An education digital intelligence teaching data analysis method and system based on big data

Abstract

The present application relates to the technical field of data analysis, in particular to an education digital teaching data analysis method and system based on big data; the method comprises the following steps: obtaining each teaching image data uploaded by a teaching institution and basic data information associated with each teaching image data; determining the processing priority of each teaching image data according to each teaching image data and each basic data information; performing resource scheduling and distribution on each teaching image data by using the processing priority of each teaching image data to obtain each resource analysis path; and performing data analysis and processing on each teaching image data by using each resource analysis path to obtain a teaching data analysis result. The present application aims to solve the problem of low processing efficiency of teaching image data and inaccurate data analysis result caused by insufficient resource scheduling under the regional education group school mode.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, specifically to a data analysis method and system for digital teaching in education based on big data. Background Technology

[0002] As the scale of education groups continues to expand, the number of students and teaching activities are increasing dramatically, leading to an exponential increase in the amount of test papers and homework that need to be processed daily. Transforming raw teaching materials such as students' daily homework and test papers into analyzable digital information facilitates better teaching data analysis.

[0003] However, under the regional education group-based school management model, different campuses exhibit variations in paper surface smoothness, ink absorption, and whiteness coefficients. Simultaneously, scanners possess inherent differences in optical resolution, color depth, and image processing methods. This results in complex inconsistencies in key visual features such as background uniformity, text edge sharpness, and contrast in scanned images. Poor image quality from the scanner severely impacts the accuracy of subsequent text recognition, significantly reducing the quality of structured text data extracted from the images. Consequently, the system frequently experiences resource contention and processing delays in image preprocessing, text recognition, and structured conversion of teaching image data. Insufficient system resource scheduling leads to rapid data queue accumulation, resulting in low processing efficiency and inaccurate data analysis results. This prevents teachers and administrators from obtaining the latest analysis reports in a timely manner, severely affecting the timeliness and effectiveness of teaching feedback. Summary of the Invention

[0004] The purpose of this invention is to provide a data analysis method and system for digital teaching based on big data, which can solve the problems of low processing efficiency of teaching image data and inaccurate data analysis results caused by insufficient resource allocation under the regional education group-based school management model.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a data analysis method for digital and intelligent teaching in education based on big data, comprising: Obtain each teaching image data uploaded by the teaching institution, as well as the basic data information associated with each teaching image data; Based on each teaching image data and its basic information, the processing priority of each teaching image data is determined. By utilizing the processing priority of each teaching image data, resource scheduling and allocation are performed on each teaching image data to obtain each resource analysis path; By utilizing each resource analysis path, data analysis and processing are performed on each teaching image data to obtain teaching data analysis results.

[0006] Preferably, the step of determining the processing priority of each teaching image data based on each teaching image data and each data basic information includes: Based on each teaching image data and each data basic information, determine the urgency of each data, the complexity of each data, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution; The processing priority of each teaching image data is determined based on the urgency of each data point, the complexity of each data point, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution.

[0007] Preferably, the steps of determining the urgency of each data point, the complexity of each data point, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution based on each teaching image data point and each data point's basic information include: Each teaching image data is analyzed for frequency components, local contrast, and edge detection to obtain frequency component parameters, data defect parameters, and edge artifact parameters. Based on the frequency component parameters, data defect parameters, and edge artifact parameters, an evaluation index for each teaching image data is determined. The evaluation index of each teaching image data is compared with the preset evaluation index to obtain the data quality ratio of each teaching image data. By using the data quality ratio of each teaching image data, data analysis is performed on each teaching image data and each data basic information to obtain the urgency of each data, the complexity of each data, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution.

[0008] Preferably, the step of using the data quality ratio of each teaching image data to perform data analysis on each teaching image data and each data basic information to obtain the urgency of each data, the complexity of each data, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution includes: A preliminary data analysis is performed on each teaching image data and each data basic information to obtain the initial urgency of each data, the initial complexity of each data, the initial demand coefficient of each teaching institution, and the initial processing load coefficient of each teaching institution. Using the data quality ratio of each teaching image data, the initial urgency, initial complexity, initial demand coefficient of each teaching institution, and initial processing load coefficient of each teaching institution are adjusted to obtain the urgency, complexity, demand coefficient, and processing load coefficient of each teaching institution.

[0009] Preferably, the step of allocating resources for each teaching image data based on the processing priority of each teaching image data to obtain each resource analysis path includes: The queuing sequence of each teaching image data is determined according to the processing priority of each teaching image data. Using the queuing sequence of each teaching image data, an initial resource scheduling allocation is performed on each teaching image data to obtain an initial analysis path for each resource. For each resource's initial analysis path, perform path planning verification to obtain each resource analysis path that passes the verification.

[0010] Preferably, the step of using the queuing sequence of each teaching image data to perform initial resource scheduling allocation for each teaching image data to obtain an initial analysis path for each resource includes: The sequence of each teaching image data is validated to obtain the valid sequence of each teaching image data. Based on the queuing sequence of each teaching image data that has passed the verification, a resource scheduling library is determined; Using the resource scheduling library and the queuing sequence of each qualified teaching image data, the initial resource scheduling allocation is performed on each teaching image data to obtain the initial analysis path for each resource.

[0011] Preferably, the steps for performing data analysis and processing on each teaching image data using each resource analysis path to obtain teaching data analysis results include: Using each resource analysis path, data analysis and processing are performed on each teaching image data to obtain the initial data analysis results and analysis quality feedback coefficient; The initial results of the data analysis are adjusted using the aforementioned analysis quality feedback coefficient to obtain the teaching data analysis results.

[0012] Preferably, the steps for adjusting the initial data analysis results using the analysis quality feedback coefficient to obtain the teaching data analysis results include: Based on the analysis quality feedback coefficient and the preset feedback coefficient, confirm the review and adjustment plan; Using the aforementioned review and adjustment scheme, the initial results of the data analysis are adjusted to obtain the teaching data analysis results.

[0013] Preferably, the step of obtaining each teaching image data uploaded by the teaching institution and the basic information of the data associated with each teaching image data includes: Obtain the original data of each teaching image uploaded by the teaching institution, as well as the original basic information of the data associated with each teaching image; Each teaching image's original data and its original basic information are preprocessed to obtain preprocessed original data and its original basic information for each teaching image. Each preprocessed original teaching image data and its basic information are verified to obtain each verified teaching image data and its associated basic information.

[0014] This invention also provides a big data-based intelligent education teaching data analysis system, which includes: The data acquisition module is used to acquire each teaching image data uploaded by the teaching institution and the basic data information associated with each teaching image data; The priority determination module is used to determine the processing priority of each teaching image data based on each teaching image data and each data basic information; The scheduling and allocation module is used to allocate resources for each teaching image data according to the processing priority of each teaching image data, so as to obtain each resource analysis path; The data analysis module is used to perform data analysis and processing on each teaching image data using each resource analysis path, and obtain teaching data analysis results.

[0015] Compared with existing technologies, the big data-based digital teaching data analysis method and system of the present invention have the following advantages: This invention acquires teaching image data and related basic information uploaded by educational institutions and determines the processing priority of each teaching image data. Subsequently, resource scheduling and allocation are performed on the data based on the processing priority to generate resource analysis paths. These paths are then used to analyze and process the teaching image data, yielding teaching data analysis results. By performing refined analysis on each teaching image data and determining its processing priority, differentiated processing based on the actual quality and characteristics of the data can be achieved, avoiding the limitations of traditional fixed-parameter preprocessing. Secondly, facing the bottlenecks in computational resource allocation and concurrent processing caused by massive and diverse test paper image data, the introduction of a processing priority and resource scheduling and allocation mechanism optimizes the data processing queue and parallel processing capabilities, effectively alleviating system pressure during peak examination periods and reducing processing latency. Furthermore, this invention can more accurately assess data processing needs, ensuring timely processing of key data and improving the timeliness and effectiveness of teaching feedback. Ultimately, it significantly improves the efficiency and accuracy of digital teaching data analysis in education. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention, the accompanying drawings used in the specific embodiments will be briefly described below. In all the drawings, the elements or parts are not necessarily drawn to scale.

[0017] Figure 1 This is a flowchart of a data analysis method for digital teaching in education based on big data, according to the present invention.

[0018] Figure 2 This is a structural block diagram of an educational digital teaching data analysis system based on big data, according to the present invention.

[0019] In the diagram: 210, Data Acquisition Module; 220, Priority Determination Module; 230, Scheduling and Allocation Module; 240, Data Analysis Module.

[0020] The implementation and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] The following drawings disclose several embodiments of the present invention. For clarity, many practical details will be described in the following description. However, it should be understood that these practical details are not intended to limit the invention. That is, in some embodiments of the invention, these practical details are not essential. Furthermore, for the sake of simplicity, some conventional structures and components will be shown in the drawings in a simple schematic manner.

[0022] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0023] Furthermore, in this invention, the use of terms such as "first" and "second" is for descriptive purposes only and does not specifically refer to any order or sequence, nor is it intended to limit the invention. They are merely used to distinguish components or operations described using the same technical terms, and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of various embodiments can be combined with each other, but only if they are feasible for those skilled in the art. If a combination of technical solutions is contradictory or impossible to implement, such a combination should be considered nonexistent and not within the scope of protection claimed by this invention.

[0024] To further understand the content, features, and effects of this invention, the following embodiments are provided, and detailed descriptions are given below in conjunction with the accompanying drawings: Please see Figure 1 This invention provides a data analysis method for digital and intelligent teaching in education based on big data, comprising the following steps: S100. Obtain each teaching image data uploaded by the teaching institution, along with the basic data information associated with each teaching image data. Teaching image data refers to various image information from teaching activities, such as scanned copies of student assignments and exam papers, hand-drawn drawings, and mixed text and image materials. Basic data information refers to metadata related to the teaching image data, such as the teaching institution name, student ID, course name, submission time, and exam paper type, which helps in classifying, indexing, and initially filtering the teaching image data. Specifically, teaching institutions can manually upload scanned teaching image files and their corresponding basic information into the system. Alternatively, they can use a batch import tool to upload pre-organized teaching image data files and basic data information files (such as CSV or Excel formats) to the system at once. This method improves upload efficiency and reduces the tediousness of manual operation. Furthermore, by integrating with the teaching institution's existing document management system or scanning system, automatic synchronous uploading of teaching image data and basic data information can be achieved. For example, after the teaching institution's scanner completes scanning of exam papers, the image data and preset basic information (such as class, subject, and date) can be directly transmitted through the interface.

[0025] S200. Based on each teaching image data and its basic information, determine the processing priority of each teaching image data. The processing priority is an assigned processing order or importance level based on the characteristics and needs of the teaching image data, guiding subsequent resource scheduling. Specifically, priorities are set for each teaching image data or a specific batch of teaching image data based on experience or preset rules. For example, for urgent final exam score analysis, its priority can be manually set to the highest. Alternatively, a preliminary judgment can be made based on certain fields in the basic data information. For example, based on the submission time field, data with earlier submission times are assigned lower priority, while data with later submission times or closer to the deadline are assigned higher priority. Or, based on the exam paper type field, exam paper data for important exams are given a higher priority than daily homework data.

[0026] S300. Utilizing the processing priority of each teaching image data, resource scheduling and allocation are performed on each teaching image data to obtain each resource analysis path. Resource scheduling and allocation refers to the rational allocation of computing and storage resources to different teaching image data processing tasks according to processing priority, in order to optimize overall processing efficiency. A resource analysis path refers to the planned data analysis and processing flow for each teaching image data, including the required algorithms, models, and computing nodes. Specifically, according to a preset priority queue, high-priority teaching image data is preferentially allocated to idle computing resources for processing. For example, when multiple processing tasks are waiting, the highest priority task is selected first, and available computing resources such as CPU and memory are allocated to it. Alternatively, teaching image data can be allocated to different processing clusters or servers according to priority. For example, high-priority data can be allocated to high-performance computing clusters to ensure rapid processing; while low-priority data can be allocated to ordinary clusters to wait in the queue. Furthermore, the resource quota for each task can be dynamically adjusted according to priority. For example, high-priority tasks can obtain more CPU time slices or larger memory spaces to accelerate their processing.

[0027] S400. Utilize each resource analysis path to perform data analysis on each teaching image data, obtaining teaching data analysis results. These results are quantitative or qualitative conclusions derived from processing and analysis to guide teaching decisions, such as students' knowledge mastery, teaching effectiveness evaluation, and learning trend prediction. Specifically, based on the algorithms and models specified in the resource analysis path, image recognition, text extraction, and semantic analysis are performed on the teaching image data. For example, for test paper images, OCR technology can be used to extract text content, followed by keyword recognition and error analysis. Alternatively, based on the resource analysis path, the processed data can be imported into specific data analysis tools or platforms for deeper statistical analysis, pattern recognition, or machine learning. For example, student answer data can be imported into statistical software to analyze the overall knowledge mastery of the class, or machine learning models can be used to predict student learning risks. Furthermore, different forms of teaching data analysis results can be generated based on the resource analysis path, such as visualization charts, reports, or structured databases.

[0028] This invention lays the foundation for subsequent data processing by acquiring each teaching image data uploaded by educational institutions and the basic information associated with each teaching image data. This ensures the comprehensiveness and accuracy of the analyzed data. Secondly, based on each teaching image data and its basic information, a processing priority is determined for each teaching image data. By introducing processing priorities, this invention can intelligently adjust the processing order based on factors such as the urgency and complexity of the data, the needs of the educational institution, and the processing load, ensuring that key data is processed in a timely manner and avoiding the problem of missing the best teaching intervention opportunity due to processing delays. Furthermore, using the processing priority of each teaching image data, resource scheduling and allocation are performed for each teaching image data, resulting in each resource analysis path. This allows for the dynamic allocation of computing and storage resources according to priority and the planning of the optimal analysis path for each data. This not only improves the overall throughput but also ensures that different types of data receive processing most suitable for their characteristics. Finally, using each resource analysis path, data analysis processing is performed on each teaching image data to obtain teaching data analysis results. Through refined processing and resource scheduling in the early stages, high-quality teaching data analysis results can be obtained. High-quality analysis results can provide teachers and administrators with more accurate and timely feedback on student learning and decision support, thereby effectively improving teaching quality and student learning efficiency.

[0029] In some embodiments of this application described above, the step of determining the processing priority of each teaching image data based on each teaching image data and each data basic information includes: Based on each teaching image data and its basic information, the urgency, complexity, demand coefficient, and processing load coefficient of each teaching institution are determined. Specifically, upcoming exams or urgent teaching feedback have a higher urgency. Data complexity refers to the amount of computational resources and time required for the analysis and processing of teaching image data; for example, image data containing a large amount of detail, high resolution, or requiring complex algorithm analysis has a higher complexity. The demand coefficient of each teaching institution refers to the degree of expectation or importance that a particular teaching institution places on the data analysis results; for example, data required for key projects or critical decisions has a higher demand coefficient. The processing load coefficient of each teaching institution refers to the current usage of the institution's data processing system or resources; for example, when the load is high, it may be necessary to adjust priorities to balance resource utilization.

[0030] The processing priority of each teaching image data is determined based on the urgency of each data point, the complexity of each data point, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution.

[0031] This embodiment, by considering the urgency, complexity, demand coefficient, and processing load coefficient of each teaching institution, enables a multi-dimensional and refined evaluation of the processing priority of each teaching image data. Specifically, data urgency ensures that critical and time-sensitive tasks receive priority processing resources; data complexity helps to rationally allocate computing resources, avoiding excessive resource consumption by simple tasks while complex tasks wait for extended periods; the demand coefficient of the teaching institution allows data analysis to better respond to actual teaching needs and strategic priorities; and the processing load coefficient of the teaching institution can dynamically adjust priorities to optimize the overall system's resource utilization efficiency and throughput. This makes the determination of processing priorities more scientific and reasonable.

[0032] In some embodiments of this application described above, the steps of determining the urgency of each data point, the complexity of each data point, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution based on each teaching image data point and each data point's basic information include: Each teaching image data point is analyzed for frequency components, local contrast, and edge detection to obtain frequency component parameters, data defect parameters, and edge artifact parameters. Specifically, frequency component analysis is performed on each teaching image data point to assess the image's texture detail and sharpness. For example, Fourier transform or wavelet transform is used to obtain the energy distribution of the image at different frequencies, thus obtaining the frequency component parameters. Local contrast analysis is used to measure the brightness difference between adjacent regions in the image, reflecting the image's visual sharpness and information content. For example, data defect parameters are obtained by calculating the gray-level standard deviation or contrast enhancement factor of local image regions. Edge detection analysis is used to identify boundaries and contours in the image, which is crucial for assessing the image's structural integrity and the presence of artifacts. For example, edge detection is performed using Canny, Sobel, or Prewitt operators to obtain edge artifact parameters. Together, these methods constitute a comprehensive quantification of the teaching image data quality.

[0033] Based on the frequency component parameters, data defect parameters, and edge artifact parameters, an evaluation index for each teaching image data is determined. This evaluation index is a comprehensive quantification of image quality. For example, a weighted average method can be used to combine the frequency component parameters, data defect parameters, and edge artifact parameters to obtain a single numerical value or vector, which characterizes the overall quality level of the image.

[0034] The evaluation index for each teaching image data is compared with the preset evaluation index to obtain the data quality ratio for each teaching image data. The purpose of this step is to quantify the deviation of the current teaching image data from the ideal or standard quality. The preset evaluation index can be set based on industry standards, expert experience, or historical high-quality data. By comparison, the data quality ratio for each teaching image data can be obtained. This ratio intuitively reflects the current quality level of the image data; for example, a higher ratio indicates better image quality.

[0035] Using the data quality ratio of each teaching image data set, data analysis is performed on each teaching image data set and its basic information to obtain the urgency, complexity, demand coefficient, and processing load coefficient for each teaching institution. The data quality ratio is introduced as a correction factor to adjust the coefficients. For example, for image data with lower quality, its processing urgency may be appropriately increased, or its complexity may be reassessed to ensure more accurate results in subsequent processing.

[0036] Specifically, after uploading multiple teaching image data sets, the educational institution first performs frequency component analysis on one of the images. This reveals a low number of high-frequency components, indicating blurred image details, and yields frequency component parameters. Next, local contrast analysis reveals low local contrast, indicating insufficient image sharpness, and yields data defect parameters. Then, edge detection analysis reveals obvious jagged artifacts at the image edges, yielding edge artifact parameters. Based on these parameters, an evaluation index for the teaching image data is derived. For example, this evaluation index might have a low score, indicating poor image quality. Subsequently, this evaluation index is compared with a preset evaluation index representing high-quality images, resulting in a low data quality ratio, such as 0.6, indicating that the image quality is below average. After this initial analysis of the teaching image data and its basic information, an initial data urgency, data complexity, educational institution demand coefficient, and processing load coefficient may be obtained. For example, the initial urgency might be medium, and the complexity average. At this point, the initial coefficients are adjusted using a data quality ratio of 0.6. Specifically, due to the low image quality, its data complexity may be adjusted upwards, for example, from general to high, to reserve more processing resources. Simultaneously, to ensure the accuracy of the analysis results, its urgency may also be appropriately increased to prioritize processing or allocate more specialized analysis resources. By using the data quality ratio as a correction factor, the final determined data urgency, data complexity, educational institution demand coefficient, and processing load coefficient can more accurately reflect the actual processing needs of the teaching image data.

[0037] This embodiment analyzes each teaching image data point using frequency components, local contrast, and edge detection, enabling a comprehensive quantification of the image's intrinsic quality from multiple dimensions. Frequency component parameters reflect the image's detail richness, local contrast parameters reveal its sharpness and information carrying capacity, while edge artifact parameters indicate the image's structural integrity and potential distortion. This allows for a deeper assessment of image quality down to its underlying features. By determining evaluation indicators for each teaching image data point and comparing them with preset evaluation indicators, a data quality ratio is obtained. This ratio, as a quantitative quality metric, objectively reflects the quality of each teaching image data point relative to a standard quality level. It allows for the use of the image's inherent quality factors as a reference when conducting subsequent data analysis on each teaching image data point and its basic information. For example, images with poor quality may be more difficult to process or require more urgent processing to avoid information loss. Therefore, the data quality ratio can serve as an adjustment factor to correct the initially determined data urgency, data complexity, educational institution demand coefficients, and processing load coefficients, making the coefficient determination more accurate and reasonable.

[0038] In some embodiments of this application described above, the step of using the data quality ratio of each teaching image data to perform data analysis on each teaching image data and each data basic information to obtain the urgency of each data, the complexity of each data, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution includes: A preliminary data analysis is performed on each teaching image data and each piece of basic data information to obtain the initial urgency, initial complexity, initial demand coefficient of each teaching institution, and initial processing load coefficient of each teaching institution. Specifically, the preliminary data analysis refers to a basic assessment of the teaching image data and basic data information without relying entirely on the data quality ratio. For example, preliminary judgments can be made based on factors such as data volume, file type, upload time, and the typical demand patterns of teaching institutions. The purpose is to establish baseline coefficients without quality correction. Among them, the initial urgency, initial complexity, initial demand coefficient of teaching institutions, and initial processing load coefficient of teaching institutions are the original assessment values ​​obtained after the preliminary analysis without adjustment for the data quality ratio. The initial values ​​reflect the urgency, processing difficulty, institutional demand, and system load of the data under ideal or average conditions.

[0039] By utilizing the data quality ratio of each teaching image dataset, adjustments are made to the initial urgency, initial complexity, initial demand coefficient, and initial processing load coefficient of each data set. This yields the urgency, complexity, demand coefficient, and processing load coefficient for each institution. The adjustment process involves modifying these initial coefficients using the data quality ratio. For example, a low data quality ratio indicates numerous defects or artifacts, requiring a higher urgency (for priority processing and repair or reacquisition), higher complexity (due to increased processing difficulty), or corresponding adjustments to the demand and processing load coefficients for each institution. For instance, low-quality data may require more resources or longer processing time, thus impacting the workload. The goal is to ensure that the final coefficients more accurately reflect the actual data quality and its impact on subsequent processing.

[0040] Specifically, after educational institutions upload a batch of teaching image data, a preliminary analysis is first conducted on the basic information of the teaching image data and its associated data. For example, based on information such as the file size of the images, the type of course they belong to, and the preset submission deadline, the initial urgency, initial complexity, initial demand coefficient, and initial processing load coefficient of each teaching institution are initially determined. Specifically, if the image file is large, the initial complexity may be high; if the deadline is approaching, the initial urgency is high. Subsequently, these initial coefficients are adjusted using the previously determined data quality ratio of each teaching image data. For example, if the data quality ratio of the teaching image data is low (indicating defects such as blurriness, noise, or artifacts), even if its initial urgency is not high, its data urgency is increased to ensure that the low-quality data can be prioritized for processing or repair; at the same time, its data complexity may also be increased to reflect the additional resources and time required to process the defective data. Conversely, if the data quality ratio is high, the initial coefficients may be maintained or fine-tuned. Through this adjustment process, the final data urgency, data complexity, demand coefficient of each teaching institution, and processing load coefficient of each teaching institution will more accurately reflect the actual situation of the data and processing needs.

[0041] This embodiment first performs preliminary data analysis on each teaching image data and its basic information to obtain the initial urgency, initial complexity, initial demand coefficient, and initial processing load coefficient for each teaching institution. This establishes a baseline assessment without quality correction. Then, the data quality ratio for each teaching image data is used to adjust these initial coefficients. This step-by-step processing approach allows the data quality ratio to serve as a refined correction factor, optimizing the preliminary assessment results. This avoids assessment biases that might result from direct analysis and ensures that the final determined coefficients more accurately reflect the actual quality of the data and its impact on subsequent processing.

[0042] In some embodiments of this application described above, the step of allocating resources for each teaching image data based on its processing priority to obtain each resource analysis path includes: Based on the processing priority of each teaching image data, a queuing sequence for each teaching image data is determined. This step involves sorting all the teaching image data to be processed according to the pre-calculated processing priority of each data point, forming an ordered processing queue. Teaching image data with higher processing priority will be placed at the front of the queue to ensure they receive resources first. The purpose of this queuing sequence is to establish a structured resource allocation order, thereby optimizing resource utilization efficiency and meeting the timeliness requirements of processing different data.

[0043] Using the queuing sequence of each teaching image data, an initial resource scheduling allocation is performed for each teaching image data to obtain an initial analysis path for each resource. This step refers to allocating available computing resources, storage resources, and network resources to each teaching image data according to the sequence after determining the queuing sequence, forming a preliminary resource analysis path. The initial allocation aims to provide a preliminary feasible processing environment for each teaching image data to facilitate subsequent data analysis.

[0044] Each initial resource analysis path undergoes path planning verification to obtain a qualified analysis path. This step involves checking and verifying the effectiveness, feasibility, and efficiency of each initial resource analysis path obtained from the initial allocation. The verification process may include, but is not limited to, checking for resource conflicts, network latency, whether storage capacity meets requirements, and whether the path meets preset performance indicators. Through verification, unqualified paths can be identified and corrected, ensuring that the resource analysis paths ultimately used for data analysis are efficient, reliable, and compliant.

[0045] Specifically, after educational institutions upload a large amount of teaching image data, it is assigned different processing priorities based on factors such as urgency and complexity. First, all teaching image data is sorted according to these priorities to form a processing queue; for example, exam analysis data requiring urgent processing is placed before ordinary teaching video data. Next, based on this queue, resources such as computing servers, storage space, and network bandwidth are initially allocated to each teaching image data, forming an initial resource analysis path. For example, high-priority data is allocated to a high-performance computing cluster, while low-priority data is allocated to a regular server. Subsequently, the initially allocated resource analysis path is validated. The validation process may include simulating data flow, checking server load, assessing network latency, and verifying storage access permissions. If a resource bottleneck (e.g., server overload) or excessively high network latency is found in an initial path, the path is automatically adjusted, and resources are reallocated until a validated, efficient, and conflict-free resource analysis path is obtained. For example, tasks on overloaded servers are migrated to other idle servers, or a better network transmission path is selected.

[0046] This embodiment ensures the orderly allocation of resources and guarantees priority by determining the queuing sequence based on processing priority. Subsequently, initial resource scheduling and allocation are performed, constructing a preliminary processing path for each teaching image data. Crucially, by verifying the initial analysis path for each resource, potential resource conflicts, inefficiencies, or unreasonable paths can be identified and corrected in a timely manner. This avoids directly using potentially flawed resource paths for data analysis without verification, effectively improving the accuracy and reliability of resource scheduling and allocation.

[0047] In some embodiments of this application described above, the step of performing initial resource scheduling allocation for each teaching image data using the queuing sequence of each teaching image data to obtain an initial analysis path for each resource includes: The queuing sequence for each teaching image data is validated to obtain a valid queuing sequence for each teaching image data. This step involves checking the completeness, validity, and consistency of each data item in the determined queuing sequence. For example, it can check for duplicate teaching image data identifiers, invalid priority settings, or discrepancies with basic data information. The purpose is to ensure the accuracy and reliability of the queuing sequence, providing high-quality input for subsequent resource scheduling and allocation.

[0048] Based on the queuing sequence of each verified teaching image data, a resource scheduling library is determined. This step dynamically identifies and integrates currently available computing, storage, and network resources to form a structured resource set based on the specific processing requirements of the teaching image data in the queuing sequence (e.g., required computing resource types, storage requirements, and processing time limits). This resource scheduling library may contain information on various processing units (such as CPUs and GPUs), storage devices, and specific analysis tools or software modules, and records their current load, available capacity, and performance indicators. Its purpose is to provide a clear and real-time view of available resources for initial resource scheduling allocation, ensuring the rationality and efficiency of resource allocation.

[0049] Using the resource scheduling library and the queue sequence of each verified teaching image data, initial resource scheduling allocation is performed for each teaching image data to obtain an initial analysis path for each resource. This step refers to preliminary resource matching and allocation based on the priority and processing requirements of each teaching image data in the verified queue sequence, combined with the availability and capabilities of various resources in the resource scheduling library. For example, high-priority teaching image data may be allocated to high-performance, low-load computing resources, while teaching image data requiring specific processing capabilities will be allocated to resources with corresponding capabilities. The purpose is to generate a preliminary resource analysis path for each teaching image data, laying the foundation for subsequent path planning and verification.

[0050] Specifically, after educational institutions upload a large amount of teaching video and image data, they are assigned different processing priorities based on factors such as urgency and complexity, forming an initial queuing sequence. To ensure the accuracy of subsequent resource allocation, this queuing sequence is first validated. For example, it checks whether each video file is complete, whether the encoding format meets requirements, and whether associated metadata (such as course ID and teacher information) is missing or incorrect. Any non-compliant data items are marked or corrected, resulting in a validated queuing sequence. Subsequently, based on the specific processing needs of each teaching image data in the validated queuing sequence (e.g., video transcoding requires GPU resources, and image recognition requires AI model inference resources), a resource scheduling library is dynamically determined. This resource scheduling library may contain multiple high-performance computing servers (equipped with different models of CPUs and GPUs), large-capacity storage arrays, and virtual machine instances pre-installed with specific data analysis software, and their load and availability are updated in real time. Finally, using the resource scheduling library containing real-time resource information and the validated queuing sequence for each teaching image data, initial resource allocation is performed for each teaching image data. For example, high-priority 4K instructional videos might be assigned to a server equipped with the latest GPUs and currently under low load for transcoding and preliminary analysis, while a batch of low-priority instructional images might be distributed across multiple CPU servers for parallel processing. This generates a dedicated initial resource analysis path for each instructional image data set, laying a solid foundation for subsequent path planning verification and final data analysis and processing.

[0051] This embodiment verifies the queuing sequence of each teaching image data, enabling timely detection and correction of potential errors or inconsistencies, thus ensuring the reliability of input data for subsequent resource scheduling and allocation. Secondly, by determining the resource scheduling library based on the verified queuing sequence, a comprehensive understanding of the currently available resources and their characteristics is achieved. This allows for a more accurate match between the processing needs of the teaching image data and the actual available resources during initial resource scheduling and allocation. Because the reliability of the queuing sequence is improved, and the resource scheduling library provides a clear resource view, a more reasonable and efficient initial resource analysis path can be generated when using both for initial resource scheduling and allocation, avoiding allocation deviations caused by inaccurate data or resource information.

[0052] In some embodiments of this application described above, the steps of performing data analysis processing on each teaching image data using each resource analysis path to obtain teaching data analysis results include: Each resource analysis path is used to process each teaching image data, resulting in initial data analysis results and an analysis quality feedback coefficient. The initial data analysis results refer to the raw analysis output obtained during the preliminary analysis phase, after processing the teaching image data based on the assigned resource analysis path, without final verification or optimization. This includes various statistical data, pattern recognition results, and behavioral analysis reports. The analysis quality feedback coefficient is an indicator that quantitatively evaluates the quality, confidence level, or reliability of the initial data analysis results. This coefficient can be generated in various ways, such as by comparing it with preset quality standards, evaluating it using internal verification algorithms, or statistically analyzing anomalies that occur during the analysis process. Its purpose is to reflect the potential biases or degree of improvement required in the initial analysis results.

[0053] The initial results of the data analysis are adjusted using the aforementioned analysis quality feedback coefficient to obtain the teaching data analysis results. The adjustment process refers to the process of correcting, optimizing, or refining the initial results of the data analysis based on the quality status indicated by the analysis quality feedback coefficient. Specifically, if the feedback coefficient indicates that the initial results have quality problems or deviations, corresponding adjustment strategies can be triggered, such as rerunning part of the analysis process, applying correction algorithms, filtering outlier data points, or combining other auxiliary information for correction.

[0054] Specifically, it is necessary to analyze students' classroom behavior image data to assess their concentration. First, using a resource analysis path, the image data is preliminarily processed to identify students' head posture, eye direction, and body movements at different time periods, thus obtaining initial data analysis results, such as a time series graph of student concentration. Simultaneously, an analysis quality feedback coefficient is generated based on factors such as the internal consistency of these initial identification results, their matching degree with preset behavioral patterns, and the presence of abnormal fluctuations. For example, if there is a significant discrepancy between the student's head posture and eye direction identification results within a certain time period, or if the concentration curve shows drastic and irregular fluctuations, the feedback coefficient will indicate a low analysis quality. Subsequently, this analysis quality feedback coefficient is used to adjust the initial data analysis results. Specifically, if the feedback coefficient is low, a more refined image recognition algorithm may be used for a secondary analysis of the image data for that time period, or outlier data points may be smoothed. Cross-validation with other sensor data (such as speech recognition results) may even be performed to correct the initial concentration time series graph, ultimately obtaining more accurate and reliable teaching data analysis results.

[0055] This embodiment, after obtaining the initial data analysis results, does not directly use them as the final output. Instead, it first generates an analysis quality feedback coefficient. This feedback coefficient, as a self-evaluation mechanism, can quantitatively reflect the quality of the initial analysis results. Because of this feedback coefficient, potential deviations or deficiencies in the initial results can be identified, and targeted adjustments can be made accordingly. This ensures that the final output of the teaching data analysis results is optimized and refined, thereby significantly improving the accuracy, reliability, and practical value of the analysis results.

[0056] In some embodiments of this application described above, the step of adjusting the initial data analysis results using the analysis quality feedback coefficient to obtain the teaching data analysis results includes: Based on the analysis quality feedback coefficient and the preset feedback coefficient, the review and adjustment plan is confirmed. Specifically, the analysis quality feedback coefficient is a quantitative indicator automatically generated after data analysis of each teaching image data, based on various indicators during the analysis process (such as data integrity, model fit, and outlier detection results) and the deviation between the preliminary analysis results and the expected goals. This coefficient reflects the reliability of the initial data analysis results and the degree of adjustment required. The preset feedback coefficient is a pre-set set of thresholds, rules, or model parameters used to compare or combine with the actually generated analysis quality feedback coefficient to guide subsequent adjustment decisions. Its purpose is to provide a benchmark or reference for the adjustment process, ensuring the rationality and consistency of the adjustments.

[0057] The initial data analysis results are adjusted using the aforementioned review and adjustment scheme to obtain the teaching data analysis results. Confirming the review and adjustment scheme means that the system intelligently selects or generates a specific set of adjustment strategies and methods based on the comparison or combination of the analysis quality feedback coefficient and the preset feedback coefficient. For example, if the analysis quality feedback coefficient indicates a significant deviation in the initial data analysis results, exceeding the range allowed by the preset feedback coefficient, a more stringent review and adjustment scheme may need to be initiated, such as re-analyzing part of the data, introducing expert manual review, or adopting a more complex correction algorithm. Conversely, if the feedback coefficient is within an acceptable range, only fine-tuning or automated correction may be required. This review and adjustment scheme may include, but is not limited to, adjusting data cleaning rules, reconfiguring model parameters, strategies for handling outlier data points, or optimizing the result presentation method. This step refers to automatically or semi-automatically correcting, optimizing, or improving the initial data analysis results according to the confirmed review and adjustment scheme. For example, if the review and adjustment plan indicates that a specific type of data needs to be re-cleaned, the corresponding cleaning operation will be performed; if it indicates that model parameters need to be adjusted, the model will be updated according to the plan; if it indicates that the results need to be visualized and optimized, clearer and more insightful charts or reports will be generated according to the plan. The purpose is to eliminate or mitigate biases, errors, or deficiencies in the initial results of data analysis through systematic adjustments, thereby improving the accuracy, reliability, and practicality of the final teaching data analysis results.

[0058] Specifically, after analyzing the image data of students' learning behaviors in the class, initial data analysis results are obtained. At this point, an analysis quality feedback coefficient is generated, for example, 0.75 (out of 1, indicating higher quality). Simultaneously, a preset feedback coefficient threshold is set, for example, 0.8. Since 0.75 is lower than 0.8, the quality of the initial result is judged to have not met the expected standard and adjustment is required. Specifically, a review and adjustment plan is confirmed. This plan may include: first, performing secondary enhancement processing on the blurred areas identified in the original image data; second, adjusting the confidence threshold of the deep learning model used for behavior recognition to reduce false positives; and finally, cross-validating the adjusted results. After confirming this review and adjustment plan, it is used to adjust the initial data analysis results, such as reprocessing some data, rerunning the model, and applying the new threshold. The final result is an optimized and corrected teaching data analysis result, whose analysis quality feedback coefficient may be improved to 0.88, thus meeting the preset standard and providing teachers with a more reliable student learning behavior analysis report.

[0059] This embodiment provides a real-time quantitative assessment of the quality of the initial data analysis results by analyzing the quality feedback coefficient, while the preset feedback coefficient provides an objective benchmark for this assessment. When the two are compared, the quality status of the current initial data analysis results and the degree and direction of adjustment required can be accurately determined. For example, when the feedback coefficient indicates low result quality, a more stringent adjustment scheme is selected; when the result quality is high, slight adjustments or no adjustment may be chosen. This ensures the intelligence and adaptability of the adjustment process, enabling adjustments to precisely target the weak points of the initial data analysis results, avoiding unnecessary resource waste, and improving the efficiency and effectiveness of the adjustments.

[0060] In some embodiments of this application described above, the step of obtaining each teaching image data uploaded by the teaching institution and the basic data information associated with each teaching image data includes: This step involves obtaining the raw data for each teaching image uploaded by the educational institution, along with the associated basic information. This means directly receiving the raw, unprocessed teaching image data and its associated basic information from the educational institution. The raw data may include images in various file formats and resolutions, as well as structured or unstructured text information.

[0061] Each teaching image's original data and its basic information are preprocessed to obtain preprocessed original data and basic information for each teaching image. This step involves a series of cleaning, transformation, and normalization operations on the received raw data. Specifically, data preprocessing may include image format conversion, size adjustment, noise removal, and brightness / contrast optimization to ensure the image data meets the unified standards for subsequent analysis. Simultaneously, for the basic information, preprocessing may include data format standardization, missing value imputation, outlier handling, and text normalization, aiming to eliminate inconsistencies, redundancy, and errors in the original data, thereby improving data quality and usability.

[0062] Each preprocessed teaching image's original data and its basic information are individually validated to obtain valid teaching image data and associated basic information. This step involves checking the completeness, consistency, and validity of the preprocessed data. For example, it checks whether image files are corrupted, whether image content matches the teaching scenario, whether basic information fields are complete, whether data types are correct, and whether data values ​​are within reasonable ranges. The purpose is to ensure that the data is accurate, reliable, and conforms to preset specifications before entering subsequent analysis processes. Data validation effectively identifies and removes unqualified data, thus ensuring the quality of subsequent analysis.

[0063] Specifically, educational institutions need to upload image data of students' online learning behavior (e.g., screenshots of students on the learning platform and video frames of their answering process) and related student information (such as student ID, course name, and learning duration). First, the raw teaching image data and the basic information associated with each teaching image are obtained. For example, images may exist in various formats such as JPEG and PNG, with varying resolutions; student information may include handwritten input, improperly formatted dates, or missing course IDs. Next, the raw data undergoes preprocessing. Specifically, image data can be uniformly converted to PNG format and adjusted to a standard resolution (e.g., 1920x1080 pixels), while image denoising algorithms are applied to remove interfering pixels from screenshots. The basic information is cleaned by text cleaning, all dates are standardized to YYYY-MM-DD, and student IDs are formatted, ensuring they are purely numeric and of fixed length. If a student's course ID is found to be missing, attempts may be made to fill it in from other associated data, or it may be marked for manual review. Finally, the preprocessed data is validated. For example, the system verifies whether image files are complete and readable, and whether the image content is clear and identifiable; it also verifies whether student IDs exist in the registered student database, whether course names match the existing course list, and whether the study duration is positive and within a reasonable range. Any data that fails verification will be isolated or marked, and a corresponding error report will be generated for manual intervention or re-uploading. Only data that passes all verifications will be confirmed as qualified teaching image data, along with the basic data information associated with each teaching image data, and will then proceed to the subsequent processing priority determination and resource scheduling allocation stages. In this way, it ensures that the data entering the analysis process is high-quality, reliable, and consistent.

[0064] This embodiment introduces data preprocessing and data verification steps to progressively filter and optimize the raw data. First, it acquires the raw teaching image data and basic data information uploaded by educational institutions, which serves as the starting point for data analysis. Then, through data preprocessing, this raw data is cleaned, transformed, and standardized to avoid potential format inconsistencies and noise interference, making the data more organized and easier to process. Furthermore, through data verification, the preprocessed data undergoes rigorous quality checks to ensure its integrity, consistency, and validity, effectively preventing analytical biases or errors caused by data quality issues. This ensures that the final teaching image data and basic data information are of high quality, reliable, and meet the analytical requirements.

[0065] For a data analysis method for digital and intelligent teaching in education based on any of the above, please refer to [link / reference needed]. Figure 2The present invention also provides an education digitalization teaching data analysis system based on big data, which includes a data acquisition module 210, a priority determination module 220, a scheduling and allocation module 230 and a data analysis module 240.

[0066] The data acquisition module 210 is used to acquire each teaching image data uploaded by the teaching institution and the basic information of the data associated with each teaching image data.

[0067] The priority determination module 220 is used to determine the processing priority of each teaching image data according to each teaching image data and each data basic information; The scheduling and allocation module 230 is used to schedule and allocate resources for each teaching image data according to the processing priority of each teaching image data, so as to obtain each resource analysis path.

[0068] The data analysis module 240 is used to perform data analysis and processing on each teaching image data using each resource analysis path to obtain teaching data analysis results.

[0069] In this embodiment, the data acquisition module 210 ensures the comprehensive and accurate input of teaching image data and its associated basic information; the priority determination module 220 intelligently evaluates and allocates the data processing order, effectively addressing data heterogeneity and urgency; the scheduling and allocation module 230 optimizes the utilization of computing resources, planning efficient analysis paths for each data task; and finally, the data analysis module 240 performs in-depth processing of the data according to the preset path, thereby outputting timely and accurate teaching data analysis results. This significantly improves the efficiency of data processing and the reliability of analysis results, enhancing the accuracy and reliability of the final teaching data analysis results and providing strong support for educational decision-making.

[0070] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the present invention specification.

Claims

1. A data analysis method for digital and intelligent teaching in education based on big data, characterized in that, include: Obtain each teaching image data uploaded by the teaching institution, as well as the basic data information associated with each teaching image data; Based on each teaching image data and its basic information, the processing priority of each teaching image data is determined. By utilizing the processing priority of each teaching image data, resource scheduling and allocation are performed on each teaching image data to obtain each resource analysis path; By utilizing each resource analysis path, data analysis and processing are performed on each teaching image data to obtain teaching data analysis results.

2. The data analysis method for digital teaching in education based on big data according to claim 1, characterized in that, The steps for determining the processing priority of each teaching image data based on each data's basic information include: Based on each teaching image data and each data basic information, determine the urgency of each data, the complexity of each data, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution; The processing priority of each teaching image data is determined based on the urgency of each data point, the complexity of each data point, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution.

3. The method for analyzing educational digital teaching data based on big data according to claim 2, characterized in that, The steps for determining the urgency, complexity, demand coefficient, and processing load coefficient of each teaching institution based on each teaching image data and its basic information include: Each teaching image data is analyzed for frequency components, local contrast, and edge detection to obtain frequency component parameters, data defect parameters, and edge artifact parameters. Based on the frequency component parameters, data defect parameters, and edge artifact parameters, an evaluation index for each teaching image data is determined. The evaluation index of each teaching image data is compared with the preset evaluation index to obtain the data quality ratio of each teaching image data. By using the data quality ratio of each teaching image data, data analysis is performed on each teaching image data and each data basic information to obtain the urgency of each data, the complexity of each data, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution.

4. The data analysis method for digital teaching in education based on big data according to claim 3, characterized in that, The steps of using the data quality ratio of each teaching image data to perform data analysis on each teaching image data and each data basic information to obtain the urgency of each data, the complexity of each data, the demand coefficient of each teaching institution, and the processing load coefficient of each teaching institution include: A preliminary data analysis is performed on each teaching image data and each data basic information to obtain the initial urgency of each data, the initial complexity of each data, the initial demand coefficient of each teaching institution, and the initial processing load coefficient of each teaching institution. Using the data quality ratio of each teaching image data, the initial urgency, initial complexity, initial demand coefficient of each teaching institution, and initial processing load coefficient of each teaching institution are adjusted to obtain the urgency, complexity, demand coefficient, and processing load coefficient of each teaching institution.

5. The method for analyzing educational digital teaching data based on big data according to claim 1, characterized in that, The steps for allocating resources for each teaching image data based on its processing priority to obtain each resource analysis path include: The queuing sequence of each teaching image data is determined according to the processing priority of each teaching image data. Using the queuing sequence of each teaching image data, an initial resource scheduling allocation is performed on each teaching image data to obtain an initial analysis path for each resource. For each resource's initial analysis path, perform path planning verification to obtain each resource analysis path that passes the verification.

6. The method for analyzing educational digital teaching data based on big data according to claim 5, characterized in that, The steps for initial resource scheduling and allocation for each teaching image data using the queuing sequence of each teaching image data to obtain the initial analysis path for each resource include: The sequence of each teaching image data is validated to obtain the valid sequence of each teaching image data. Based on the queuing sequence of each teaching image data that has passed the verification, a resource scheduling library is determined; Using the resource scheduling library and the queuing sequence of each qualified teaching image data, the initial resource scheduling allocation is performed on each teaching image data to obtain the initial analysis path for each resource.

7. The method for analyzing educational digital teaching data based on big data according to claim 1, characterized in that, The steps for performing data analysis and processing on each teaching image data using each resource analysis path to obtain the teaching data analysis results include: Using each resource analysis path, data analysis and processing are performed on each teaching image data to obtain the initial data analysis results and analysis quality feedback coefficient; The initial results of the data analysis are adjusted using the aforementioned analysis quality feedback coefficient to obtain the teaching data analysis results.

8. The method for analyzing educational digital teaching data based on big data according to claim 7, characterized in that, The steps for adjusting the initial data analysis results using the aforementioned analysis quality feedback coefficient to obtain the teaching data analysis results include: Based on the analysis quality feedback coefficient and the preset feedback coefficient, confirm the review and adjustment plan; Using the aforementioned review and adjustment scheme, the initial results of the data analysis are adjusted to obtain the teaching data analysis results.

9. The data analysis method for digital teaching in education based on big data according to claim 1, characterized in that, The steps for obtaining each teaching image data uploaded by the teaching institution and the basic information of the data associated with each teaching image data include: Obtain the original data of each teaching image uploaded by the teaching institution, as well as the original basic information of the data associated with each teaching image; Each teaching image's original data and its original basic information are preprocessed to obtain preprocessed original data and its original basic information for each teaching image. Each preprocessed original teaching image data and its basic information are verified to obtain each verified teaching image data and its associated basic information.

10. A data analysis system for digitalized teaching in education based on big data, characterized in that: The system includes: The data acquisition module is used to acquire each teaching image data uploaded by the teaching institution and the basic data information associated with each teaching image data; The priority determination module is used to determine the processing priority of each teaching image data based on each teaching image data and each data basic information; The scheduling and allocation module is used to allocate resources for each teaching image data according to the processing priority of each teaching image data, so as to obtain each resource analysis path; The data analysis module is used to perform data analysis and processing on each teaching image data using each resource analysis path, and obtain teaching data analysis results.

Images