Big data-based english cross-scene learning data integration and teaching collaborative management system

Abstract

The application discloses an English cross-scene learning data integration and teaching collaborative management system based on big data, and relates to the technical field of teaching collaborative management.The system comprises a data perception and gathering module, an integration effective accuracy scheduling module, a dynamic adaptive response quantification module and a dynamic adaptive response scheduling module.The application collects multi-scene learning behavior flow through the data perception module, calculates and integrates the effective accuracy rate to quantify data quality, judges whether to trigger data accuracy scheduling, judges whether to execute response scheduling based on the teaching collaborative dynamic adaptive response rate to optimize teaching synchronization, decides to enter the next round of data integration cycle or directly perform data life cycle management according to the scheduling result, realizes closed-loop self-adaptive regulation and control of data quality and teaching efficiency, and solves the problem of low dynamic responsiveness of English cross-scene learning data integration and teaching collaborative management due to the heterogeneity and inefficiency of English cross-scene learning data pipeline in the prior art.

Description

Technical Field

[0001] This invention relates to the field of collaborative teaching management technology, and in particular to a big data-based cross-scenario English learning data integration and collaborative teaching management system. Background Technology

[0002] Firstly, in the multi-source data collection stage, classroom interaction data (such as smart blackboard answer records and pronunciation data from the voice assessment system) is captured through IoT devices. After-class self-study trajectories (including vocabulary memorization progress, reading time, and listening practice frequency) are collected through mobile learning apps. Group collaborative discussion records and cross-cultural communication dialogue audio are captured through social learning platforms. At the same time, offline simulated scenario training data such as oral scenario simulation videos and OCR (Optical Character Recognition) recognition results of paper scans of writing tasks are also integrated to achieve comprehensive data collection across online and offline scenarios.

[0003] Next, the data integration and processing stage begins. ETL (Extract-Transform-Load) tools are used to clean, transform, and standardize heterogeneous data. Big data distributed storage technologies (such as Hadoop) are used to securely store massive amounts of structured data (answer accuracy, learning time) and unstructured data (audio and written text). Natural language processing technology is then used to analyze grammatical errors and semantic logic in the text data. Speech recognition and evaluation technologies are used to transform spoken data into quantifiable indicators of pronunciation accuracy and fluency. Machine learning algorithms are combined to build personalized learning profiles for students. At the same time, blockchain technology is used to ensure data traceability and privacy security.

[0004] Finally, in the application of collaborative teaching management, based on the integrated data analysis results, the intelligent teaching platform enables collaborative interaction among teachers, students, and administrators: teachers obtain data on students' weaknesses through the AI ​​lesson preparation system, generate customized lesson plans and differentiated assignments, and conduct targeted tutoring using the real-time interactive platform; students obtain suitable learning resources and tasks through the personalized learning path recommendation function and participate in collaborative learning with the help of intelligent peer assessment tools; administrators monitor teaching quality and resource utilization efficiency in real time through data visualization dashboards, and coordinate the allocation of teaching resources and the division of labor among teachers using the collaborative office system.

[0005] For example, the multifunctional university English teaching management device disclosed in Chinese invention patent CN103440563A includes: an English teaching attendance module, an English teaching testing module, a multi-major English self-study module, and other teaching function modules. The front of the housing is equipped with a storage box, a fingerprint reader, a scanner, a USB interface, a microphone, a speaker, a touch screen display, a scrolling information display screen, a translation switch, and a retrieval switch. The sides of the housing are equipped with a power interface, a data interface, a working switch, and a serial port interface. The top of the housing is equipped with a working indicator light, a translation status indicator light, a box retrieval success indicator light, and a data transmission success indicator light. The housing contains a central controller, a storage unit, a translator, and a data transmission system.

[0006] For example, Chinese invention patent CN119671805A discloses an English course management system for English teaching, which includes: collecting multi-source English teaching data and cleaning and extracting features; constructing a knowledge graph based on the extracted feature parameters; calculating course configuration parameter factors using feature filtering and weighting formulas based on the knowledge graph; inputting the course configuration parameter factors into a deep reinforcement learning model for strategy training; and outputting the optimal combination of course content and difficulty configuration in the current teaching environment and dynamically updating the teaching strategy.

[0007] However, in the process of implementing the inventive technical solution in the embodiments of this application, it was found that the above-mentioned technology has at least the following technical problems:

[0008] In existing technologies, the integration of cross-scenario English learning data and collaborative teaching management face specific and interconnected problems. Firstly, at the data integration level, due to unreasonable technical architecture design, existing ETL tools lack sufficient parallel processing and dynamic scheduling capabilities when dealing with massive, high-concurrency, multi-dimensional learning data. This results in inefficient extraction, transformation, and cleaning processes, easily leading to data loss and duplication. Secondly, due to the domain adaptability limitations of natural language processing technology, existing models are generally trained on general corpora, lacking specialized corpora for English learning errors and cross-cultural expressions. This leads to biases in semantic understanding of polysemous words, slang, and grammatical errors when parsing writing and dialogue content, severely impacting the accuracy of learning profile construction. Furthermore, data storage solutions suffer from a lack of functional balance. Whether it's distributed storage that prioritizes security and encryption at the expense of read speed, or blockchain technology that suffers from low query efficiency due to decentralization, neither can meet the comprehensive requirements of high security, high access frequency, and scalability for English learning data. Technology selection tends to be singular, failing to achieve synergistic optimization of security, efficiency, and scalability.

[0009] Secondly, in the collaborative application of teaching, due to the fundamental defects in the training data and model design of AI algorithms, the learning profiles rely on incomplete data lacking cognitive dimensions, and the models lack dynamic adjustment mechanisms. As a result, the resource recommendations and lesson plan designs provided by the intelligent teaching platform remain at the level of common group analysis, and cannot accurately adapt to the individual weaknesses and real-time changes in students' learning status, resulting in a serious lack of personalized teaching adaptation.

[0010] In terms of collaboration, due to limitations in data transmission bandwidth and the lack of uniformity in communication protocols between heterogeneous platforms, cross-scenario data synchronization faces problems such as transmission congestion and protocol incompatibility. This results in significant delays in data updates between teachers, students, and administrators, making it impossible to guarantee the real-time nature of teaching feedback. Furthermore, the heterogeneity and inefficiency of cross-scenario English learning data pipelines lead to low dynamic responsiveness in the integration of cross-scenario English learning data and collaborative teaching management. Summary of the Invention

[0011] To address the technical problem of low dynamic responsiveness in English cross-scenario learning data integration and collaborative teaching management caused by the heterogeneity and inefficiency of existing English cross-scenario learning data pipelines, this invention provides a big data-based English cross-scenario learning data integration and collaborative teaching management system. The technical solution is as follows:

[0012] On the one hand, a big data-based English cross-scenario learning data integration and teaching collaboration management system is provided. This system includes: a data perception and aggregation module, an integration effectiveness accuracy scheduling module, a dynamic adaptation response quantification module, and a dynamic adaptation response scheduling module. The data perception and aggregation module collects heterogeneous learning behavior data streams from multiple independent English learning application scenarios and performs data integration. It obtains integration effectiveness parameters during the integration process and, based on these parameters, obtains the English cross-scenario data integration effectiveness accuracy rate, which quantifies the effectiveness and accuracy of the learning data collected across English scenarios during the integration process. The integration effectiveness accuracy scheduling module determines whether to execute integration effectiveness accuracy scheduling based on the English cross-scenario data integration effectiveness accuracy rate to maintain the accuracy when the data integration quality meets the standard. An efficient processing flow enhances the reliability of cross-scenario learning. If the data lifecycle management step is performed after scheduling, it proceeds directly; otherwise, it proceeds directly. The dynamic adaptation response quantification module acquires adaptation response parameters from the data lifecycle management step and obtains the teaching collaboration dynamic adaptation response rate based on these parameters. This rate is used to quantify the adaptation and response effectiveness of the entire teaching collaboration process across English scenarios. The dynamic adaptation response scheduling module determines whether to execute dynamic adaptation response scheduling based on the teaching collaboration dynamic adaptation response rate. This ensures real-time synchronization between the teaching process and learners' cognitive states, improving the overall efficiency of collaborative learning. If the data integration is successful, it proceeds directly; otherwise, it proceeds directly.

[0013] Beneficial effects

[0014] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0015] 1. By collecting heterogeneous learning behavior data streams from multiple independent English learning application scenarios and performing data integration, effective integration parameters are obtained. Based on these parameters, the effective accuracy rate of cross-scenario English data integration is obtained, which quantifies the effectiveness and accuracy of learning data collected across English scenarios during the integration process. The effectiveness accuracy rate determines whether to execute integration accuracy scheduling to automatically trigger data cleaning and verification processes when data integration quality is substandard, ensuring high credibility of data input to the downstream teaching collaboration system and thus guaranteeing the reliability of learning analysis and decision-making. 2. Adaptation response parameters are obtained from the data lifecycle management stage, and the dynamic adaptation response rate of teaching collaboration is obtained, quantifying the adaptability and response effectiveness of the entire teaching collaboration process across English scenarios. The dynamic adaptation response rate determines whether to execute dynamic adaptation response scheduling, which automatically triggers dynamic optimization of teaching strategies and resource allocation when a mismatch between teaching pace and learner status is detected, ensuring that the teaching process remains synchronized with the learner's cognitive progress, thereby improving overall teaching efficiency and personalized learning outcomes, and achieving closed-loop adaptive control of data quality and teaching effectiveness.

[0016] 2. The learning data retention period is dynamically adjusted based on the total amount of English cross-scenario data collected. When the total amount of data increases, the retention time of non-core data can be automatically shortened, and when the total amount of data decreases, the retention period of high-value data can be extended. This achieves intelligent balance and continuous optimization between storage resource costs and the reuse value of historical data. The query latency threshold of cross-scenario data availability zones is dynamically adjusted based on the number of concurrent access connections. Under high concurrency pressure, latency tolerance is proactively tightened to prioritize service availability, while under low load, restrictions are relaxed to improve the completeness and accuracy of query results. This achieves dynamic balance and optimization between system throughput and data service quality, and realizes closed-loop adaptive control of data quality and teaching effectiveness.

[0017] 3. The intervention response time window width is dynamically adjusted based on the number of effective responses to teaching collaboration. This automatically extends the waiting and observation period when there are insufficient effective responses and shortens the window to accelerate the pace when there are sufficient responses. This dynamically matches the cognitive load and participation status of the learning group, achieving precise synchronization between the timing of teaching intervention and individual learning progress. The real-time interactive data processing window threshold is dynamically adjusted based on the teaching feedback response time. This automatically widens the processing window to ensure the depth of analysis when the group response is slow and narrows the window to improve the real-time feedback when the response is rapid. This achieves a dynamic fit between the data processing rhythm and the frequency of teaching interaction, optimizing the timeliness and accuracy of teaching decisions, and realizing closed-loop adaptive control of data quality and teaching effectiveness. Attached Figure Description

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

[0019] Figure 1 A schematic diagram of the structure of the big data-based English cross-scenario learning data integration and teaching collaboration management system provided in this application embodiment;

[0020] Figure 2 Flowchart of adaptive adjustment of query latency threshold for the big data-based English cross-scenario learning data integration and teaching collaboration management system provided in this application embodiment;

[0021] Figure 3 The flowchart illustrates the adaptive adjustment of the response time window width in the big data-based English cross-scenario learning data integration and teaching collaboration management system provided in this application embodiment. Detailed Implementation

[0022] The technical solution provided in this application will now be described with reference to the accompanying drawings.

[0023] To facilitate understanding of the embodiments of this application, the following points will be explained first:

[0024] First, in this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates an "or" relationship between the preceding and following related objects, but it does not exclude the possibility of indicating an "and" relationship; the specific meaning can be understood in context. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c; a and b; a and c; b and c; or a and b and c. Here, a, b, and c can be single or multiple.

[0025] Second, the use of prefixes such as "first" and "second" in this application is solely for the purpose of distinguishing and describing different things belonging to the same category, and does not constrain the order, size, or quantity of things. For example, "first message" and "second message" are simply different messages, and there is no chronological, size, or priority relationship between them.

[0026] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0027] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0028] like Figure 1 The diagram shown is a structural schematic of the English cross-scenario learning data integration and teaching collaboration management system based on big data provided in this application embodiment. The English cross-scenario learning data integration and teaching collaboration management system based on big data provided in this application embodiment includes: a data perception and aggregation module, an integration effectiveness and accuracy scheduling module, a dynamic adaptation response quantification module, and a dynamic adaptation response scheduling module.

[0029] The data perception and aggregation module is used to collect heterogeneous learning behavior data streams from multiple independent English learning application scenarios, and to perform data integration. It obtains effective integration parameters in the data integration process, and obtains the effective accuracy rate of cross-scenario English data integration based on the effective integration parameters. This is used to quantify the effectiveness and accuracy of the learning data collected in cross-scenario English learning in the integration process.

[0030] It needs to be explained that the specific steps to obtain the effective accuracy rate of cross-scenario data integration in English are as follows:

[0031] The effective parameters to be integrated include ETL dynamic scheduling response time, learning data integration rate, and number of data integration connection timeout retries. ETL dynamic scheduling response time refers to the total time taken from receiving a high-concurrency data processing request trigger signal to initiating dynamic resource scheduling (such as adjusting the number of processing threads, allocating server nodes, and splitting task priorities) until the scheduling configuration is completed and the data extraction-transformation-cleaning task begins execution. The difference between the timestamp of completing the scheduling configuration and starting execution extracted from the ETL system's built-in log module and the timestamp of receiving the high-concurrency data processing request trigger signal is recorded as the ETL dynamic scheduling response time. Learning data integration rate refers to the amount of high-quality data successfully converted per unit time in the entire English cross-scenario data integration process. High-quality data refers to data without loss, duplication, or semantic bias. The learning data integration rate is obtained by extracting the amount of data without loss / duplication from the ETL processing logs, combining it with the NLP (Natural Language Processing) parsing results report to filter semantically bias-free data, and combining the results. Data integration connection timeout retries refer to the actual number of times that automatic connection retries are initiated during the English cross-scenario data integration process. The data integration connection timeout retries are obtained by filtering connection timeout-related records from the ETL's built-in connection management module logs and extracting the number of retries corresponding to each timeout record.

[0032] The response time weighting coefficient and response time threshold are interactively processed with the results of the ETL dynamic scheduling response time ratio analysis to obtain the response time impact value; the integration rate weighting coefficient is interactively processed with the results of the learning data integration rate and integration rate threshold ratio analysis to obtain the integration rate impact value; the retry count weighting coefficient and retry count threshold are interactively processed with the results of the data integration connection timeout retries ratio analysis to obtain the retry count impact value; the response time impact value, integration rate impact value, and retry count impact value are coupled to obtain the effective accuracy rate of English cross-scenario data integration. The specific constraint expression for the effective accuracy rate of English cross-scenario data integration is as follows:

[0033] ;

[0034] In the formula, R represents the effective accuracy rate of cross-scenario data integration in English; s1 represents the response time weight coefficient obtained from the teaching collaboration management database; s2 represents the integration rate weight coefficient obtained from the teaching collaboration management database; s3 represents the retry count weight coefficient obtained from the teaching collaboration management database; T0 represents the response time threshold obtained from the teaching collaboration management database; M0 represents the integration rate threshold obtained from the teaching collaboration management database; N0 represents the retry count threshold obtained from the teaching collaboration management database; T represents the ETL dynamic scheduling response time; M represents the learning data integration rate; and N represents the number of data integration connection timeout retries.

[0035] It's important to explain that a longer ETL dynamic scheduling response time leads to resource scheduling lag when dealing with concurrent data from multiple scenarios (such as untimely CPU / memory allocation and task queue backlog). This results in a delay in the time it takes for the first piece of data to enter the processing flow, causing subsequent processing stages to "idle" due to insufficient resources or task congestion. Furthermore, high-concurrency data is prone to partial data loss / duplication due to untimely scheduling, leading to a lower learning data integration rate. A longer ETL dynamic scheduling response time also results in insufficient resource scheduling capabilities when dealing with data from multiple scenarios (such as insufficient thread allocation and uneven node load). This leads to excessive consumption of communication resources between the ETL system, data source, and storage nodes, causing insufficient bandwidth in network transmission channels due to resource congestion, and data transmission connections are prone to timeouts. Simultaneously, scheduling lag causes some connection requests to queue, triggering retries after exceeding the preset timeout threshold, thus increasing the number of data integration connection timeout retries. The more data integration connection timeout retries, the more delays subsequent data transmission occur, potentially causing the ETL processing module to experience "data supply interruptions." When idle, the NLP parsing module cannot work continuously because it is waiting for data; at the same time, multiple retries will fail and data will be lost. Additional time is needed to supplement the collection or clean up invalid data. The amount of effective data to be integrated per unit time is greatly reduced, and the learning data integration rate is lower. Meanwhile, there is a negative correlation between ETL dynamic scheduling response time and the effective accuracy of cross-scenario data integration in English. The longer the ETL dynamic scheduling response time, the more delayed resource scheduling becomes, leading to untimely CPU / memory allocation and task queue accumulation. High-concurrency data is prone to "overload" issues, resulting in missed extractions and duplicate imports, thus lowering the effective accuracy of cross-scenario data integration in English. Conversely, there is a positive correlation between learning data integration rate and the effective accuracy of cross-scenario data integration in English. A higher learning data integration rate indicates an efficient closed loop formed in the data preprocessing, ETL processing, NLP parsing, and storage stages, resulting in a higher effective accuracy of cross-scenario data integration in English. Furthermore, there is a negative correlation between the number of data integration connection timeout retries and the effective accuracy of cross-scenario data integration in English. The more data integration connection timeout retries, the more transmission channel and system resources are consumed, leading to data loss due to "timeout retransmission" or data corruption due to network fluctuations during transmission, thus lowering the effective accuracy of cross-scenario data integration in English.

[0036] The integrated effective accuracy scheduling module is used to determine whether to perform integrated effective accuracy scheduling based on the integrated effective accuracy rate of English cross-scenario data. This ensures that when the data integration quality meets the standards, the efficient processing flow is maintained and the reliability of cross-scenario learning is improved. If so, the data lifecycle management stage is performed after scheduling; otherwise, the data lifecycle management stage is performed directly.

[0037] Furthermore, the specific steps to determine whether to perform integrated effective and accurate scheduling are as follows:

[0038] If the effective accuracy rate of cross-scenario data integration in English is higher than the effective accuracy reference value, then the integration effective accuracy scheduling will not be executed; otherwise, the integration effective accuracy scheduling will be executed. The integration effective accuracy scheduling includes adaptive adjustment of retention period and adaptive adjustment of query latency threshold.

[0039] As further explained in detail, the specific process for adaptive adjustment of retention period is as follows:

[0040] The total data volume of English learning behavior data streams from multiple independent application scenarios is aggregated and calculated in real time.

[0041] Based on a preset total data volume threshold range and a mapping relationship between the total data volume and retention period, the overall retention period for learning data across all scenarios is dynamically adjusted. This achieves an intelligent balance and continuous optimization between storage resource costs and the value of reusing historical data.

[0042] Historical data exceeding the dynamically adjusted retention period will be automatically archived and deleted. A predefined non-increasing function is used, which is: Where T represents the learning data retention period, G represents the total amount of English cross-scenario data collected within a preset time window, and the function f(G) passes through a set of discrete threshold intervals {[S0, S1), [S1, S2), [S2, S3)...[S...]}. n-1 S n )}, and the corresponding retention period values ​​{T0,T1,T2..T n-1 To achieve this; {S0<S1<S2<S3..<S n}, {T0>T1>T2>..T n-1}

[0043] Adaptive retention period adjustment also includes:

[0044] The total amount of aggregated English cross-scenario data is calculated at a preset frequency. This total amount of data is then compared to a preset threshold interval sequence to determine its specific interval. If the learning data retention period corresponding to the specific interval to which the total amount of aggregated English cross-scenario data belongs differs from the currently effective learning data retention period, an adjustment mechanism is triggered. By introducing a data quality factor for correction, the period is further tightened when data quality declines, prioritizing the elimination of low-quality data. This optimizes storage efficiency while ensuring the effectiveness of basic data assets.

[0045] The adjustment mechanism includes: periodic updates and recalculation of the lifecycle of existing data. The specific steps for periodic updates are as follows:

[0046] The total amount of English cross-scenario data collected and the effective accuracy offset are input into the total data volume-retention period mapping relationship to obtain the retention period correction factor. The current learning data retention period and the retention period correction factor are weighted and fused to obtain the target learning data retention period. The effective accuracy offset represents the negative difference between the effective accuracy rate of English cross-scenario data integration and the effective accuracy reference value. This ensures that global strategy changes can be applied to all existing data in a timely and consistent manner, realizing centralized, accurate and automated management of the data lifecycle and avoiding the disconnect between strategy and execution.

[0047] The specific steps for recalculating the lifecycle of existing data are as follows:

[0048] Iterate through all stored historical data entries. For each historical data entry, extract its lifecycle anchor timestamp. Couple the target learning data retention period with the lifecycle anchor timestamp to obtain the target absolute failure timestamp. Based on the comparison between the target absolute failure timestamp and the current failure timestamp, synchronously update the time status flag of the data entry.

[0049] If the current expiration timestamp is greater than or equal to the target absolute expiration timestamp, then the expiration status flag of the data entry will be updated to pending cleanup.

[0050] If the current expiration timestamp is less than the target absolute expiration timestamp, the expiration status flag of the data entry is updated to valid. Through the above closed-loop adaptive control and synchronization mechanism, the system ultimately achieves continuous and precise release of storage space based on the dynamic changes in data scale and quality without human intervention, and prioritizes the use of limited storage resources to retain high-value, high-quality learning data, thereby improving the overall health and management efficiency of data assets.

[0051] In this embodiment, by monitoring the total amount of data across scenarios in real time and dynamically adjusting the global data retention period based on a preset non-increasing function mapping relationship, an adaptive balance between storage costs and data value is achieved. The system introduces a data quality offset to correct the periodic decision, ensuring that inefficient data is prioritized for removal when data quality deteriorates. By synchronously recalculating the lifecycle of all existing data and updating its expiration status, the system can automatically identify and clean up expired data. Overall, this solution constructs a closed-loop intelligent storage management system that, while ensuring the retention of high-value data, continuously optimizes storage resource utilization, improving the overall health and availability of data assets.

[0052] It is necessary to understand that, such as Figure 2 The diagram shows the flowchart of the adaptive adjustment of query latency threshold in the English cross-scenario learning data integration and teaching collaboration management system based on big data provided in this application embodiment. The specific process is as follows: First, it starts with "monitoring the number of concurrent access connections". The core is to determine the range of the number of connections. If it is within the critical range of the number of connections, no adjustment is performed and the current latency threshold is maintained. If it is below the critical lower limit, it is determined to be a low load state. After calculating the connection number offset, the gain factor is obtained by querying the mapping relationship with the effective and accurate offset. The latency threshold is adjusted to relax the restriction and prioritize scheduling remote replicas with higher data consistency to improve data completeness. If it is above the critical upper limit, it is determined to be a high load state. After calculating the connection number correction, the attenuation coefficient is obtained by querying the mapping relationship with the effective and accurate offset. The latency threshold is tightened to reduce the congestion risk of cross-network requests, and finally, dynamic adjustment under different load states is completed.

[0053] It should be further explained that the specific steps for adaptive adjustment of the query latency threshold are as follows:

[0054] If the number of concurrent access connections falls below the critical lower limit of the connection count, the system is considered to be in a low-load state. The connection count offset and the effective accurate offset are input into a predefined connection count-latency threshold mapping relationship to obtain a latency threshold gain factor. This latency threshold gain factor is then interacted with the cross-scenario data availability zone query latency threshold to obtain the target cross-scenario data availability zone query latency threshold. This allows for dynamic relaxation of latency restrictions on cross-availability zone queries while ensuring data query accuracy. This prioritizes scheduling queries to remote replicas with higher data consistency but slightly higher network latency, improving data completeness during low-load periods. The connection count offset represents the negative difference between the number of concurrent access connections and the critical lower limit of the connection count. By dynamically relaxing latency restrictions on cross-availability zone queries while ensuring data query accuracy, query requests are guided to prioritize scheduling to remote data replicas with higher data consistency but slightly higher network latency. This fully utilizes surplus system resources during low-load periods, significantly improving query result completeness and global data consistency at the cost of an acceptable small increase in latency, thus optimizing the data service quality during low-load periods.

[0055] If the number of concurrent access connections is within the critical range, no adaptive adjustment of the query latency threshold will be performed, and the current cross-scenario data availability zone query latency threshold will be maintained. The critical range refers to the closed interval formed by the critical lower limit and critical upper limit of the connection count. This effectively avoids unnecessary threshold adjustments that could cause oscillations in the query routing strategy within the normal load fluctuation range, ensuring the stability and predictability of the system strategy and query service.

[0056] If the number of concurrent access connections exceeds the critical upper limit, the system is considered to be under high load. The connection count correction and the effective accurate offset are input into a predefined connection count-latency threshold mapping relationship to obtain a latency threshold attenuation coefficient. This coefficient is then interacted with the cross-scenario data availability zone query latency threshold to obtain the target cross-scenario data availability zone query latency threshold. This proactively tightens the latency tolerance for cross-availability zone queries, effectively reducing the queuing and congestion risks of cross-network requests. The connection count correction represents the positive difference between the number of concurrent access connections and the critical upper limit. This proactive and rapid tightening of the latency tolerance for cross-availability zone queries forces new query requests to be preferentially routed to the local or near-end data replica with the lowest latency, effectively reducing long-path cross-network queries under high concurrency and lowering the risk of network congestion and request queuing. Its core objective is to sacrifice some global data consistency (by reading potentially outdated near-end replicas) to prioritize the availability and overall response speed of core query services, achieving system self-protection and service degradation under high pressure, and ensuring uninterrupted critical business operations.

[0057] In this embodiment, by introducing connection number offset / correction and data quality offset as dual decision factors, and based on the fundamental control objective of load state switching, the system achieves a leap from "single load response" to "multi-objective contextualized intelligent decision-making." Ultimately, it dynamically balances the relationship between "data completeness," "service availability," and "system stability," enabling the data query service to intelligently adapt to different business pressure scenarios and optimize service quality and resource utilization.

[0058] The dynamic adaptation response quantification module is used to obtain adaptation response parameters in the data lifecycle management process. Based on the adaptation response parameters, the dynamic adaptation response rate of teaching collaboration is obtained, which is used to quantify the adaptation and response effectiveness of the entire teaching collaboration process in cross-scenario English.

[0059] It should be noted that the specific steps to obtain the dynamic adaptation response rate of teaching collaboration are as follows:

[0060] The adaptation response parameters include the effective accuracy rate of the integrated reference data, the learning profile update lag time, and the time required for multi-terminal command synchronization. Specifically, the effective accuracy rate of the integrated reference data refers to the rate of accuracy of the newly acquired English cross-scenario data integration if the effective accuracy scheduling was executed; otherwise, the current effective accuracy rate is used. The learning profile update lag time refers to the total time taken for students to generate new effective learning behavior data (such as classroom answer errors, after-class oral practice, offline training grammar correction, etc.) in English cross-scenario learning, after transmission, processing, and parsing, and finally synchronized to the student's learning profile to complete the profile dimension update (such as weakness tag adjustment, cognitive style correction). The learning profile management system will retrieve the data triggered by this data and successfully complete and store the updated profile dimensions (such as "tense error" tag reinforcement, "visual learning style" weight adjustment, etc.). The difference between the time of the data collection platform and the time of the occurrence of the student's effective learning behavior is recorded as the learning profile update lag time. The multi-terminal instruction synchronization time refers to the total time taken for a certain terminal (teacher / student / management terminal) to receive, parse, and display the synchronization success after cross-platform transmission and protocol adaptation, and finally after the instruction is received, parsed, and displayed on all other related terminals (e.g., after the teacher initiates the instruction, it is synchronized to the student terminal and management terminal). The time when the instruction is extracted from the collaboration platform logs of each related terminal and the time when the instruction is displayed and subsequent operations can be performed is taken as the difference between the latest time among all related terminals and the time when the instruction is extracted from the initiating terminal (e.g., the teacher's collaboration platform logs) and the time when the instruction is clicked to publish / submit is extracted.

[0061] The effective accuracy weight coefficient is integrated with the reference data, and the results of the effective accuracy and effective accuracy threshold ratio analysis are interactively processed to obtain the effective accuracy impact value. The lag time weight coefficient and lag time threshold are interactively processed with the results of the learning profile update lag time ratio analysis to obtain the lag time impact value. The achievement time weight coefficient and achievement time threshold are interactively processed with the results of the multi-terminal instruction synchronization achievement time ratio analysis to obtain the achievement time impact value. The effective accuracy impact value, lag time impact value, and achievement time impact value are coupled to obtain the teaching collaboration dynamic adaptation response rate. The specific constraint expression for the teaching collaboration dynamic adaptation response rate is as follows:

[0062] ;

[0063] In the formula, D represents the dynamic adaptation response rate of teaching collaboration; v1 represents the effective accuracy weight coefficient obtained from the teaching collaboration management database; v2 represents the lag time weight coefficient obtained from the teaching collaboration management database; v3 represents the achievement time weight coefficient obtained from the teaching collaboration management database; C0 represents the effective accuracy threshold obtained from the teaching collaboration management database; H0 represents the lag time threshold obtained from the teaching collaboration management database; P0 represents the achievement time threshold obtained from the teaching collaboration management database; C represents the effective accuracy rate of the integrated reference data; H represents the learning profile update lag time; and P represents the multi-terminal instruction synchronization achievement time.

[0064] It needs to be explained that the higher the effective accuracy of the integrated reference data, the less likely the learning data collected across scenarios will be to be lost / duplicated after ETL processing, and the less likely it will be to have semantic bias after NLP parsing. The higher the proportion of high-quality data and the faster it can be transmitted directly to the learning profile management system, the shorter the learning profile update lag time will be. The higher the effective accuracy of the integrated reference data, the stronger the consistency of the integrated cross-scenario data (such as the learning performance and weakness data obtained by teachers, students and management are completely consistent). Teaching instructions initiated based on this data (such as teachers issuing personalized homework based on precise weakness points) do not require additional data verification and correction. The instruction content can be directly transmitted through protocol conversion, reducing the time spent on "data conflicts" during the synchronization process, and the shorter the time to achieve multi-terminal instruction synchronization. The longer the learning profile update lag time, the more likely it is that transmission channel congestion will cause instructions to queue and wait, and low protocol adaptation efficiency will increase the time spent on instruction conversion. At the same time, the learning profile update lag may trigger a second verification of "instructions based on outdated data", further prolonging the time to achieve multi-terminal instruction synchronization. Meanwhile, there is a positive correlation between the effective accuracy rate of the integrated reference data and the dynamic adaptation response rate of teaching collaboration. The higher the effective accuracy rate of the integrated reference data, the better the cross-scenario learning data is processed by ETL without loss / duplication and parsed by NLP without semantic bias. High-quality data can fully support the construction of learning profiles, resulting in a higher dynamic adaptation response rate of teaching collaboration. There is a negative correlation between the learning profile update lag time and the dynamic adaptation response rate of teaching collaboration. The longer the learning profile update lag time, the more likely teachers are to formulate teaching strategies based on outdated data (such as assigning exercises for knowledge points that have already been mastered), resulting in the loss of effectiveness of teaching feedback, a significant reduction in the number of effective responses, and a lower dynamic adaptation response rate of teaching collaboration. There is also a negative correlation between the synchronization time of multi-terminal instructions and the dynamic adaptation response rate of teaching collaboration. The longer the synchronization time of multi-terminal instructions, the more likely the collaborative process will break down, resulting in a lower dynamic adaptation response rate of teaching collaboration.

[0065] The dynamic adaptation response scheduling module is used to determine whether to execute dynamic adaptation response scheduling based on the dynamic adaptation response rate of teaching collaboration, thereby ensuring real-time synchronization between the teaching process and the learner's cognitive state and improving the overall efficiency of collaborative learning. If yes, the next preset period of English cross-scenario learning data integration will be carried out after scheduling; otherwise, the next preset period of English cross-scenario learning data integration will be carried out directly.

[0066] Furthermore, the specific steps for determining whether to execute dynamic adaptive response scheduling are as follows:

[0067] If the teaching collaboration dynamic adaptation response rate is higher than or equal to the adaptation response reference value, dynamic adaptation response scheduling will not be executed; otherwise, dynamic adaptation response scheduling will be executed. Dynamic adaptation response scheduling includes adaptive adjustment of response time window width and adaptive adjustment of processing window threshold.

[0068] It is necessary to understand that, such as Figure 3The diagram shows the adaptive adjustment flowchart of the response time window width of the English cross-scenario learning data integration and teaching collaboration management system based on big data provided in this application embodiment. The specific process is as follows: starting from monitoring the teaching collaboration adaptation response behavior and counting the cumulative value of the effective response count, the core is to determine whether the cumulative value has reached the reference value of the effective response count. If it has, no adjustment is made and the current time window width is maintained; if it has not reached, the effective count offset (the negative deviation between the reference value and the actual cumulative value) and the adaptation response offset (the negative deviation between the adaptation response reference value and the actual adaptation response rate) are calculated. These two offsets are input into a preset mapping table to obtain the time window width gain coefficient. The target time window width is obtained by adjusting the coefficient. Finally, the time window is dynamically widened to reduce the collaboration threshold and promote deeper teaching collaboration, thus completing the entire adjustment process.

[0069] It should be further explained that the specific steps for adaptive adjustment of the response time window width are as follows:

[0070] Monitor adaptive response behavior in collaborative teaching tasks, identify and count the number of effective responses, and adaptively adjust the intervention response time window width based on the real-time cumulative value of the number of effective responses. Specifically:

[0071] If the number of effective responses in the collaborative learning approach is higher than or equal to the reference value for the number of effective responses, then no adaptive adjustment of the response time window width will be performed, and the current intervention response time window width will be maintained. This avoids unnecessary rhythmic disruptions to learners who are already in a good interactive state, ensuring their smooth and continuous learning experience and coherence of thought, reflecting the precise control principle of "no intervention without problems".

[0072] If the number of effective responses to collaborative teaching is less than the reference value, the effective response offset and the adaptation response offset are input into a predefined effective response-time window width mapping table to obtain the time window width gain coefficient. This time window width gain coefficient is then interacted with the intervention response time window width to obtain the target intervention response time window width. This reduces time constraints and promotes deeper collaborative teaching interaction. The effective response offset represents the negative difference between the number of effective responses to collaborative teaching and the reference value, while the adaptation response offset represents the negative difference between the dynamic adaptation response rate and the adaptation response reference value. When insufficient effective responses are detected, the system intelligently and differentially relaxes the response waiting time limit. By extending the intervention response time window, the system effectively reduces the anxiety and frustration caused by time pressure, providing learners with more time for thinking, organizing, and responding. This is particularly helpful in promoting deeper cognitive processing and encouraging hesitant or slower learners to participate more deeply in collaborative teaching interaction, thereby improving the overall quality and inclusivity of the interaction.

[0073] In this embodiment, by introducing effective count offset and adaptive response offset, the system can identify different degrees and natures of "insufficient participation" (such as simply too few counts, or too few counts with poor quality) and apply differentiated, non-linear support. Ultimately, while ensuring the progress of the teaching process, it provides each learner with time flexibility matching their current cognitive readiness, elevating teaching collaboration from "rhythm synchronization" to "cognitive state synchronization," significantly improving the effectiveness of interaction and the humanization of teaching.

[0074] It should be further explained that the specific steps for adaptive adjustment of the window threshold are as follows:

[0075] The system monitors and records the feedback response times of multiple individual learners during the teaching interaction process, and adaptively adjusts the threshold of the real-time interactive data processing window based on the feedback response times. Specifically:

[0076] If the teaching feedback response time is within the critical range, adaptive adjustment of the processing window threshold will not be performed. The critical range refers to the closed interval formed by the lower and upper critical limits of the response time. This maintains the stability of teaching interaction and the efficiency of data processing, ensuring that the accuracy and response speed of real-time feedback are in an optimal balance.

[0077] The adaptive adjustment of the window threshold also includes:

[0078] If the teaching feedback response time is less than the critical lower limit of the response time, the response time offset and the adapted response offset are input into a predefined response time-window threshold mapping table for lookup. This retrieves the window threshold narrowing coefficient, which is then interacted with the current real-time interactive data processing window threshold to obtain the target real-time interactive data processing window threshold. This intelligently and controllably narrows the data processing window, thereby increasing processing frequency and real-time feedback speed. The response time offset represents the negative difference between the teaching feedback response time and the critical lower limit of the response time. On one hand, this significantly increases the data processing frequency, matching the data processing rhythm with the actual state of rapid feedback, compressing the processing cycle to a range compatible with feedback delays, and improving real-time feedback speed. This ensures that learners receive immediate responses to their rapid feedback, enhancing their interactive engagement. On the other hand, it reduces the inclusion of invalid data, lowers the computational resource consumption of data processing, improves data processing efficiency, and avoids the feedback delay caused by wide windows. This allows instructors to capture learners' rapid feedback in real time, adjust the teaching pace promptly, and achieve a virtuous cycle of "immediate feedback - rapid adjustment."

[0079] If the teaching feedback response time exceeds the critical upper limit, the response time correction and the adaptive response offset are input into a predefined response time-window threshold mapping table for querying, obtaining the window threshold gain coefficient. This window threshold gain coefficient is then interacted with the current real-time interactive data processing window threshold to obtain the target real-time interactive data processing window threshold. The data processing window is intelligently widened, transforming the originally inefficient waiting period into a window for generating in-depth teaching insights. The response time correction represents the positive difference between the teaching feedback response time and the critical upper limit. Firstly, this avoids the loss of effective feedback data, ensuring that learners' delayed feedback during the thinking period is fully captured, guaranteeing the integrity of interactive data. More importantly, it transforms the originally inefficient waiting period caused by response delays into a window for generating in-depth teaching insights. Within the widened window, the system can simultaneously analyze learners' feedback hesitation duration, multiple feedback modification behavior trajectories, and the distribution characteristics of group feedback delays, generating teaching insight reports such as learner comprehension difficulty analysis and group interaction heat assessment. This provides data support for instructors to adjust teaching depth, slow down the pace of explanation, or supplement Q&A content, achieving a technological upgrade from passively waiting for feedback to actively exploring teaching value.

[0080] In this embodiment, by accurately monitoring the teaching feedback response time of multiple individual learners and dynamically adapting it to the threshold of the real-time interactive data processing window, the system maintains the stability of teaching interaction and the efficiency of data processing when the feedback response time is within the critical range, ensuring that the accuracy and response speed of real-time feedback are in an optimal balance. When the feedback response time is less than the critical lower limit, the data processing window is intelligently and controllably narrowed, significantly improving the data processing frequency and real-time feedback speed, reducing the inclusion of invalid data and the consumption of computing resources, and helping educators to capture rapid feedback and adjust the teaching pace in a timely manner, achieving a virtuous cycle of "instant feedback - rapid adjustment". When the feedback response time is greater than the critical upper limit, the data processing window is intelligently widened, which not only avoids the loss of effective feedback data during the learner's thinking period and ensures the integrity of interactive data, but also transforms the originally inefficient waiting period into a window for generating in-depth teaching insights. The system can simultaneously analyze in-depth data such as feedback hesitation duration and group delay distribution, generating reports such as analysis of comprehension difficulties and evaluation of interaction popularity, providing data support for educators to optimize the depth and pace of teaching, and ultimately realizing the shift from "passively waiting for feedback" to "actively exploring teaching value". The technological upgrade comprehensively considers the real-time nature, accuracy, and depth of data utilization in teaching feedback.

[0081] The various features and processes described above can be used independently of each other or can be combined in various ways. All possible combinations and sub-combinations are intended to fall within the scope of this disclosure. Furthermore, certain method or process blocks may be omitted in some embodiments. The methods and processes described herein are not limited to any particular order, and the blocks or states associated with them may be performed in other suitable orders. For example, the described blocks or states may be performed in an order different from the order specifically disclosed, or multiple blocks or states may be combined in a single block or state. Example blocks or states may be performed serially, in parallel, or in some other manner. Blocks or states may be added to or removed from the disclosed example embodiments. The exemplary systems and components described herein may be configured differently from those described. For example, elements may be added to, removed from, or rearranged compared to the disclosed example embodiments.

[0082] The various operations of the example methods described herein can be performed at least in part by an algorithm. This algorithm can be contained in program code or instructions stored in memory (e.g., the aforementioned non-transitory computer-readable storage medium). Such an algorithm may include a machine learning algorithm. In some embodiments, the machine learning algorithm may not be explicitly programmed into the computer to perform the function, but can learn from training data to create a predictive model that performs the function.

[0083] The various operations of the example methods described herein can be performed, at least in part, by one or more processors that are temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors can constitute the engine of a processor implementation that operates to perform one or more of the operations or functions described herein.

[0084] Similarly, the methods described herein can be implemented at least in part by a processor, where one or more specific processors are examples of hardware. For example, at least some operations of a method can be performed by one or more processors or an engine implemented by a processor. Furthermore, one or more processors can also be operated to support the performance of related operations in a “cloud computing” environment or as “Software as a Service” (SaaS). For example, at least some operations can be performed by a set of computers (as an example of a machine including processors), where these operations are accessible via a network (e.g., the Internet) and via one or more suitable interfaces (e.g., application programming interfaces (APIs)).

[0085] The performance of certain operations can be distributed across processors, residing not only within a single machine but also deployed across multiple machines. In some example embodiments, the processor or processor-implemented engine may reside in a single geographic location (e.g., within a home environment, office environment, or server cluster). In other example embodiments, the processor or processor-implemented engine may be distributed across multiple geographic locations.

[0086] In this specification, multiple instances may implement components, operations, or structures described as single instances. Although individual operations of one or more methods are shown and described as separate operations, one or more of the separate operations may be performed simultaneously and do not need to be performed in the order shown. Structures and functions presented as separate components in the example configuration may be implemented as composite structures or components. Similarly, structures and functions presented as single components may be implemented as separate components. These and other variations, modifications, additions, and improvements fall within the scope of this document.

[0087] While an overview of the subject matter has been described with reference to specific example embodiments, various modifications and changes can be made to these embodiments without departing from the broader scope of embodiments of this disclosure. Such embodiments of the subject matter are referred to herein, individually or collectively, by the term "invention," and are used for convenience only and are not intended to limit the scope of this application to any single disclosure or concept, should more than one disclosure or concept be disclosed in fact.

[0088] The embodiments described herein have been described in sufficient detail to enable those skilled in the art to practice the disclosed teachings. Other embodiments may be used and derived therefrom, such that structural and logical substitutions and changes may be made without departing from the scope of this disclosure. Therefore, the detailed description should not be construed as limiting, and the scope of the various embodiments is defined only by the appended claims and the full scope of their equivalents.

Claims

1. A big data-based English cross-scenario learning data integration and teaching collaboration management system, characterized in that: It includes a data perception and aggregation module, an integrated effective and accurate scheduling module, a dynamic adaptation response quantization module, and a dynamic adaptation response scheduling module. The data perception and aggregation module is used to collect heterogeneous learning behavior data streams from multiple independent English learning application scenarios, and to perform a data integration process. It obtains effective integration parameters in the data integration process and obtains the effective accuracy rate of cross-scenario English data integration based on the effective integration parameters. This is used to quantify the effectiveness and accuracy of the learning data collected in cross-scenario English learning in the integration process. The integrated effective accuracy scheduling module is used to determine whether to perform integrated effective accuracy scheduling based on the integrated effective accuracy rate of English cross-scenario data. This is to maintain an efficient processing flow and improve the reliability of cross-scenario learning when the data integration quality meets the standard. If yes, the data lifecycle management stage is performed after scheduling; otherwise, the data lifecycle management stage is performed directly. The dynamic adaptation response quantification module is used to obtain adaptation response parameters in the data lifecycle management process, and obtain the dynamic adaptation response rate of teaching collaboration based on the adaptation response parameters. This is used to quantify the adaptation and response effectiveness of the entire teaching collaboration process in English cross-scenario scenarios. The dynamic adaptation response scheduling module is used to determine whether to execute dynamic adaptation response scheduling based on the teaching collaboration dynamic adaptation response rate, thereby ensuring real-time synchronization between the teaching process and the learner's cognitive state and improving the overall efficiency of collaborative learning. If yes, the next preset period of English cross-scenario learning data integration will be carried out after scheduling; otherwise, the next preset period of English cross-scenario learning data integration will be carried out directly.

2. The English cross-scenario learning data integration and teaching collaboration management system based on big data as described in claim 1, characterized in that, The specific steps for obtaining the effective accuracy of cross-scenario data integration in English are as follows: The effective parameters for integration include ETL dynamic scheduling response time, learning data integration rate, and number of data integration connection timeout retries. The response time weighting coefficient and response time threshold are interactively processed with the results of the ETL dynamic scheduling response time ratio analysis to obtain the response time impact value. The integration rate weight coefficient is interactively processed with the results of the analysis of the proportion of integration rate and integration rate threshold of learning data to obtain the integration rate influence value. The retry number weighting coefficient and retry number threshold are combined with the data and the results of the timeout retry number ratio analysis are interactively processed to obtain the retry number impact value. By coupling the impact values ​​of response time, integration rate, and retry count, the effective accuracy of cross-scenario data integration in English is obtained. The specific steps for determining whether to perform integrated effective accuracy scheduling are as follows: if the effective accuracy rate of cross-scenario data integration in English is higher than the effective accuracy reference value, then integrated effective accuracy scheduling is not performed; otherwise, integrated effective accuracy scheduling is performed. The integrated effective accuracy scheduling includes adaptive adjustment of retention period and adaptive adjustment of query latency threshold.

3. The English cross-scenario learning data integration and teaching collaboration management system based on big data as described in claim 2, characterized in that, The specific process for adaptive adjustment of the retention period is as follows: Collect English learning behavior data streams from multiple independent application scenarios; The total data volume of English learning behavior data streams from the multiple independent application scenarios is aggregated and calculated in real time. Based on the preset total data volume threshold range and the mapping relationship between the total data volume and the retention period, the overall retention period of learning data for all scenarios is dynamically adjusted. Historical data exceeding the dynamically adjusted retention period will be automatically archived and deleted. A non-increasing function is predefined, and the non-increasing function is: Where T represents the learning data retention period, and G represents the total amount of English cross-scenario data collected within a preset time window, the function f(G) is implemented through a set of discrete threshold intervals and corresponding retention period values.

4. The English cross-scenario learning data integration and teaching collaboration management system based on big data as described in claim 3, characterized in that, The adaptive adjustment of the retention period also includes: The total amount of cross-scenario English data collected at the current frequency is calculated. The total amount of cross-scenario English data collected is compared with the preset threshold interval sequence to determine the specific interval to which it belongs. If the learning data retention period corresponding to the specific interval to which the total amount of cross-scenario English data collected belongs is different from the currently effective learning data retention period, the adjustment mechanism is triggered. The adjustment mechanism includes: periodic updates and recalculation of the lifecycle of existing data. The specific steps of the periodic update are as follows: The total amount of English cross-scenario data collected and the effective accuracy offset are input into the total data amount-retention period mapping relationship to obtain the retention period correction factor. The current learning data retention period and the retention period correction factor are weighted and fused to obtain the target learning data retention period. The effective accuracy offset represents the degree of negative deviation between the effective accuracy rate of English cross-scenario data integration and the effective accuracy reference value.

5. The English cross-scenario learning data integration and teaching collaboration management system based on big data as described in claim 4, characterized in that, The specific steps for the lifecycle recalculation of existing data are as follows: Iterate through all stored historical data entries, extract the lifecycle anchor timestamp for each historical data entry, couple the target learning data retention period with the lifecycle anchor timestamp to obtain the target absolute failure timestamp, and synchronously update the time status flag of the data entry based on the comparison between the target absolute failure timestamp and the current failure timestamp. If the current expiration timestamp is greater than or equal to the target absolute expiration timestamp, then update the expiration status flag of the data entry to pending cleanup; If the current expiration timestamp is less than the target absolute expiration timestamp, then the expiration status flag of the data entry will be updated to valid.

6. The English cross-scenario learning data integration and teaching collaboration management system based on big data as described in claim 2, characterized in that, The specific steps for adaptive adjustment of the query latency threshold are as follows: If the number of concurrent access connections is lower than the critical lower limit of the number of connections, it is determined to be in a low load state. The connection number offset and the effective accurate offset are input into the predefined connection number-latency threshold mapping relationship to obtain the latency threshold gain factor. The latency threshold gain factor is interacted with the cross-scenario data availability zone query latency threshold to obtain the target cross-scenario data availability zone query latency threshold. This realizes the dynamic relaxation of the latency limit for cross-availability zone queries while ensuring the accuracy of data queries, thereby prioritizing scheduling to remote replicas with higher data consistency but slightly higher network latency, improving data integrity during low load periods. The connection number offset represents the degree of negative deviation between the number of concurrent access connections and the critical lower limit of the number of connections. If the number of concurrent access connections is within the critical interval of the number of connections, no adaptive adjustment of the query latency threshold will be performed, and the current cross-scenario data availability zone query latency threshold will be maintained. The critical interval of the number of connections refers to the closed interval formed by the critical lower limit of the number of connections and the critical upper limit of the number of connections. If the number of concurrent access connections exceeds the critical upper limit of the number of connections, it is determined to be in a high-load state. The connection number correction amount and the effective accurate offset are input into the predefined connection number-latency threshold mapping relationship to obtain the latency threshold attenuation coefficient. The latency threshold attenuation coefficient is then interactively processed with the cross-scenario data availability zone query latency threshold to obtain the target cross-scenario data availability zone query latency threshold. The latency tolerance of cross-availability zone queries is actively tightened, thereby effectively reducing the queuing and congestion risks of cross-network requests. The connection number correction amount represents the degree of positive deviation between the number of concurrent access connections and the critical upper limit of the number of connections.

7. The English cross-scenario learning data integration and teaching collaboration management system based on big data as described in claim 1, characterized in that, The specific steps for obtaining the dynamic adaptation response rate of teaching collaboration are as follows: The adaptation response parameters include the effective accuracy of the integrated reference data, the learning profile update lag time, and the time to achieve multi-terminal command synchronization. The effective accuracy weighting coefficient is integrated with the reference data, and the results of the analysis of the proportion of effective accuracy and effective accuracy threshold are interactively processed to obtain the effective accuracy impact value. The results of the analysis of the lag time weight coefficient, the lag time threshold, and the proportion of lag time in the learning profile update are interactively processed to obtain the lag time impact value. The results of the analysis of the achievement time weighting coefficient, achievement time threshold and the proportion of achievement time of multi-terminal instruction synchronization are interactively processed to obtain the achievement time impact value. The impact values ​​of effective accuracy, lag time, and achievement time are coupled and processed to obtain the dynamic adaptation response rate of teaching collaboration. The specific steps for determining whether to perform dynamic adaptive response scheduling are as follows: If the teaching collaboration dynamic adaptation response rate is higher than or equal to the adaptation response reference value, then dynamic adaptation response scheduling will not be executed; otherwise, dynamic adaptation response scheduling will be executed. The dynamic adaptation response scheduling includes adaptive adjustment of the response time window width and adaptive adjustment of the processing window threshold.

8. The English cross-scenario learning data integration and teaching collaboration management system based on big data as described in claim 7, characterized in that, The specific steps for adaptive adjustment of the response time window width are as follows: Monitor adaptive response behavior in collaborative teaching tasks, identify and count the number of effective responses, and adaptively adjust the width of the intervention response time window based on the real-time cumulative value of the number of effective responses. Specifically: If the number of effective responses to the teaching collaboration adaptation is higher than or equal to the reference value for the number of effective responses, then no adaptive adjustment of the response time window width will be performed, and the current intervention response time window width will be maintained. If the number of effective responses to collaborative teaching is less than the reference value for the number of effective responses, the offset of the number of effective responses and the offset of the effective responses are input into a predefined effective responses-time window width mapping table to obtain the time window width gain coefficient. The time window width gain coefficient is then interacted with the intervention response time window width to obtain the target intervention response time window width, thereby reducing the time factor and promoting deeper collaborative teaching interaction. The offset of the number of effective responses represents the degree of negative deviation between the number of effective responses to collaborative teaching and the reference value for the number of effective responses, and the offset of the effective responses represents the degree of negative deviation between the dynamic adaptation response rate of collaborative teaching and the reference value for the adaptation response.

9. The English cross-scenario learning data integration and teaching collaboration management system based on big data as described in claim 7, characterized in that, The specific steps for adaptive adjustment of the processing window threshold are as follows: The system monitors and records the feedback response times of multiple individual learners during the teaching interaction process, and adaptively adjusts the threshold of the real-time interactive data processing window based on these feedback response times. Specifically: If the teaching feedback response time is within the critical interval of the response time, then the adaptive control of the processing window threshold will not be performed. The critical interval of the response time refers to the closed interval formed by the critical lower limit of the response time and the critical upper limit of the response time.

10. The English cross-scenario learning data integration and teaching collaboration management system based on big data as described in claim 1, characterized in that, The adaptive adjustment of the processing window threshold also includes: If the teaching feedback response time is less than the critical lower limit of the response time, the response time offset and the adapted response offset are input into the predefined response time-window threshold mapping table for querying to obtain the window threshold narrowing coefficient. The window threshold narrowing coefficient is then interacted with the current real-time interactive data processing window threshold to obtain the target real-time interactive data processing window threshold. The data processing window is narrowed intelligently and in a controlled manner, thereby improving the processing frequency and real-time feedback speed. The response time offset represents the degree of negative deviation between the teaching feedback response time and the critical lower limit of the response time. If the teaching feedback response time exceeds the critical upper limit, the response time correction and the adaptive response offset are input into the predefined response time-window threshold mapping table for querying to obtain the window threshold gain coefficient. The window threshold gain coefficient is then interacted with the current real-time interactive data processing window threshold to obtain the target real-time interactive data processing window threshold. The data processing window is intelligently widened, thereby transforming the originally inefficient waiting period into a window for generating deep teaching insights. The response time correction represents the degree of positive deviation when the teaching feedback response time exceeds the critical upper limit.

Images