Error question mining and accurate printing management system of learning situation big data
By constructing modules for evaluating the value of incorrect questions and perceiving student status, and dynamically adjusting the priority of printing tasks, the problem of delayed printing of incorrect questions in existing technologies has been solved, and the accurate and efficient transformation of learning data into physical teaching processes has been achieved.
Patent Information
- Application Number
- CN202512003710.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-29
- Publication Date
- 2026-04-07
- Estimated Expiration
- 2045-12-29
AI Technical Summary
The existing print management system cannot identify the differences in the teaching value of incorrect questions, which leads to high-urgency incorrect question printing tasks being blocked by low-value tasks, resulting in delayed teaching feedback and resource misallocation.
A big data-driven error analysis and precise printing management system for student learning is constructed, including a learning data collection module, an error value assessment module, a student status perception module, and a dynamic priority scheduling module. Through multi-dimensional quantitative analysis of error value and student status, the system dynamically adjusts the priority of printing tasks to achieve intelligent printing management.
It enables timely processing of highly urgent incorrect questions, ensuring that students receive targeted practice during the optimal learning window, optimizing the allocation of teaching resources, reducing operating costs and paper consumption, and improving the accuracy and efficiency of teaching resources.
Smart Images

Figure CN121411720B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application belongs to the technical field of educational data processing, and particularly relates to a wrong question mining and accurate printing management system of learning situation big data. BACKGROUND
[0002] In the field of educational informatization, the collection, analysis and application of learning situation big data have become a key technical direction for improving teaching quality and realizing personalized learning. By integrating students' behavior data and evaluation data on various learning platforms, a multi-dimensional student portrait can be constructed to provide data support for precise teaching intervention. Wrong question mining and targeted practice based on learning situation big data is an important practical path to improve learning efficiency. This path aims to analyze students' historical wrong question data, identify their knowledge weaknesses, and generate personalized consolidation practice materials.
[0003] In the prior art, a learning situation big data wrong question mining system can generally identify common wrong questions and individual wrong questions of students according to an algorithm model, and generate corresponding wrong question sets. However, in the process of converting these digital wrong question sets into physical practice materials, i.e. in the output link through the printing management system, there are technical bottlenecks.
[0004] The task queue of the existing printing management system generally adopts a simple first-in-first-out scheduling strategy, which completely ignores the differences in teaching value of different wrong question tasks. Specifically, it cannot identify high-urgency common wrong questions and low-priority general wrong questions, nor can it dynamically adjust according to the learning progress and priority of individual students.
[0005] This leads to the possibility that high-value, timely consolidation wrong question printing tasks may be blocked by a large number of low-value tasks, resulting in high printing delay rate, which in turn makes it impossible for teachers to obtain timely teaching feedback and for students to perform targeted practice in the best learning window period, ultimately causing delays in teaching opportunities and mismatches in teaching resources. Therefore, how to realize an accurate management system that can intelligently perceive the value of wrong questions and the needs of students and dynamically schedule the priority of printing tasks has become a technical problem to be solved. SUMMARY
[0006] The present application aims to provide a wrong question mining and accurate printing management system of learning situation big data to solve the contradiction in the prior art that the printing task scheduling strategy is single and cannot distinguish the differences in teaching value of wrong questions, leading to delays in high-value tasks and delays in teaching opportunities.
[0007] The present application provides a wrong question mining and accurate printing management system of learning situation big data, which includes a learning situation data collection module, a wrong question value evaluation module, a student state perception module, a dynamic priority scheduling module and an intelligent printing execution module.
[0008] The learning data collection module is used to continuously collect and integrate student behavior log data and assessment results data from multiple learning platforms to build a raw learning database in a unified format.
[0009] The error value assessment module is used to perform multi-dimensional quantitative analysis of error data in the original learning database, and calculate the global teaching value score and individual urgency score for each error.
[0010] The student status awareness module is used to analyze and update each student's learning progress profile and current learning focus status in real time based on the original learning information database.
[0011] The dynamic priority scheduling module receives output data from the error value assessment module and the student status perception module. Based on the preset composite priority calculation model, it generates a dynamic, real-time comprehensive priority score for each error task to be printed, and reconstructs the printing task queue accordingly.
[0012] The intelligent printing execution module is used to control the connected physical printing devices to execute printing jobs based on the task queue order reconstructed by the dynamic priority scheduling module, and to provide feedback on the printing status.
[0013] Furthermore, the error assessment module includes a common value analysis unit and an individual urgency analysis unit. The common value analysis unit is used to calculate the global teaching value score of the error, and its calculation process is as follows:
[0014] First, the error rate of the wrong question in the target student group is calculated. Wrong questions with an error rate greater than 20% of the preset threshold are marked as high-frequency common wrong questions.
[0015] Secondly, the weight of the knowledge points associated with the incorrect question in the curriculum standards is analyzed, and the weight values are assigned by the pre-set knowledge graph.
[0016] Next, assess the level of thinking involved in solving the incorrect problem. The level of thinking is divided into five levels: memory, understanding, application, analysis and synthesis, and each level corresponds to a different coefficient.
[0017] Ultimately, the overall teaching value score is calculated by weighting the error rate, the knowledge point weight, and the thinking level coefficient.
[0018] The individual urgency analysis unit is used to calculate the individual urgency score for incorrect answers. The calculation process is as follows:
[0019] First, we tracked the number of errors the student made on the same knowledge point throughout their history, and the number of errors was positively correlated with the urgency score.
[0020] Secondly, calculate the time interval between the last occurrence of the wrong question and the previous occurrence of the question. The shorter the time interval, the higher the urgency score.
[0021] Secondly, considering the student's recent accuracy rate trend in exercises related to this knowledge point, if the accuracy rate shows a downward trend, the urgency score should be increased.
[0022] Furthermore, the student status perception module includes a learning progress tracking unit and a focus assessment unit.
[0023] The learning progress tracking unit is used to maintain and update each student's learning progress profile, which includes the following data:
[0024] The system includes a list of completed knowledge points, a mastery score for each knowledge point, a marker for the knowledge points currently being studied, and a pre-set learning schedule.
[0025] The focus assessment unit is used to evaluate students' current learning focus in real time. Its assessment is based on real-time behavioral log data provided by the learning data collection module, specifically including:
[0026] The effective learning time of students in the current learning session, the frequency of page switching, and the response time of interactive quizzes;
[0027] By comparing the above behavioral data with a preset focus model, a quantitative focus index is output. When the focus index is less than 60%, it is judged as a low focus state.
[0028] Furthermore, the composite priority calculation model of the dynamic priority scheduling module is defined as follows: the comprehensive priority score is equal to the product of the basic value score, the dynamic adjustment factor, and the state decay factor.
[0029] The basic value score is a linear combination of the overall teaching value score for incorrect answers and the individual urgency score. A dynamic adjustment factor is used to incorporate time sensitivity and resource optimization considerations; its calculation method is as follows:
[0030] For tasks marked as frequently occurring common errors with an error rate greater than 30%, an acceleration factor greater than 1 is assigned; for tasks in the print task queue with a waiting time greater than a preset threshold of 10 minutes, a waiting compensation factor that increases linearly with time is assigned.
[0031] The state decay factor is associated with the output of the student state awareness module. Specifically, when a student is in a low-attention state, a decay factor of less than 1, such as 0.8, is assigned to all of their pending print jobs to reduce their immediate printing priority.
[0032] When a student's progress profile indicates that they are about to enter the learning stage of new knowledge points, a promotion coefficient greater than 1 is assigned to the old knowledge weak points and wrong questions that are strongly related to the new knowledge points.
[0033] Furthermore, the dynamic priority scheduling module also includes a queue reconstruction unit and a conflict resolution unit.
[0034] The queue refactoring unit sorts all pending print tasks in descending order based on the latest comprehensive priority score at a fixed period, such as every 5 minutes, and generates a new print task queue.
[0035] The conflict resolution unit is used to handle resource conflicts that may occur during the task queue reconstruction process. Its logic is as follows:
[0036] When multiple high-priority tasks compete for the same printing device, priority is given to executing tasks for which the associated student is highly focused and which involve frequently occurring common mistakes.
[0037] For tasks that are temporarily postponed due to conflicts, the waiting compensation factor in their dynamic adjustment factor will receive an additional increment to ensure that they are not delayed indefinitely.
[0038] Furthermore, the intelligent printing execution module includes a task parsing unit, a device driving unit, and a feedback unit.
[0039] The task parsing unit is used to extract the first task from the task queue output by the dynamic priority scheduling module, and convert the incorrect questions, format templates and student identification information contained therein into a standard print job description file.
[0040] The device driver unit is used to generate specific printer control commands based on the print job description file, and monitor the real-time status of the connected physical printing devices, including paper level, toner level, and online status.
[0041] The feedback unit is used to send confirmation information to the system core database after the print job is completed, update the status of the print job to complete, and record the completion timestamp; if a fault occurs during the printing process, it immediately sends an interrupt signal to the dynamic priority scheduling module to trigger task rescheduling.
[0042] Furthermore, the system also integrates a resource optimization module. This module operates on top of the intelligent printing execution module, and its functions include:
[0043] Analyze historical print job data to identify peak and off-peak periods in print volume;
[0044] During off-peak hours, the system automatically schedules and executes non-urgent, large-volume printing tasks of common error sets to balance the equipment load.
[0045] Meanwhile, the resource optimization module intelligently typeset and merges the content based on the output of the error value assessment module. When multiple students' errors involve the same high-value question, they are merged and printed on one sheet of paper, with clear labels indicating different students, thus saving paper consumption.
[0046] Furthermore, the system's operation flow is as follows:
[0047] The learning data collection module continuously inputs data; the error value assessment module and the student status perception module process the data in parallel and output the assessment results.
[0048] The dynamic priority scheduling module periodically obtains the above results, calculates the comprehensive priority score, and reconstructs the task queue;
[0049] The intelligent printing execution module executes printing according to the queue order;
[0050] The resource optimization module performs load balancing and content merging analysis in the background;
[0051] The entire process forms a closed-loop, adaptive, precision printing management pipeline.
[0052] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0053] 1. This invention, by constructing a module for assessing the value of incorrect questions and a module for perceiving student status, is the first to quantitatively integrate the intrinsic teaching value of incorrect questions with students' real-time learning status, providing a multi-dimensional and precise decision-making basis for printing scheduling. The dynamic priority scheduling module, based on a comprehensive priority score calculated by a composite model, completely replaces the simple first-in-first-out strategy, enabling high-urgency common incorrect questions and tasks related to students' high-focus states to be prioritized, reducing delays in key teaching feedback, and ensuring that students receive targeted practice materials during the optimal learning window, thereby effectively solving the fundamental problem of delayed teaching opportunities.
[0054] 2. This invention achieves coordinated optimization of system resources and student states through the design of a conflict resolution unit and a state decay factor in the dynamic priority scheduling module. The system can intelligently identify and postpone the immediate printing of tasks for students with low concentration, prioritizing the learning continuity of students with high engagement, while ensuring that postponed tasks are not lost through a waiting compensation mechanism. This design transforms print management from a mechanical task execution into an environmental regulator that promotes effective learning behavior, improving the accuracy and psychological effectiveness of teaching resource allocation.
[0055] 3. This invention, by introducing a resource optimization module, balances system efficiency and economic benefits while ensuring the priority of teaching. This module analyzes historical workloads for off-peak scheduling and significantly reduces paper consumption by intelligently merging common incorrect answers. This allows the system to effectively reduce operating costs and resource waste during long-term, large-scale deployments, demonstrating the forward-looking and practical nature of the technical solution. The entire system constitutes a complete closed-loop intelligent management system from data mining to physical output, realizing the precise and efficient transformation of the value of student learning big data into the actual teaching process. Attached Figure Description
[0056] Figure 1 This is a schematic diagram of the overall technical solution architecture of the error mining and precise printing management system for big data learning proposed in this invention;
[0057] Figure 2 This is a schematic diagram of the core principle framework of the dynamic priority scheduling module in this invention;
[0058] Figure 3 This is a logical flowchart of the error value assessment module and the student status perception module in this invention;
[0059] Figure 4 This is a schematic diagram of the multi-level interaction relationship and data flow between the error value assessment module, the student status perception module and the dynamic priority scheduling module in this invention;
[0060] Figure 5 This is a framework diagram of the collaborative workflow between the intelligent printing execution module and the resource optimization module in this invention. Detailed Implementation
[0061] This embodiment details the specific technical implementation of the error mining and precise printing management system based on big data of student learning.
[0062] Please refer to the attached document. Figures 1 to 5 This system is a closed-loop intelligent management system composed of multiple tightly coupled functional modules. Its core objective is to dynamically schedule printing tasks based on the teaching value of incorrect questions and the real-time status of students, so as to achieve the optimal and accurate delivery of teaching resources.
[0063] The system includes a learning data collection module, a wrong question value assessment module, a student status perception module, a dynamic priority scheduling module, an intelligent printing execution module, and a resource optimization module.
[0064] These modules interact with each other through preset data interfaces and communication protocols, forming a continuously operating data processing and physical execution pipeline.
[0065] The learning data acquisition module is the system's basic data entry point, responsible for continuously and stably acquiring raw learning data from heterogeneous learning environments.
[0066] This module is deployed on the system server and establishes secure connections with multiple third-party learning platforms through a set of standardized application programming interfaces.
[0067] These learning platforms include online assessment systems, digital assignment platforms, and interactive teaching software.
[0068] The data acquisition process of the learning data collection module is carried out around the clock, and its collection frequency can be adjusted according to the configuration. A typical setting is to poll the data update interface of each platform every 30 seconds.
[0069] The collected data is mainly divided into two categories:
[0070] Student behavior log data and assessment results data.
[0071] Student behavior log data specifically includes the sequence of student operations on the learning platform, such as login and logout timestamps, viewing duration and pause points of course videos, page browsing paths and dwell times of e-textbooks, click and answer time sequences of practice questions, and forum posting and reply records.
[0072] The assessment results data are more structured, including detailed records of each test or assignment, such as question labels, student answers, standard answers, scores, time spent answering, and the knowledge point codes associated with the questions.
[0073] The collected raw data are in various formats, and the learning data collection module has a built-in data cleaning and standardization sub-unit.
[0074] This subunit first performs validity checks on the raw data, removing obviously abnormal records, such as those with a response time of less than 100 milliseconds or more than 1 hour.
[0075] Subsequently, the data cleaning and standardization subunit maps data fields from different platforms into a unified internal representation based on a predefined global data dictionary.
[0076] For example, the "Question ID" field in Platform A and the "Question Number" field in Platform B can be uniformly mapped to the "Globally Unique Identifier for Questions" within the system.
[0077] After mapping is completed, the data is converted into a unified JSON format or a specific binary format and tagged with metadata tags such as the data source platform identifier, collection timestamp, and data version number.
[0078] Ultimately, the processed standardized data is written into the system's original learning database in real time.
[0079] The original learning data database adopts a hybrid architecture of distributed time-series database and relational database. High-frequency behavioral log data is stored in the time-series database for fast writing and time-range querying, while structured assessment result data is stored in the relational database to ensure transaction integrity and efficiency of complex relational queries.
[0080] The database table structure design includes student information table, behavior log table, assessment record table, knowledge point mapping table, etc. The tables are linked together by a globally unique identifier for each student and a globally unique identifier for each question.
[0081] The error value assessment module is one of the core analysis engines of the system. Its function is to conduct in-depth quantitative analysis of error records in the original learning database and assign two key scores to each error: a global teaching value score and an individual urgency score.
[0082] Please refer to the attached document. Figure 3 With appendix Figure 4 The error assessment module is logically divided into two independent but collaborative analysis units:
[0083] The system consists of a common value analysis unit and an individual urgency analysis unit. These two units typically operate in parallel computing to process the new stream of incorrect questions injected by the learning data acquisition module.
[0084] The common value analysis unit is responsible for evaluating the value of incorrect answers from a group teaching perspective. Its calculation process is a multi-step quantitative analysis procedure.
[0085] The first step is to calculate the error rate of the target question within a specific student group. This unit extracts records of all students answering the question within a pre-defined statistical period, such as the most recent 14 days, from the original student learning database.
[0086] The statistical period can be dynamically adjusted according to the teaching stage. When calculating the error rate, the numerator is the number of students who answered incorrectly, and the denominator is the total number of students who answered correctly.
[0087] The system presets a threshold for frequently missed questions, with a typical value of 20%.
[0088] When the error rate of a question is greater than 20%, the question is automatically marked as a "high-frequency common error" by the system. This mark will be used as an important feature in subsequent calculations.
[0089] The second step is to analyze the inherent weights of the knowledge points associated with the incorrect questions. The system maintains a knowledge graph built based on the curriculum standards. This graph stores the pre-, post-, and parallel relationships between knowledge points in a graph structure, and assigns a weight value to each knowledge point node.
[0090] The weight values are predefined by subject matter experts, reflecting the importance, difficulty, and frequency of the knowledge point in the entire curriculum and in entrance examinations.
[0091] The common value analysis unit obtains the weight value of the knowledge point corresponding to the wrong question by querying the knowledge graph. The weight value is usually normalized to the range of 0 to 1.
[0092] The third step is to assess the level of thinking behind the incorrect answers. The system has a pre-set classification system for thinking levels, which includes five levels: memorization, understanding, application, analysis, and synthesis.
[0093] Each level corresponds to a preset coefficient, for example, the memory coefficient is 0.6, the comprehension coefficient is 0.8, the application coefficient is 1.0, the analysis coefficient is 1.2, and the comprehensive coefficient is 1.5.
[0094] The common value analysis unit performs a shallow analysis of the question text and solution process through the natural language processing subunit, or directly reads the preset thinking level tags in the question metadata to determine the thinking level of the question and obtain the corresponding coefficient.
[0095] After extracting data from the three dimensions mentioned above, the common value analysis unit performs a weighted calculation to derive the global teaching value score. The calculation uses a weighted linear summation model. Let the normalized error rate be denoted as... The knowledge point weight value is The thinking level coefficient is Overall teaching value score The calculation formula is:
[0096] ;
[0097] The preset weighting coefficients are used, and they satisfy the following conditions: These coefficients reflect the relative importance that system designers place on error rate, knowledge point importance, and thinking level. A typical configuration could be... Calculated The score is a real number greater than 0. The higher the score, the more valuable the wrong question is to be addressed from the perspective of group teaching.
[0098] The individual urgency analysis unit assesses the urgency of incorrect answers from the perspective of individual student learning trajectories.
[0099] The calculations in this unit are performed on a specific combination of a particular student and a specific wrong answer.
[0100] The calculation process also involves multiple dimensions. The first dimension is the frequency of historical errors.
[0101] The unit query records the student's answers to all questions on the same knowledge point over a relatively long period of time, such as the entire semester, and counts the total number of errors.
[0102] The number of errors is directly positively correlated with urgency. The system maps the original number of errors to a standardized urgency component using a nonlinear function, such as a logarithmic function or a piecewise linear function, to prevent an excessive influence from a single dimension.
[0103] The second dimension is the error reproduction time interval.
[0104] The unit calculates the number of days between the time when the current mistake occurred and the time when the student last made a mistake on the same knowledge point.
[0105] The shorter the interval, the more frequently the knowledge gap recurs, and the greater the urgency.
[0106] The system uses a preset time decay function; the shorter the interval, the higher the score in this dimension. The third dimension is the recent accuracy trend.
[0107] The unit obtains the student's accuracy data for all exercises involving this knowledge point in a recent period, such as the last 5 times, and judges the trend by linear regression or simple comparison of the first and last values.
[0108] If the accuracy rate shows a downward trend, it indicates that the student's mastery of the knowledge point is deteriorating, and the system will assign a higher positive score to this dimension.
[0109] If the trend is stable or rising, a lower score or zero score will be given.
[0110] The Individual Urgency Analysis Unit combines the scores from the three dimensions mentioned above to derive the Individual Urgency Score (IP).
[0111] The comprehensive approach can be a weighted summation or a logical combination based on business rules. The final IP score is also a real value; a higher score indicates that the specific student needs to practice and reinforce that particular question as soon as possible.
[0112] The student status awareness module is another core analysis engine. It is dedicated to building and updating dynamic learning profiles of students in real time, focusing on two key states: learning progress and current concentration.
[0113] Please refer to the appendix. Figure 3 With appendix Figure 4 This module includes a learning progress tracking unit and a focus assessment unit.
[0114] The learning progress tracking unit is responsible for maintaining a long-term learning progress profile for each student.
[0115] This profile is a structured data object stored in a dedicated student profile database. The profile contains the following core fields:
[0116] The system includes a globally unique student identifier, a list of completed knowledge points and their mastery scores, an identifier for the knowledge points currently being studied, a pre-set personal learning plan schedule, and a summary of recent learning activities.
[0117] The mastery score is a value between 0 and 100, calculated by combining the student's average accuracy rate across all historical assessments on the corresponding knowledge point, the accuracy rate of the most recent assessment, and the stability of their answers.
[0118] The identification of the knowledge point currently being learned is automatically inferred by the learning progress tracking unit by analyzing the student's recent behavior logs. For example, when a student accesses the learning resources for a certain knowledge point multiple times in a row and completes the related exercises, the system determines that the student is in the learning stage of that knowledge point.
[0119] The study schedule may be imported from the academic affairs system or set by the students themselves.
[0120] This unit updates the learning progress profile on a daily basis. Each time new assessment or learning behavior data is received, the relevant fields are recalculated and refreshed.
[0121] The focus assessment unit is responsible for evaluating students' real-time learning status. This unit runs at a higher frequency, such as once per minute.
[0122] The assessment is based entirely on the real-time behavior log data stream provided by the learning data collection module.
[0123] The attention assessment unit defines a set of behavioral indicators related to attention:
[0124] Effective learning time, page switching frequency, and response time for interactive quizzes.
[0125] Effective learning time refers to the cumulative time during which students continuously perform learning-related operations without prolonged interruptions within the assessment window.
[0126] Page switching frequency refers to the number of times a student switches between different learning resource pages per unit of time.
[0127] The response time for interactive quizzes refers to the interval between when the question is presented and when the student begins to answer.
[0128] The unit has a pre-set focus calculation model, which can be a rule-based scoring system or a lightweight machine learning model.
[0129] The model receives the above behavioral indicators as input and calculates and outputs a quantified focus index, which typically ranges from 0 to 100.
[0130] The system sets a threshold for judging focus level, with a typical value of 60%.
[0131] When the calculated focus index is less than 60%, the focus assessment unit determines that the student is currently in a "low focus state" and generates a state identifier.
[0132] Conversely, a "high level of focus" state is indicated by a low level of focus. This state identifier, along with the focus index itself, serves as the core output of the student's real-time status.
[0133] The dynamic priority scheduling module is the decision-making center of the entire system. It receives the global teaching value score and individual urgency score from the error value assessment module, as well as the learning progress profile and concentration status from the student status perception module. Based on a composite calculation model, it calculates a dynamic comprehensive priority score for each error task to be printed, and then intelligently schedules the printing queue.
[0134] Please refer to the attached document. Figure 2 With appendix Figure 4 This module is key to the system's precise scheduling.
[0135] The core of the dynamic priority scheduling module is the composite priority calculation model. This model defines the comprehensive priority score as the product of three factors: a base value score, a dynamic adjustment factor, and a state decay factor. The base value score is a direct reflection of the error assessment module's output, derived from the global teaching value score. With individual urgency scores It is composed of linear combinations, that is:
[0136] ;
[0137] in and This is a weighted combination used to balance the relative importance of group value and individual urgency.
[0138] Dynamic adjustment factors are used to introduce system-level optimization and time sensitivity considerations.
[0139] The calculation of this factor involves multiple sub-rules.
[0140] The first sub-rule is an acceleration rule targeting frequently asked common mistakes:
[0141] For tasks marked as "high-frequency common mistakes" by the error value assessment module and whose error rate is greater than another higher threshold of 30%, the system assigns an acceleration factor greater than 1, such as 1.2, to increase their priority.
[0142] The second sub-rule is a compensation rule for task waiting time: the system records the timestamp of each print task entering the print queue.
[0143] When the task waiting time exceeds the preset waiting threshold, which is typically 10 minutes, the system will activate a waiting compensation factor that increases linearly over time.
[0144] For example, after more than 10 minutes, the compensation factor increases by 0.05 for every additional minute of waiting, until it reaches the upper limit such as 2.0, thus ensuring that no task is put on hold indefinitely.
[0145] The state decay factor is used to incorporate the real-time state of students into scheduling decisions.
[0146] The calculation of this factor is closely related to the output of the student status awareness module.
[0147] Specifically, it includes two core rules.
[0148] The first rule is the focus decay rule:
[0149] When the attention assessment unit determines that a student is in a low-attention state, the system assigns a decay coefficient of less than 1 to all of the student's pending print tasks, with a typical value of 0.8.
[0150] This means that when students are not focused, the overall priority of their error printing tasks will be systematically lowered, and system resources will be prioritized for students who are highly focused, in order to match their optimal learning window.
[0151] The second rule is the learning progress promotion rule:
[0152] When the learning progress tracking unit outputs that a student is about to enter the learning stage of new knowledge points according to the learning plan or actual progress, the system will scan the student's queue of unprinted wrong questions to find those old knowledge weaknesses that have a strong prerequisite relationship with the new knowledge points to be learned.
[0153] For these incorrect answers, the system assigns a promotion coefficient greater than 1, such as 1.3, to prioritize printing these materials, helping students consolidate prior knowledge and clearing obstacles for learning new knowledge.
[0154] The final calculation of the overall priority score is to multiply the three factors mentioned above.
[0155] This score is a dynamic, real-valued score that is constantly updated over time, depending on changes in the student's status and the addition of new tasks.
[0156] The dynamic priority scheduling module also includes a queue reconstruction unit and a conflict resolution unit.
[0157] The queue reconfiguration unit runs automatically at a fixed period, typically set every 5 minutes.
[0158] At the start of each cycle, the queue reconstruction unit retrieves all tasks with a status of "pending printing" from the internal cache or database and reads the latest comprehensive priority score for each task.
[0159] Subsequently, the unit sorts all tasks in descending order according to their overall priority scores, generating a new, ordered queue of printing tasks.
[0160] This new queue will replace the old queue and serve as the basis for the execution of the intelligent printing execution module.
[0161] The conflict resolution unit is responsible for handling resource contention issues that may arise during queue reconstruction and task execution.
[0162] Its logic is activated in two scenarios.
[0163] The first scenario involves multiple tasks competing for a single device:
[0164] When multiple high-priority print jobs are directed to the same physical printing device, and the device cannot process them simultaneously, the conflict resolution unit makes a decision based on a set of preset arbitration rules.
[0165] The rules prioritize task combinations where "the associated students are in a state of high concentration" and "the task consists of frequently asked common mistakes".
[0166] This design ensures that instructional interventions are delivered promptly when students are most engaged and when the problems are most prevalent.
[0167] The second scenario is that the task is delayed:
[0168] For tasks that are temporarily removed from the execution sequence due to resource conflicts, the conflict resolution unit will notify the dynamic adjustment factor calculation logic to provide an additional increment for the "waiting compensation factor" sub-item of the task, such as adding an additional base of 0.1 on the original linear growth, thereby accelerating its priority recovery and avoiding task starvation.
[0169] The intelligent printing execution module is the terminal actuator for the system's interaction with the physical world.
[0170] Please refer to the attached document. Figure 5 This module drives the actual printing device to complete the job based on the ordered task queue output by the dynamic priority scheduling module.
[0171] The module contains a task parsing unit, a device driver unit, and a feedback unit.
[0172] The task parsing unit is responsible for converting logical printing tasks in the queue into executable physical instructions. It retrieves the first task from the head of the queue, whose data packet contains information such as the incorrect questions, student identifiers, and a preset printing format template index.
[0173] The task parsing unit first loads the corresponding print template from the template library based on the student identifier and the format template index. This template defines the page layout, header and footer, font size, and other styles.
[0174] Next, the unit combines the incorrect answers with the template to generate a standard page description file, which can be in PDF, PostScript, or a page description language specific to a particular printer.
[0175] At the same time, the unit will check whether the content contains the "merge print" instruction issued by the resource optimization module. If it does, it will merge the multiple contents according to the instruction requirements.
[0176] The device driver unit is a component that communicates directly with the printer hardware.
[0177] It receives the page description file generated by the task parsing unit and converts it into a specific sequence of printer control instructions, such as PCL or PJL commands, based on the target printer's model and communication protocol.
[0178] The device driver unit establishes a connection with the printer via a wired or wireless network and continuously monitors the printer's real-time status.
[0179] The monitored parameters include paper balance, toner or ink balance for each color, current online status, paper jam status, and paper shortage status.
[0180] The device driver unit maintains the printer status table and periodically, for example, polls all registered printers every 30 seconds.
[0181] Before sending a print job, the device driver unit checks the status of the target printer to ensure it is ready.
[0182] The feedback unit is responsible for completing the closed-loop management of printing tasks.
[0183] Once the device driver unit confirms that the print job has been successfully received by the printer and has started execution, the feedback unit will start the monitoring timer.
[0184] If the printer returns a "job completed" status code within the preset job timeout period, such as 3 minutes, the feedback unit determines that the task was successfully executed.
[0185] Subsequently, it sends a confirmation message to the system's core database, updates the status field of the print job to "completed", and records the precise completion timestamp.
[0186] These completed data will become historical data for analysis by the resource optimization module.
[0187] If a malfunction occurs during printing, such as the printer going offline, running out of paper, or running out of toner, the device driver unit will immediately capture the error code and interrupt the current job.
[0188] Upon receiving an interrupt signal, the feedback unit will not only update the task status to "printing failed", but will also immediately send an interrupt signal containing the task identifier and error type to the dynamic priority scheduling module.
[0189] Upon receiving this signal, the dynamic priority scheduling module will trigger conflict resolution logic, which may reschedule the failed task to another available printer or introduce a fault recovery factor into its priority calculation to arrange a retry.
[0190] The resource optimization module runs in the system background and is dedicated to improving overall resource utilization efficiency.
[0191] This module works on two levels: load balancing and content merging.
[0192] At the load balancing level, the resource optimization module analyzes historical printing task data fed back by the intelligent printing execution module, including task volume, execution time, and printer used.
[0193] Through data mining, the module identifies peak and off-peak periods for printing workloads. For example, the peak is from 9:00 to 11:00 AM, the secondary peak is from 1:00 to 3:00 PM, and the off-peak is in the evening.
[0194] During off-peak periods, the resource optimization module sends scheduling suggestions to the dynamic priority scheduling module, proactively triggering a batch of non-urgent but high-value-for-all common error sets to be printed.
[0195] These tasks may come from pre-designed exercises by teachers to reinforce common mistakes made in the class.
[0196] Through this staggered scheduling, the system effectively balances the load on printing equipment, avoids congestion during peak hours, and extends the lifespan of the equipment.
[0197] At the content merging level, the resource optimization module and the error value assessment module work together in a deep collaboration.
[0198] It periodically scans the queue of tasks to be printed, utilizing the globally unique identifiers and global teaching value scores provided by the error assessment module.
[0199] When multiple students' print jobs are found to contain the same high-value wrong answer, for example, when the global teaching value score is greater than 0.8, the resource optimization module will initiate the merging logic.
[0200] The merge logic first determines whether these tasks meet the merge conditions, such as having the same target printer or compatible printing formats.
[0201] If the conditions are met, the module will generate a "Merge Print" instruction, which includes a list of student identifiers to be merged and detailed parameters for the merged layout. For example, the original questions will be printed once, and the relevant student identifiers and their individual error analysis summaries will be listed in a table below.
[0202] The resource optimization module attaches this instruction to the relevant task and sends it together to the task parsing unit of the intelligent printing execution module.
[0203] In this way, the system significantly reduces paper consumption for repetitive content while ensuring that each student receives targeted feedback. It is particularly suitable for the centralized processing of common mistakes in a class, demonstrating both economic benefits and environmental value.
[0204] The entire system's operation process forms an efficient and adaptive closed loop.
[0205] The learning data collection module serves as a continuous source of data input.
[0206] The error assessment module and the student status perception module serve as a dual-core analysis engine, processing data streams in parallel and continuously generating value scores and status profiles.
[0207] The dynamic priority scheduling module acts as the decision-making brain, periodically acquiring these analysis results, performing complex composite model calculations, and outputting a dynamically sorted task queue.
[0208] The intelligent printing execution module acts as the execution arm, strictly following the queue order to drive the hardware to complete the printing and feeding back the results to form a closed loop.
[0209] The resource optimization module acts like a background optimizer, adjusting load and resources from a global perspective.
[0210] All modules are decoupled and coordinated through message queues or database sharing of state, ensuring that the system can run stably 24 / 7 and realize the accurate and efficient transformation of the value of in-depth mining of learning big data into physical teaching links.
[0211] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.
[0212] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
Claims
1. A big data-based error analysis and precise printing management system for student learning, characterized in that: include: The learning data collection module is used to continuously collect and integrate student behavior log data and assessment result data from multiple learning platforms to build a raw learning database in a unified format. The error value assessment module is used to perform multi-dimensional quantitative analysis on the error data in the original learning database, and calculate the global teaching value score and individual urgency score for each error. The student status perception module is used to analyze and update each student's learning progress profile and current learning focus status in real time based on the original learning information database. The dynamic priority scheduling module is used to receive output data from the wrong question value assessment module and the student status perception module, generate a dynamic comprehensive priority score for each wrong question task to be printed according to the preset composite priority calculation model, and reconstruct the printing task queue accordingly. The intelligent printing execution module is used to control the connected physical printing device to execute printing jobs according to the task queue order reconstructed by the dynamic priority scheduling module, and to provide feedback on the printing status. The composite priority calculation model of the dynamic priority scheduling module is defined as follows: The overall priority score is equal to the product of the base value score, the dynamic adjustment factor, and the state decay factor. The basic value score is a linear combination of the overall teaching value score of the incorrect questions and the individual urgency score. The dynamic adjustment factor is calculated as follows: For tasks marked as frequently occurring common mistakes with an error rate greater than 30%, an acceleration factor greater than 1 is assigned. For tasks in the print task queue whose waiting time exceeds a preset threshold, a waiting compensation factor that increases linearly with time is assigned; the state decay factor is calculated as follows: when the student state perception module determines that the student is in a low-focus state, a decay coefficient of less than 1 is assigned to all the student's pending print tasks. When the student status perception module indicates that a student is about to enter the learning stage of new knowledge points, it assigns a promotion coefficient greater than 1 to the old knowledge weak points and wrong questions that are strongly related to the new knowledge points. The dynamic priority scheduling module also includes a queue reconstruction unit and a conflict resolution unit; The queue reconstruction unit sorts all pending print tasks in descending order according to the latest comprehensive priority score at a fixed period to generate a new print task queue. The conflict resolution unit is used to handle resource conflicts that may occur during the task queue reconstruction process. Its logic is as follows: when multiple high-priority tasks compete for the same printing device at the same time, the task that is associated with a student in a highly focused state and is a frequently asked common wrong question is executed first. For tasks that are temporarily postponed due to conflict, the waiting compensation factor in their dynamic adjustment factor will receive an additional increment.
2. The error mining and precise printing management system for big data learning information according to claim 1, characterized in that, The error assessment module includes a common value analysis unit and an individual urgency analysis unit; The common value analysis unit is used to calculate the global teaching value score of incorrect questions, and the calculation process is as follows: The error rate of the incorrect question in the target student group is statistically analyzed, and questions with an error rate greater than a preset threshold are marked as high-frequency common errors. The weight of the knowledge points associated with the incorrect question in the preset knowledge graph is analyzed. The thinking levels involved in the solution process of the incorrect question are evaluated. The thinking levels are divided into five levels: memory, understanding, application, analysis and synthesis, and each level corresponds to a different coefficient. The overall teaching value score is calculated by weighting the error rate, the weight of knowledge points, and the thinking level coefficient. The individual urgency analysis unit is used to calculate the individual urgency score for incorrect answers, and the calculation process is as follows: Track the number of times the student made mistakes on the same knowledge point throughout their history; Calculate the time interval between the last occurrence of the incorrect question and its occurrence; combine this with the student's recent accuracy trend in related exercises on this knowledge point.
3. The error mining and precise printing management system for big data learning according to claim 2, characterized in that, The student status perception module includes a learning progress tracking unit and a focus assessment unit; The learning progress tracking unit is used to maintain and update the learning progress profile of each student. The learning progress profile includes a list of completed knowledge points, a mastery score for each knowledge point, an identifier of the knowledge point currently being learned, and a preset learning plan timetable. The focus assessment unit is used to assess the student's current learning focus in real time. The assessment is based on the real-time behavior log data provided by the learning data acquisition module, which specifically includes the student's effective learning time, page switching frequency, and response time for interactive questions in the current learning session. The focus assessment unit compares the above behavioral data with a preset focus model and outputs a quantified focus index. When the focus index is less than 60%, it is determined to be a low focus state.
4. The error mining and precise printing management system for big data learning information according to claim 1, characterized in that, The intelligent printing execution module includes a task parsing unit, a device driving unit, and a feedback unit; The task parsing unit is used to extract the first task from the task queue output by the dynamic priority scheduling module, and convert the incorrect questions, format templates and student identification information contained therein into a standard print job description file. The device driver unit is used to generate specific printer control instructions based on the print job description file, and monitor the real-time status of the connected physical printing device, including paper balance, toner balance, and online status. The feedback unit is used to send confirmation information to the system core database after the print job is completed, update the status of the print job to be completed and record the completion timestamp; If a fault occurs during printing, an interrupt signal is immediately sent to the dynamic priority scheduling module.
5. The error mining and precise printing management system for big data learning according to claim 4, characterized in that, It also includes a resource optimization module; The resource optimization module is used to analyze historical printing task data, identify peak and off-peak periods of printing volume, and automatically schedule non-urgent, large-volume printing tasks of common error sets during off-peak periods to balance the equipment load.
6. The error mining and precise printing management system for big data learning information according to claim 5, characterized in that, The resource optimization module is also used to intelligently format and merge content based on the output of the error value assessment module; When multiple students' incorrect answers involve the same high-value question, they should be combined and printed on one sheet of paper, with clear labels indicating which student's answer is correct.
7. The error mining and precise printing management system for big data learning information according to claim 6, characterized in that, The learning data acquisition module establishes a secure connection with multiple third-party learning platforms through a set of standardized application programming interfaces, and the collected data includes student behavior log data and assessment result data. The learning data acquisition module has a built-in data cleaning and standardization subunit, which is used to convert the acquired raw data into a unified internal representation format and write it into the raw learning database.
8. The error mining and precise printing management system for big data learning information according to claim 1, characterized in that, The system's operation flow is as follows: The learning data acquisition module continuously inputs data; The error assessment module and the student status perception module process data in parallel and output assessment results. The dynamic priority scheduling module periodically obtains the above results, calculates the comprehensive priority score, and reconstructs the task queue; the intelligent printing execution module executes printing according to the queue order. The entire process forms a closed-loop, adaptive, precision printing management pipeline.
Citation Information
Patent Citations
Method and system for printing wrong questions based on data analysis
CN121209806A
Printer
JP2005103924A
