni","hao","zhong","guo"]，基础拼音单元列表即为该数组。列表结构采用二维字符数组表示：拼音单元列表P_base＝[["ni"],["hao"],["zhong"],["guo"]]，其中每个元素对应一个拼音子串，列表长度即拼音单元数量，索引从0开始，便于后续定位。

[0121] 加载平台内置的拼音纠错映射表，包括声母替换规则、韵母归并规则及音调容差表；

[0122] 本实施例中，加载内置拼音纠错映射表。该映射表包括三大类规则，分别为声母替换、韵母归并和音调容差表。表结构以字典形式存储，如：声母替换规则S_rule＝{"n":["l"],"zh":["z","j"],...}；韵母归并规则Y_rule＝{"eng":["en"],"in":["ing"],...}；音调容差表T_rule＝{1:[1,2],2:[1,2,3],...}，其中声母替换规则覆盖常见混淆对，韵母归并规则兼顾发音相近归类，音调容差表允许音调上下浮动范围设定为±1级。

[0123] 根据拼音纠错映射表对基础拼音单元列表中各拼音子串进行声母-韵母错误匹配扩展，得到拼音纠错候选集；

[0124] 本实施例中，对基础拼音单元列表中每个拼音子串执行声母和韵母的错误匹配扩展。例如对拼音"zhong”，先分解为声母"zh”和韵母"ong”，分别查询S_rule和Y_rule，得到扩展集合：扩展声母集S_expand("zhong")＝["zh","z","j"]；扩展韵母集Y_expand("zhong")＝["ong","eng"]；拼接生成扩展拼音候选项集合：拼音纠错候选集C_candidat es("zhong")＝["zhong","zheng","zong","jeng"]，该扩展操作遵循严格长度匹配原则，确保新候选拼音长度与原拼音长度差异不超过1。

[0125] 提取拼音片段流中左右上下文区间±2个拼音单元的词首词尾拼音搭配概率，结合拼音纠错候选集中每个候选项的自身结构误差计算输入上下文误判概率，并根据输入上下午误判概率计算置信重构值，保留置信重构值≥0.6的候选项；

[0126] 本实施例中，提取拼音片段流中每个拼音单元在上下文区间±2范围内的词首词尾拼音搭配概率。该概率由预先统计的用户输入大数据词频库提供，库内存储格式如下：上下文搭配概率矩阵m为拼音单元数量。M_context[i,j]＝词首拼音i到词尾拼音j的搭配概率，取值范围[0,1]；例如，针对"hao”与"zhong”间的搭配，概率为0.87，若概率低于0.1则视为不常用搭配，影响候选置信度。结合拼音纠错候选集内各项自身的结构误差值(取值范围[0,1]，0表示完全匹配，0.3为最大容错阈值)，计算每个候选拼音的输入上下文误判概率。输入上下文误判概率计算公式内隐包含权重参数：输入上下文误判概率＝0.7×(1-词首词尾搭配概率)+0.3×结构误差值；置信重构值为该概率的补数，定义为：置信重构值＝1-输入上下文误判概率；例如，"zhong”候选"zheng”对应词首词尾概率为0.4，结构误差值0.2，则输入上下文误判概率为：＝0.7×(1-0.4)+0.3×0.2＝0.7×0.6+0.06＝0.42+0.06＝0.48，置信重构值＝1-0.48＝0.52<0.6，故该候选项被剔除。最后，保留置信重构值≥0.6的候选项作为有效纠错选项集合。

[0127] 将各拼音子串与相对应的拼音纠错候选集中置信重构值最高的候选项进行匹配，得到标准化拼音动作项集合。

[0128] 本实施例中，在标准化阶段，为每个拼音子串匹配其对应的候选集中置信重构值最高的项。该映射存储于如下矩阵结构：标准化拼音动作项集合每行包含：[原拼音子串,纠错后拼音,置信重构值]。例如：[["ni","ni",0.98],["hao","hao",0.95],["zhong","zhong",0.97],["guo","guo",0.99]]。该集合作为后续拼音动作序列构建的核心输入，保障输入拼音标准化一致性，有效提升了系统对拼音误输入的纠正能力与用户体验。

[0129] 可选地，步骤S2中汉字结构路径重建包括：

[0130] 利用字词结构索引表对拼音-汉字映射关系中每个汉字进行结构路径查询与定位，得到结构记录序列；

[0131] 本实施例中，利用预设的字词结构索引表对拼音-汉字映射关系中的每个汉字进行结构路径查询与定位。该索引表采用哈希映射结构存储，键为汉字Unicode编码，值为对应的结构路径记录。结构路径记录以二维矩阵形式表示，例如，汉字"林”(Unicode：U+6797)的结构路径记录序列如下：结构记录序列L_林＝[[主部件,子部件1,子部件2],["木”,"木”,null]]。其中，主部件和子部件均以汉字部件编码表示，null表示无下级部件。查询时，系统对输入的每个汉字检索该表，得到对应的结构路径矩阵，矩阵维度为n×3，n为部件层级数。

[0132] 基于结构记录序列进行结构层级展开，从而构建汉字结构拓扑图序列；

[0133] 本实施例中，以邻接矩阵形式构造拓扑图，矩阵维度m×m，m为部件总数。邻接矩阵元素为0或1，表示部件间的连接关系，例如：A_桥＝[[0,1,1],[0,0,0],[0,0,0]]，表示"木”指向"乔”及无其他子部件。矩阵的行和列分别对应汉字"桥”的部件节点，假设第1个节点是"木”，第2个是"乔”，第3个是另一个部件(比如"丿”或无具体部件，用null表示)。矩阵中的元素A_{ij}表示节点i是否指向节点j：如果A_{ij}＝1，表示从部件i指向部件j，即部件j是部件i的子部件或依附部件。如果A_{ij}＝0，表示部件i与部件j之间没有直接的层级连接关系。第一行[0,1,1]，具体含义是：A1,1＝0：部件"木”不指向自己(没有自环)；A1,2＝1：部件"木”直接连接指向"乔”，说明"乔”是"木”的子部件或者依赖部件；A1,3＝1：部件"木”也直接连接指向第三个部件，表明"木”与该部件也存在层级或依附关系。这样一来，"木”作为主部件，连接了"乔”和另一个部件，因此第一行是[0,1,1]，表示"木”指向这两个子部件。第二行和第三行全为0，表示"乔”和第三部件没有再往下的子部件。此拓扑图序列覆盖所有汉字的部件结构层级，体现部件间的父子关系，构成输入字词的结构骨架。

[0134] 将汉字结构拓扑图序列中各图从主部件向次部件的遍历路径进行线性编码，得到结构路径编码序列；

[0135] 本实施例中，采用深度优先遍历策略，按访问顺序将部件节点编码为字符串路径。编码格式如"M1->M2”，其中M1、M2为部件编号或简写。示例如下：编码序列＝["木->乔","乔->丿"]，编码序列以数组形式保存，长度取决于部件层级和数量，便于后续快速比较和匹配。

[0136] 对结构路径编码序列进行语义压缩，形成结构路径片段集合；

[0137] 本实施例中，设定相似度阈值α＝0.7，利用编辑距离计算编码序列之间的相似性，若相似度大于该阈值，则将两条路径合并为单一片段。压缩后的路径片段集合去除冗余，显著降低存储空间需求。示例压缩结果：片段集合＝["木->乔"]，结构路径片段以字符串数组形式存储，支持快速索引。

[0138] 将结构路径片段集合与拼音对齐输入帧中的用户输入行为ID绑定，生成用户输入字词结构路径集。

[0139] 本实施例中，用户输入行为ID采用32位整型唯一标识，建立关联矩阵M_bind(维度k×2)，第一列为行为ID，第二列为结构路径片段索引，实现一对多映射关系。例如：关联矩阵M_bind＝[[行为ID_1,片段索引_1],[行为ID_2,片段索引_1],...]；该矩阵用于快速定位用户输入行为对应的结构路径片段，构成完整的用户输入字词结构路径集，便于后续内容匹配和学习轨迹分析。

[0140] 可选地，步骤S3中拆分内容片段结构单元包括：

[0141] 根据平台用户交互数据中的用户输入行为ID和课程编号，通过在线汉语言学习平台获取对应课程内容集合，并加载对应课程内容集合中课程文本单元，从而得到学习内容数据；

[0142] 本实施例中，依据平台用户交互数据中的用户输入行为ID和课程编号，通过在线汉语言学习平台访问课程管理系统，调用课程内容数据库接口，获取对应课程编号下的完整课程内容集合。课程内容集合包括多层次文本资源，例如章节标题、自然段落、词汇表及示例句等结构化文本单元，所有文本单元均存储为UTF-8编码的字符串，统一格式方便后续解析。系统基于课程内容集合中的自然段落进行后续处理，保证文本完整性和上下文关联，加载得到学习内容数据。

[0143] 对学习内容数据中的自然段落进行语义切分，构建语义片段单元；

[0144] 本实施例中，针对学习内容数据中每个自然段落，通过语言理解模型内置的文本分割框架，将自然段落切分成语义连贯且主题一致的片段单元。该切分依据句法标点符号和句子主谓宾结构，同时结合上下文词向量相似度阈值0.65进行分割，确保语义片段长度保持在20至80个汉字范围内，避免片段过短导致语义碎片化或过长影响处理效率，最终构建出完整的语义片段单元集合。

[0145] 根据语义片段单元执行字词标注与结构特征标记，得到结构化标注结果；

[0146] 本实施例中，对构建的语义片段单元，执行逐字词级的标注与结构特征标记，应用预训练的汉语词性标注模型及词汇依存关系标注框架，获得包括词性类别(名词、动词、形容词等)、词组边界及层级依存关系矩阵(以邻接矩阵形式存储，形如A＝\begin{bmatrix}0&1\\0&0\end{bmatrix})的结构化标注结果。该矩阵用于指示词语之间的依存连接关系，便于后续语义分析和内容理解。

[0147] 对结构化标注结果进行语义紧密度分析，将语义紧密度≥0.75的语义片段标记为知识焦点，得到知识焦点集合；

[0148] 本实施例中，针对上述结构化标注结果，基于词向量加权平均方法计算每个语义片段的内在关联强度，设定阈值为0.75，筛选出语义紧密度高于该阈值的语义片段，标记为知识焦点。知识焦点集合通过节点列表形式存储，每个节点包含片段标识符、起止位置及紧密度得分，用于精确定位重点内容。

[0149] 根据知识焦点集合将语义片段单元进行内容片段单元划分，并对各内容片段单元进行时间戳重定位，得到内容片段结构单元。

[0150] 本实施例中，根据知识焦点集合，对原始语义片段单元进行重新划分，合并相邻且均属知识焦点的片段形成内容片段单元，同时将片段间时间戳基于用户学习行为日志进行校正，精确重定位每个内容片段的起止时间，时间戳精度达到毫秒级，确保内容片段结构单元在时间轴上的连续性和准确性。最终内容片段结构单元以包含文本内容、结构标识及时间戳的多字段数据结构保存，便于后续多模态嵌入与分析使用。

[0151] 可选地，步骤S3中多模态内容节点嵌入包括：

[0152] 将用户输入字词结构路径集进行结构拓扑向量化编码，得到结构向量编码集；

[0153] 本实施例中，用户输入字词结构路径集经过结构拓扑向量化编码时，首先基于每个汉字的部件层级关系，利用预设的部件编码字典将部件类别转换为向量形式。部件编码字典是一种预先构建的结构化数据表，将汉字的基本部件映射为唯一的数字编码，以便于计算机进行结构化分析和特征提取。该部件编码字典基于汉字部首表和标准汉字字典，通过形态学分析和图像特征提取相结合的方法，为每个部件生成统一的特征向量编码，包含形状、笔画和结构位置信息，确保编码的准确性和一致性。部件"口”：编码为[0.12,0.45,0.78,0.33,0.09,…](长度固定为64的浮点数向量)，表示其典型的方形轮廓、四笔画结构及在字中的常见位置特征。部件"氵”(三点水)：编码为[0.05,0.67,0.23,0.55,0.11,…]，体现其波浪形状特征、三点笔画数及偏旁位置属性。部件"亻”(单人旁)：编码为[0.22,0.31,0.48,0.39,0.15,…]，反映其竖笔形态、两笔结构和在字中的左侧偏旁位置。该字典支持动态更新与版本管理，为汉字结构解析和向量化表示提供了基础数据保障。每个部件编码为长度为128的定长特征序列，序列中的每个元素均为0-1之间的小数，反映部件激活权重。编码过程中，采用最大部件层级深度为5层，编码过程中部件权重衰减系数设为0.85，保证深层次部件贡献逐渐降低，避免冗余信息。编码结果保存为结构向量编码集，按字词顺序依次排列，形成列表形式，便于后续检索和匹配。

[0154] 根据内容片段结构单元提取对应课程内容集合中的文本词义嵌入、配套图文中汉字部件图形纹理特征、片段配音音频中的音素序列特征及文本语法标签分布特征，从而构建多模态片段特征集；

[0155] 本实施例中，对于内容片段结构单元，分别从文本、图形、音频及语法四个维度提取特征。文本维度采用预训练语言模型输出的词义嵌入，词向量维度为768维；图形维度通过图像处理提取汉字部件的纹理特征，采用64维向量表示；音频维度使用MFCC提取音素特征，长度固定为50维；语法维度统计文本中常见语法标签的分布，编码成30维多热向量。所有特征合并后构成多模态片段特征集，以列表形式存储，方便快速索引。每个特征维度均在0-1范围内，且提取过程中的归一化系数为0.99，防止极端值影响匹配。

[0156] 对结构向量编码集及多模态片段特征集执行双向匹配，得到关联匹配矩阵；

[0157] 本实施例中，对结构向量编码集与多模态片段特征集进行双向匹配时，采用相似度评分机制，具体为将两个向量组分别遍历，每对向量根据内积结果计算相似度得分，得分范围限定在0至1之间。匹配结果以"字词编号-内容片段编号-相似度得分”的三元组形式存储，例如(字词3，片段2，0.85)。设定阈值为0.7，只有得分高于该阈值的匹配关系才被保存，以保证匹配的精准度和实用性。

[0158] 从关联匹配矩阵中筛选匹配度≥0.7的匹配对映对，构建内容节点匹配集合；

[0159] 本实施例中，根据筛选出的匹配对，构建内容节点匹配集合。该集合以哈希表方式实现，键为字词结构路径编号，值为对应内容片段编号列表及其匹配得分。此结构便于快速查询字词对应的相关内容片段，实现内容与输入行为的高效关联。匹配对的数目根据具体输入量变化，一般单次会话匹配对数量控制在100以内，确保处理效率。

[0160] 根据内容节点匹配集合进行内容片段归组聚合，生成内容联动片段簇。

[0161] 本实施例中，根据内容节点匹配集合，执行内容片段归组聚合。将匹配度较高且时间戳接近的内容片段组合成簇，每个簇包含不超过10个片段，保证聚合的语义紧密性和计算效率。聚合规则基于时间差阈值0.5秒以内视为同簇，且同簇内片段的匹配得分均不低于0.7，避免无关内容混入。生成的内容联动片段簇以列表形式保存，包含簇内片段标识及其对应的匹配强度信息，便于后续拓扑轨迹构建和分析。

[0162] 可选地，步骤S4包括：

[0163] 步骤S41：提取内容联动片段簇中每个簇内的片段列表，并依据每个内容片段单元的触发时间对片段列表进行时间排序，得到初始片段轨迹序列；

[0164] 本实施例中，从内容联动片段簇里取出每个簇包含的所有学习内容片段，这些片段有各自的触发时间。接着，按照时间先后顺序对这些片段进行排列，确保学习轨迹顺序清晰，方便后续分析。时间戳格式采用标准的毫秒数表示，比如"1623456789123”，保证时间排序精准无误。通过这一步，可以得到用户在一次学习会话中，内容访问的时间序列，反映出学习的先后顺序。

[0165] 步骤S42：对初始片段轨迹序列进行语义相邻性分析，生成片段转移关系图；

[0166] 本实施例中，根据刚排好序的内容片段，分析每个片段与相邻片段之间的语义相似性。具体来说，会用已有的语义模型，将每个片段的文本转换成向量表示，再计算相邻片段向量的相似程度，数值范围0～1。设置一个阈值，比如0.6，超过这个阈值则认为两片段语义相关。还会结合课程知识点之间的关联度，综合判断片段间的转移关系，最终生成一个带权重的有向图，图中节点代表内容片段，边代表片段间的语义关联。

[0167] 步骤S43：计算片段转移关系图中的边集合的转移路径权重，并对转移路径权重<0.45的边进行剪枝，得到优化学习子轨迹集；

[0168] 本实施例中，对转移关系图中的每条边的权重进行定量计算，权重来源为语义相似度(权重0.6)和知识点关联度(0.4)的加权平均，权重阈值设定为0.45。低于该阈值的边被视为弱连接，予以剪除，以降低噪声影响和复杂度，从而得到简化且更具代表性的学习路径子集。此举有助于提高后续图结构聚合的准确性和计算效率。

[0169] 步骤S44：将优化学习子轨迹集进行有向图结构聚合，得到单次会话学习拓扑轨迹图；

[0170] 本实施例中，采用有向图聚合方法，将剪枝后的学习子轨迹集进行节点合并与路径整合，构建单次会话的学习拓扑轨迹图。图中节点仍为内容片段，边代表有效学习路径关系，整体反映用户在该会话中知识点访问及学习内容的拓扑结构。图结构的构建支持后续路径排序和轨迹分析，节点数一般控制在100以内以保持图的可读性与处理效率。

[0171] 步骤S45：对单次会话学习拓扑轨迹图进行双向拓扑排序，得到学习轨迹拓扑图。

[0172] 本实施例中，针对构建的学习拓扑轨迹图，分别执行正向和反向的节点序列扫描，形成双向拓扑排序。正向扫描依赖于节点入度为零的起点，反向扫描则以出度为零的终点开始，双向排序序列覆盖了所有可能的学习路径。最终，合并两方向排序结果，剔除冗余路径，形成完整且连贯的学习轨迹拓扑图，节点排序反映内容学习的时间及逻辑顺序，支持后续状态聚合和行为分析。

[0173] 尤其重要的是，步骤S45包括：

[0174] 步骤S451：对单次会话学习拓扑轨迹图进行前向拓扑扫描，生成正向片段路径集合；

[0175] 本实施例中，针对单次会话学习拓扑轨迹图从起始节点开始，依次遍历所有可达后续片段，形成正向片段路径集合。每条路径以内容片段的时间序列为基础，路径长度限定在不超过20个片段，确保路径完整同时避免计算资源过度消耗，形成结构化的有序路径列表。路径节点信息包括片段ID、触发时间及对应的语义标签，路径集合以列表形式存储，便于后续处理。

[0176] 步骤S452：对单次会话学习拓扑轨迹图进行逆向拓扑回溯，生成反向片段路径集合；

[0177] 本实施例中，从单次会话学习拓扑轨迹图中终止节点逆向遍历拓扑轨迹图，沿着可达的前驱节点依次回溯，生成反向片段路径集合。该集合与正向路径集合对应，最长路径同样限定为20个片段，确保前后路径长度一致性。路径数据结构同样包含片段ID、时间戳与语义标签，形成反向有序列表，便于交叉比对。

[0178] 步骤S453：根据正向片段路径集合及反向片段路径集合进行路径交并比对，筛选路径节点相交且重合比例≥0.6的路径对，得到一致路径子集；

[0179] 本实施例中，对正向片段路径集合及反向片段路径集合进行比对，针对每对路径计算节点集合的交集比例。若交集节点占比达到或超过60％(即相交比例≥0.6)，则判定该对路径为一致路径。该判断以节点ID集合形式执行，保证对比的准确性与效率。筛选出的路径对组成一致路径子集，作为下一步处理的基础。

[0180] 步骤S454：将一致路径子集进行顺序融合，消除冗余分支，保留路径主干，从而构建轨迹主路径链；

[0181] 本实施例中，将一致路径子集内的路径按时间序列顺序进行合并，消除重复节点和分支，保留路径中最具代表性的主干部分。合并过程中，依据节点的触发时间和语义标签保持路径逻辑连续性和语义一致性。最终形成轨迹主路径链，该链以序列结构保存，代表用户学习过程中的主要认知轨迹。

[0182] 步骤S455：将正向片段路径集合及反向片段路径集合中非属于轨迹主路径链的路径作为辅助路径，并将辅助路径的语义特征映射至轨迹主路径链中的相交节点中，得到学习轨迹拓扑图。

[0183] 本实施例中，将正向与反向路径集合中未包含于轨迹主路径链的路径识别为辅助路径，这些路径提供补充的学习轨迹信息。辅助路径中的语义特征被映射至主路径链中相交节点，丰富主路径的语义表达。映射操作通过节点ID匹配实现，确保语义特征准确对应，生成的结果构成完整的学习轨迹拓扑图，反映用户多层次的学习行为动态。

[0184] 可选地，步骤S5包括：

[0185] 步骤S51：对学习轨迹拓扑图进行语义阶段划分，生成语义阶段片段组；

[0186] 本实施例中，针对学习轨迹拓扑图中记录的片段，依据其语义特征进行阶段划分，形成若干语义阶段片段组。划分过程主要通过分析片段内关键词共现频率与主题一致性，设定相似度阈值为0.7，确保片段组内内容语义连贯。各阶段片段按时间顺序排列，且其对应的时间区间长度一般控制在1分钟以内，形成结构化的阶段列表，包含阶段标识、时间起止、主要语义标签等信息，便于后续处理和索引。

[0187] 步骤S52：依据语义阶段片段组中的起止时间间隔与结构跳跃度进行阶段聚合，构建学习状态演化链；

[0188] 本实施例中，通过对上述语义阶段片段组的时间起止点和结构跳跃度进行聚合，构建学习状态演化链。时间起止间隔阈值设为不超过30秒，结构跳跃度作为衡量连续两个阶段间内容差异的指标，其取值范围0～1，限定阈值不超过0.3以保证阶段间连贯。该演化链以有序时间段列表形式体现，反映学习过程中文本语义与结构的连续发展轨迹，为后续压缩和索引提供基础数据支持。

[0189] 步骤S53：基于学习状态演化链，提取主成分特征字段，从而构建汉语言状态压缩帧；

[0190] 本实施例中，基于学习状态演化链，对每个阶段片段组的多维特征数据进行主成分特征提取，目的是从高维语义及结构信息中提炼出最具代表性的核心指标。首先，将阶段组中所有包含的主题密度、结构复杂度、学习时长等原始特征整理成特征矩阵，矩阵行表示各个阶段片段，列对应各项指标。然后，针对该矩阵应用主成分选择机制，通过分析各特征的方差贡献率和相关性，筛选出方差贡献率累计超过85％的特征子集，确保压缩后特征集既保留大部分信息，又大幅减少冗余。所选特征主要包括主题密度(关键词出现频率比例)、结构复杂度(分层嵌套层数)、学习时长段(秒级时间长度)等。接着，对筛选出的特征进行加权融合，结合用户行为偏好权重(例如主题密度权重0.5，结构复杂度权重0.3，学习时长权重0.2)，计算出每个阶段的综合主成分特征值。最后，将各阶段的主成分特征字段以结构化数据形式保存，形成汉语言状态压缩帧，为后续多维索引建立和行为模式识别提供精准、简洁的特征基础。

[0191] 步骤S54：提取汉语言状态压缩帧中的主题密度、结构复杂度与学习时长段，将主题密度、结构复杂度与学习时长段转换为多主键组合，从而生成主键索引集合；

[0192] 本实施例中，对压缩帧中的主题密度、结构复杂度及学习时长段三项指标进行组合，形成多主键索引集合。索引格式采用三元组表示，例如(0.8,3,150)，对应主题密度0.8、结构复杂度3和学习时长150秒。所有索引均存储于多维索引表，方便基于多属性的快速查询和匹配，提升系统检索效率。

[0193] 步骤S55：将主键索引集合中具有同类主题密度与结构复杂度的主键聚合成行为模式，建立触发映射表；

[0194] 本实施例中，依据多主键索引集合，将具有相近主题密度(允许±0.05误差)和结构复杂度(允许±1误差)的索引项聚合，形成行为模式组。每组行为模式表示用户在特定语义和结构环境下的典型学习轨迹，如语法细节重点学习或词汇反复练习。基于这些行为模式，建立触发映射表，将模式映射到具体的学习触发事件，如学习提醒或重点推送，映射表采用键值对结构，支持实时响应。

[0195] 步骤S56：将主键索引集合与触发映射表进行归并打包，得到动态索引分布块，并将动态索引分布块传输至在线汉语言学习平台，以部署机制编排任务。

[0196] 本实施例中，将多主键索引集合与触发映射表进行归并打包，构成动态索引分布块。该分布块采用JSON格式封装，包含完整的索引数据和触发规则，便于在平台间传输和解析。动态索引分布块实时传输至在线汉语言学习平台，支持根据行为模式动态调度学习资源和任务，实现个性化教学管理，有效提升学习效率与体验。

[0197] 尤其重要的是，步骤S55具体为：

[0198] 提取主键索引集合中每个主键中包括主题密度、结构复杂度与学习时长段的三维属性，并基于主题密度对三维属性执行密度聚类，得到主题密度簇；

[0199] 本实施例中，从主键索引集合中抽取包含主题密度、结构复杂度与学习时长段的三维属性数据。针对主题密度维度，采用基于密度的空间聚类方法对样本点进行分类。具体地，设置邻域半径ε为0.05，最小邻域点数MinPts为6，确保聚类能够识别紧密的主题密度高聚集区。通过遍历样本数据，依次判定每个数据点是否属于核心点、边界点或噪声点，形成密度可达的点簇集合。聚类过程中，自动识别具有显著主题密度集中趋势的簇，剔除主题密度分布稀疏的异常数据。聚类完成后，基于主题密度簇的划分结果，将对应的三维属性分组归类，为后续结合结构复杂度及学习时长进行更细粒度的行为模式提取打下基础。此步骤有效地过滤了主题密度维度的噪声影响，提高整体聚合的准确性和稳定性。

[0200] 对主题密度簇中的索引项，按照结构复杂度进行K均值聚合，得到结构复杂度子簇；

[0201] 本实施例中，针对每个主题密度簇，继续根据结构复杂度属性划分子集。结构复杂度通过分析汉字部件层级数获得，数值范围为1至5，其中1代表结构简单(如单一偏旁)，5代表结构复杂(多层嵌套部件)。系统在主题密度簇内部，依据结构复杂度的数值区间(简单1-2级、中等3级、复杂4-5级)对主键进一步分组。该分组过程保证每个子集中的内容在结构层级上相对一致，有助于识别不同复杂度对学习行为的影响，实现更精准的行为模式划分。

[0202] 对结构复杂度子簇执行用户行为路径聚合，形成行为轨迹模板；

[0203] 本实施例中，在结构复杂度分组的基础上，利用用户学习行为轨迹数据，对同一子组内的学习路径进行聚合。学习轨迹数据以时间序列形式存储，包含内容访问顺序及对应停留时间，例如[(内容A,100秒),(内容B,200秒)]。系统采用路径相似度度量(设定阈值为0.8)对行为路径进行比对，识别相似轨迹，将相似行为轨迹归并为典型行为模板。该模板代表该类内容条件下用户的主要学习模式，便于后续的行为分析和个性化推荐设计。

[0204] 基于行为轨迹模板中的轨迹模版与统计指标，定义行为触发规则；

[0205] 本实施例中，结合行为模板及统计分析结果，定义行为触发规则。触发规则包括匹配度阈值(≥0.85)、访问频率限制和停留时间阈值(如300秒以上视为深度学习)。系统根据历史数据反馈动态调整阈值，确保规则既不过于宽松导致误触，也不过于严格导致漏判。该规则用于识别用户何时进入重点学习阶段，触发对应的辅助教学或资源推荐，提升学习效率和体验。

[0206] 将行为轨迹模板中的用户行为模式与对应行为触发规则进行映射，构建触发映射表。

[0207] 本实施例中，将行为模板与触发规则结构化整合，构建触发映射表。映射表采用JSON或表格形式存储，内容包含行为模式ID、轨迹模板内容ID序列、匹配度和触发条件参数等信息。例如：{

[0208] "模式ID”:"Pattern_001”,"轨迹内容”:["课文1段落3”,"课文1段落4”],"触发条件”:{"匹配度”:0.85,"停留时间”:300},"触发动作”:"推送练习题”}，该表支持实时查询与触发机制调用，保障平台智能反馈系统的高效运作，实现个性化动态学习指导。

[0209] 可选地，本说明书还提供一种在线汉语言学习平台的数据信息储存系统，用于执行如上所述的在线汉语言学习平台的数据信息储存方法，该在线汉语言学习平台的数据信息储存系统包括：

[0210] 汉字结构对齐模块，用于获取平台用户交互数据，并根据平台用户交互数据进行拼音输入轨迹解析与汉字部件结构编码，得到拼音对齐输入帧；

[0211] 映射关系分析模块，用于提取拼音对齐输入帧中的拼音-汉字映射关系，利用预设的字词结构索引表对拼音-汉字映射关系进行汉字结构路径重建，得到用户输入字词结构路径集；

[0212] 内容片段嵌入模块，用于通过在线汉语言学习平台调取与平台用户交互数据所关联的学习内容数据，并根据学习内容数据拆分内容片段结构单元；将用户输入字词结构路径集与内容片段结构单元进行多模态内容节点嵌入，得到内容联动片段簇；

[0213] 学习轨迹分析模块，用于根据内容联动片段簇构建单次会话学习拓扑轨迹，并对单次会话学习拓扑轨迹进行双向拓扑排序，得到学习轨迹拓扑图；

[0214] 学习状态聚合模块，用于对学习轨迹拓扑图进行学习状态分段聚合，生成汉语言状态压缩帧；根据汉语言状态压缩帧建立多维主键索引及行为触发机制，得到动态索引分布块，并将动态索引分布块传输至在线汉语言学习平台，以部署机制编排任务。

[0215] 因此，无论从哪一点来看，均应将实施例看作是示范性的，而且是非限制性的，本发明的范围由所附权利要求而不是上述说明限定，因此旨在将落在申请文件的等同要件的含义和范围内的所有变化涵括在本发明内。

[0216] 以上所述仅是本发明的具体实施方式，使本领域技术人员能够理解或实现本发明。对这些实施例的多种修改对本领域的技术人员来说将是显而易见的，本文中所定义的一般原理可以在不脱离本发明的精神或范围的情况下，在其它实施例中实现。因此，本发明将不会被限制于本文所示的这些实施例，而是要符合与本文所发明的原理和新颖特点相一致的最宽的范围。

Claims

1. A data information storage method for an online Chinese language learning platform, characterized in that, The method comprises the following steps: Step S1: obtaining platform user interaction data, and performing pinyin input track analysis and Chinese character component structure coding according to the platform user interaction data to obtain a pinyin alignment input frame; Step S1 comprises: Step S11: obtaining platform user interaction data, and performing unified timestamp coding to obtain an original input behavior record set; Step S12: performing dynamic pinyin segmentation on the original input behavior record set, and labeling pinyin segmentation boundaries to obtain a pinyin action sequence matrix; Step S13: calling a preset local word library index to perform pinyin-Chinese character mapping alignment on the pinyin action sequence to obtain a pinyin-Chinese character alignment table; Step S14: performing Chinese character component structure coding on the pinyin-Chinese character alignment table to generate a component structure vector set; Step S15: associating and fusing the pinyin action sequence matrix and the component structure vector set according to the alignment relationship in the pinyin-Chinese character alignment table to generate a pinyin alignment input frame; Step S2: extracting a pinyin-Chinese character mapping relationship in the pinyin alignment input frame, and performing Chinese character structure path reconstruction on the pinyin-Chinese character mapping relationship by using a preset word structure index table to obtain a user input word structure path set; the Chinese character structure path reconstruction in step S2 comprises: performing structure path query and positioning on each Chinese character in the pinyin-Chinese character mapping relationship by using the word structure index table to obtain a structure record sequence; performing structure level expansion based on the structure record sequence to construct a Chinese character structure topology graph sequence; linearly coding traversal paths of each graph in the Chinese character structure topology graph sequence from a main component to a secondary component to obtain a structure path coding sequence; performing semantic compression on the structure path coding sequence to form a structure path segment set; binding the structure path segment set and a user input behavior ID in the pinyin alignment input frame to generate a user input word structure path set; Step S3: calling learning content data associated with the platform user interaction data through an online Chinese language learning platform, and splitting content segment structure units according to the learning content data; performing multi-modal content node embedding on the user input word structure path set and the content segment structure units to obtain a content linkage segment cluster; Step S4: constructing a single-session learning topology track according to the content linkage segment cluster, and performing bidirectional topology sorting on the single-session learning topology track to obtain a learning track topology graph; step S4 comprises: Step S41: extracting a segment list in each cluster in the content linkage segment cluster, and performing time sorting on the segment list according to a trigger time of each content segment unit to obtain an initial segment track sequence; Step S42: performing semantic adjacency analysis on the initial segment track sequence to generate a segment transition relationship graph; Step S43: calculating transition path weights of an edge set in the segment transition relationship graph, and pruning edges with a transition path weight less than 0.45 to obtain an optimized learning sub-track set; Step S44: performing directed graph structure aggregation on the optimized learning sub-track set to obtain a single-session learning topology track graph; Step S45: performing bidirectional topology sorting on the single-session learning topology track graph to obtain a learning track topology graph; Step S5: learning state segmentation aggregation is performed on the learning trajectory topology graph to generate Chinese language state compression frames; multi-dimensional primary key indexes and behavior triggering mechanisms are established according to the Chinese language state compression frames to obtain dynamic index distribution blocks, and the dynamic index distribution blocks are transmitted to an online Chinese language learning platform to arrange tasks by means of a mechanism.

2. The data information storage method of the online Chinese language learning platform according to claim 1, wherein, Step S12 includes: Step S121: a character input sequence in the original input behavior record set is extracted, and a pinyin segment stream is constructed; Step S122: the pinyin segment stream is divided into pinyin sub-strings according to a preset initial-middle-final three-tuple matching logic to obtain a candidate pinyin segment set; Step S123: pinyin distribution statistics are performed on the candidate pinyin segment set, and syllable confidence values are calculated according to the pinyin distribution statistics results; Step S124: the candidate pinyin segment set is weighted and scored based on the pinyin length, input continuity and corresponding syllable confidence values, the highest-scoring sub-string combination is selected as a segmentation boundary rule, and the candidate pinyin segment set is segmented according to the segmentation boundary rule to obtain a pinyin segmentation boundary sequence; Step S125: pinyin unit error correction expansion is performed on the pinyin segmentation boundary sequence to construct a pinyin action item set; Step S126: the pinyin action item set is aligned with the corresponding start and end positions, error correction states and pinyin types in the pinyin segment stream to generate a pinyin action sequence matrix.

3. The data information storage method of the online Chinese language learning platform according to claim 2, wherein, Step S125 includes: A pinyin sub-string in the pinyin segmentation boundary sequence is extracted to construct a basic pinyin unit list; A built-in pinyin error correction mapping table of the platform is loaded, including initial replacement rules, middle merging rules and tone tolerance tables; According to the pinyin error correction mapping table, each pinyin sub-string in the basic pinyin unit list is expanded for initial-middle-final error matching to obtain a pinyin error correction candidate set; The word initial and final pinyin collocation probability of the context interval ±2 pinyin units in the pinyin segment stream is extracted, the input context misjudgment probability is calculated in combination with the structure error of each candidate item in the pinyin error correction candidate set, and the confidence reconstruction value is calculated according to the input context misjudgment probability, and the candidate items with a confidence reconstruction value ≥0.6 are retained; Each pinyin sub-string is matched with the candidate item with the highest confidence reconstruction value in the corresponding pinyin error correction candidate set to obtain a standardized pinyin action item set.

4. The data information storage method of the online Chinese language learning platform according to claim 1, wherein, The content segment structure unit in step S3 includes: According to the user input behavior ID and course number in the platform user interaction data, the corresponding course content set is obtained through the online Chinese language learning platform, and the course text unit in the corresponding course content set is loaded to obtain learning content data; The natural paragraphs in the learning content data are semantically segmented to construct semantic segment units; Word labeling and structural feature marking are performed on the semantic segment units to obtain a structured labeling result; Semantic closeness analysis is performed on the structured labeling result, and semantic segments with a semantic closeness ≥0.75 are marked as knowledge focuses to obtain a knowledge focus set; The semantic segment units are divided into content segment units according to the knowledge focus set, and the time stamp of each content segment unit is relocated to obtain a content segment structure unit.

5. The data information storage method of the online Chinese language learning platform according to claim 1, wherein, The multi-modal content node embedding in step S3 includes: The user input word structure path set is topologically vectorized to obtain a structure vector coding set; According to the content segment structure unit, the text word meaning embedding in the corresponding course content set, the Chinese character component graphic texture feature in the matching text and image, the phoneme sequence feature in the segment dubbing audio, and the text grammar tag distribution feature are extracted to construct a multi-modal segment feature set; Bidirectional matching is performed on the structure vector coding set and the multi-modal segment feature set to obtain an association matching matrix; From the association matching matrix, a matching pair mapping pair with a matching degree greater than or equal to 0.7 is selected to construct a content node matching set; According to the content node matching set, content segment grouping and aggregation are performed to generate a content linkage segment cluster.

6. The data information storage method for the online Chinese language learning platform according to claim 1, characterized in that, Step S5 includes: Step S51: The semantic stage division is performed on the learning trajectory topology graph to generate a semantic stage segment group; Step S52: The stage aggregation is performed according to the start and end time interval and the structure jump degree in the semantic stage segment group to construct a learning state evolution chain; Step S53: Based on the learning state evolution chain, the principal component feature field is extracted to construct a Chinese language state compression frame; Step S54: The theme density, structure complexity, and learning time length segment in the Chinese language state compression frame are extracted, the theme density, structure complexity, and learning time length segment are converted into a multi-primary key combination, and a primary key index set is generated; Step S55: The primary keys with the same theme density and structure complexity in the primary key index set are aggregated into a behavior mode, and a trigger mapping table is established; Step S56: The primary key index set and the trigger mapping table are merged and packaged to obtain a dynamic index distribution block, and the dynamic index distribution block is transmitted to an online Chinese language learning platform to deploy a mechanism to arrange tasks.

7. A data information storage system of an online Chinese language learning platform, characterized in that, A data information storage method for executing an online Chinese language learning platform as claimed in claim 1, the data information storage system of the online Chinese language learning platform comprising: A Chinese character structure alignment module for obtaining platform user interaction data and performing pinyin input trajectory analysis and Chinese character component structure coding according to the platform user interaction data to obtain a pinyin alignment input frame; A mapping relationship analysis module for extracting a pinyin-Chinese character mapping relationship in the pinyin alignment input frame, reconstructing a Chinese character structure path of the pinyin-Chinese character mapping relationship by using a preset word structure index table, and obtaining a user input word structure path set; A content segment embedding module for calling learning content data associated with the platform user interaction data through the online Chinese language learning platform, splitting content segment structure units according to the learning content data, and performing multi-modal content node embedding on the user input word structure path set and the content segment structure units to obtain a content linkage segment cluster; A learning trajectory analysis module for constructing a single-session learning topology trajectory according to the content linkage segment cluster, and performing bidirectional topology sorting on the single-session learning topology trajectory to obtain a learning trajectory topology graph; A learning trajectory analysis module for constructing a single-session learning topology trajectory according to the content linkage segment cluster, and performing bidirectional topology sorting on the single-session learning topology trajectory to obtain a learning trajectory topology graph; The learning state aggregation module is configured to aggregate learning state segments of the learning trajectory topology to generate a Chinese language state compression frame, establish a multi-dimensional primary key index and a behavior triggering mechanism according to the Chinese language state compression frame, obtain a dynamic index distribution block, and transmit the dynamic index distribution block to the online Chinese language learning platform to arrange tasks by means of a deployment mechanism.

Images