Correction timeline determination and handwriting analysis method, server, medium and product

Abstract

The invention discloses a correction timeline determination and handwriting analysis method, a server, a medium and a product. The method comprises the following steps: detecting whether each student has a class correction moment and / or a student correction moment; if the class correction moment and the student correction moment exist, setting the correction timeline as the student correction moment; if the class correction moment exists and the student correction moment does not exist, setting the correction timeline as the class correction moment; if the class correction moment and the student correction moment do not exist, obtaining the first submission moment of the question; detecting whether there is a new handwriting of the already-made question within a preset duration after the first submission moment of the question; if not, determining a question submission cut-off moment as a correction timeline according to a question first submission moment and a preset duration; and marking the handwriting of which the writing time is later than the correction time line as corrected handwriting, and determining corrected statistical data according to the corrected handwriting. By implementing the method, correction statistical deviation is reduced, and the accuracy of teaching quality analysis is improved.

Description

Technical Field

[0001] The present application relates to the field of information technology, and in particular to a correction timeline determination and handwriting analysis method, server, medium and product. Background Art

[0002] In the teaching process, teachers often need to mark students' homework, and students also need to make corrections based on the teacher's comments. The process of marking and correcting homework can not only help teachers understand the students' mastery of the course content, but also help students identify their weak knowledge points.

[0003] At present, most schools use electronic homework systems to manage the marking and correction of homework. The electronic homework system can track the completion of homework based on the student's homework submission timestamp. When the teacher marks the homework, the electronic homework system obtains the homework correction timestamp and presents the teacher's correction comments to the students simultaneously. Students can view the teacher's correction comments in the electronic homework system and continue to write and correct the original homework based on the original homework. The electronic homework system distinguishes the original answer content and the corrected content by comparing the order of the homework submission timestamp and the homework correction timestamp.

[0004] In actual applications, due to the differences in the progress and rhythm of teachers in marking homework, some students may start to supplement and modify their homework before the teacher marks it, and some students may submit their homework multiple times when completing it, making it difficult for the electronic homework system to accurately define which content falls into the category of correction, thus causing correction statistical deviations and affecting the accuracy of teaching quality analysis. Summary of the invention

[0005] This application provides a correction timeline determination and handwriting analysis method, server, medium and product for reducing correction statistical deviation and improving the accuracy of teaching quality analysis.

[0006] In a first aspect, the present application provides a correction timeline determination and handwriting analysis method, which is applied to a server. The method includes: obtaining teacher correction records and student question handwriting data, wherein the teacher correction records include teacher correction time information for student questions, and the student question handwriting data includes student handwriting and writing time for the questions; based on the teacher correction records, detecting whether there is a class correction time and / or student correction time for each student, wherein the class correction time refers to the time when the teacher corrects the first question of the first student, and the student correction time refers to the time when the teacher corrects the first question of each student; if the first student has a class correction time and a student correction time, then setting the correction timeline of the first student is the student correction time; if the second student has a class correction time but no student correction time, the second student's correction timeline is set to the class correction time; if the third student does not have a class correction time and a student correction time, the third student's first question submission time is obtained; based on the student's question handwriting data, detect whether there is new handwriting of the done question within the preset time after the first question submission time; if not, determine the question submission deadline as the third student's correction timeline based on the first question submission time and the preset time; mark the handwriting in the student's question handwriting data that is written later than the correction timeline as correction handwriting, and determine the correction statistical data based on the correction handwriting.

[0007] By adopting the above technical solution, the server adopts different correction timeline determination strategies for the three situations: the existence of class correction time and student correction time, the existence of only class correction time but no student correction time, and the existence of neither class correction time nor student correction time. The correction timeline is defined in combination with the handwriting changes within a preset time after the first submission of the question, which effectively solves the limitation of traditional electronic homework systems that only rely on homework submission timestamps and homework correction timestamps to distinguish between original answer content and corrected content. This multi-dimensional correction timeline determination method can more accurately identify students' correction behavior, reduce correction statistical deviations caused by inconsistent teacher correction progress or multiple submissions by students, and improve the accuracy of correction behavior analysis.

[0008] In combination with some embodiments of the first aspect, in some embodiments, after the step of detecting whether there is new handwriting of the completed questions within a preset time period after the first submission of the questions based on the student's handwriting data, the method also includes: if there is, setting the time when the new handwriting appears as the correction timeline of the third student.

[0009] By adopting the above technical solution, in the absence of class correction time and student correction time, the server detects whether there is new handwriting within the preset time after the question is first submitted. If so, the server sets the time when the new handwriting appears as the correction timeline, taking into account the situation that students start to modify the questions before the teacher corrects them. This method of dynamically adjusting the correction timeline based on actual handwriting changes can more realistically reflect the student's correction process and avoid misjudging self-modified content as corrected content. At the same time, this method also takes into account the homework completion rhythm of different students, making the correction statistics more in line with the actual situation and providing more reliable data support for subsequent teaching quality analysis.

[0010] In combination with some embodiments of the first aspect, in some embodiments, after the step of obtaining the time when the third student first submitted the question if there is no class correction time and student correction time for the third student, the method also includes: if there is a class correction time and a student correction time in the teacher correction record, setting the third student's correction timeline to the student correction time; if there is a class correction time and no student correction time in the teacher correction record, setting the third student's correction timeline to the class correction time.

[0011] By adopting the above technical solution, in the absence of class correction time and student correction time, the server supplements the teacher correction record to check whether the class correction time and student correction time are subsequently generated, and dynamically updates the correction timeline according to the class correction time and student correction time. This backward checking mechanism ensures that even if the teacher delays correction, the server can still accurately capture the real correction timeline, optimizes the integrity of the correction timeline determination, avoids correction identification deviation caused by teacher correction delay, and improves the reliability and accuracy of correction analysis.

[0012] In combination with some embodiments of the first aspect, in some embodiments, after marking the handwriting in the student's question handwriting data that is written later than the correction timeline as corrected handwriting, and determining the correction statistical data based on the corrected handwriting, the method also includes: obtaining the student's original handwriting before the correction timeline; comparing the handwriting features of the corrected handwriting with the original handwriting to obtain a comparison result, the handwriting features include stroke sequence, stroke strength, writing speed and handwriting continuity; determining a correction quality score based on the comparison result, the higher the correction quality score, the higher the correction quality; classifying the student's correction behavior according to the correction quality score to obtain a correction behavior classification result, the correction behavior classification result includes careful correction, mechanical correction and perfunctory correction.

[0013] By adopting the above technical solution, the server compares the stroke order, stroke strength, writing speed and handwriting coherence of the corrected handwriting with the original handwriting, and realizes an in-depth evaluation of the correction quality. This analysis method based on handwriting features can effectively identify the degree of cognitive involvement of students during correction, so as to help teachers understand students' correction attitudes and thus more objectively evaluate students' correction effects.

[0014] In combination with some embodiments of the first aspect, in some embodiments, handwriting in the student's question handwriting data whose writing time is later than the correction timeline is marked as corrected handwriting, and correction statistical data is determined based on the corrected handwriting, specifically including: determining the handwriting in the student's question handwriting data whose writing time is later than the correction timeline as corrected handwriting; counting the questions corresponding to the corrected handwriting to determine the number of corrected questions; classifying the questions corresponding to the corrected handwriting according to preset dimensions to determine the number of corrections in each category; performing time distribution statistics based on the correction time corresponding to the corrected handwriting to determine the correction time period; determining the correction statistical data based on the number of corrected questions, the number of corrections in each category and the correction time period.

[0015] By adopting the above technical solution, the server conducts multi-dimensional statistical analysis of the handwriting corrections, including the number of correction questions, the number of corrections in each category, and the correction period, and builds a comprehensive correction behavior analysis framework. This multi-angle statistical method provides data support for teachers to optimize teaching strategies and helps improve teaching effectiveness.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after the questions corresponding to the corrected handwriting are classified according to preset dimensions and the number of corrections in each category is determined, the method also includes: determining the students' weak knowledge areas based on the number of corrections in each category; determining the students' weak knowledge graph based on the correlation between the weak knowledge areas; generating the students' weak knowledge courses and personalized exercises based on the weak knowledge graph, the personalized exercises including exercises corresponding to different weak knowledge points.

[0017] By adopting the above technical solution, the server determines the students' weak knowledge areas based on the number of corrections in each category, and then constructs the students' weak knowledge graph. This method of constructing a weak knowledge graph based on correction data can not only deeply understand the students' knowledge structure defects through the correlation between weak knowledge points, but also automatically generate weak knowledge courses and personalized exercises to provide accurate learning support for each student, realizing a closed loop from problem discovery to solution, and effectively improving the pertinence and effectiveness of teaching.

[0018] In combination with some embodiments of the first aspect, in some embodiments, weak knowledge courses and personalized exercises for students are generated based on the weak knowledge graph, and the personalized exercises include exercises corresponding to different weak knowledge points respectively, specifically including: determining a weak knowledge point association strength matrix based on the weak knowledge graph, and the matrix element values ​​in the weak knowledge point association strength matrix are used to represent the degree of association between different weak knowledge points; based on the weak knowledge point association strength matrix, hierarchical clustering is performed on the weak knowledge points to obtain a weak knowledge point hierarchy tree; based on the weak knowledge point hierarchy tree, the learning priority of the weak knowledge points is determined; based on the weak knowledge point learning priority, a learning path and multiple stage learning objectives are generated; according to the learning path, weak knowledge courses and personalized exercises that match the learning objectives of each stage are screened from a preset knowledge question bank.

[0019] By adopting the above technical solutions, the server realizes the scientific organization of knowledge point learning by constructing the weak knowledge point association strength matrix and the weak knowledge point hierarchy tree. The server identifies the hierarchical relationship between weak knowledge points and determines the optimal learning path and stage learning goals based on this, ensuring the comprehensiveness and scientificity of remedial learning. The server accurately matches the stage learning goals with the weak knowledge courses and personalized exercises in the preset knowledge question bank, thereby providing students with the most suitable learning resources, avoiding the redundancy of learning resources, ensuring the pertinence and efficiency of the learning process, and achieving personalized learning goals.

[0020] In a second aspect, an embodiment of the present application provides a server, comprising: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the server to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product comprising instructions, which, when executed on a server, enables the server to execute the method described in the first aspect and any possible implementation of the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, comprising instructions, which, when executed on a server, enable the server to execute the method described in the first aspect and any possible implementation of the first aspect.

[0023] It is understandable that the server provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiment of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By adopting the above technical solution, the server adopts different correction timeline determination strategies for the three situations: the existence of class correction time and student correction time, the existence of only class correction time but no student correction time, and the existence of neither class correction time nor student correction time. The correction timeline is defined in combination with the handwriting changes within a preset time after the first submission of the question, which effectively solves the limitation of traditional electronic homework systems that only rely on homework submission timestamps and homework correction timestamps to distinguish between original answer content and corrected content. This multi-dimensional correction timeline determination method can more accurately identify students' correction behavior, reduce correction statistical deviations caused by inconsistent teacher correction progress or multiple submissions by students, and improve the accuracy of correction behavior analysis.

[0025] 2. By adopting the above technical solution, the server compares the stroke order, stroke strength, writing speed and handwriting coherence of the corrected handwriting with the original handwriting, and realizes an in-depth evaluation of the correction quality. This analysis method based on handwriting features can effectively identify the degree of cognitive involvement of students during correction, so as to help teachers understand students' correction attitudes and thus more objectively evaluate students' correction effects.

[0026] 3. By adopting the above technical solutions, the server realizes the scientific organization of knowledge point learning by constructing the weak knowledge point association strength matrix and the weak knowledge point hierarchy tree. The server identifies the hierarchical relationship between weak knowledge points and determines the optimal learning path and stage learning goals based on this, ensuring the comprehensiveness and scientific nature of remedial learning. The server accurately matches the stage learning goals with the weak knowledge courses and personalized exercises in the preset knowledge question bank, thereby providing students with the most suitable learning resources, avoiding the redundancy of learning resources, ensuring the pertinence and efficiency of the learning process, and achieving personalized learning goals. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flowchart of the correction timeline determination and handwriting analysis method in the embodiment of the present application; Figure 2 is another flowchart of the correction timeline determination and handwriting analysis method in the embodiment of the present application; Figure 3It is a schematic diagram of a physical device structure of a server in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification of the present application, the singular expressions "one", "a kind of", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more of the listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0030] The following is a description of the process of the method provided by this implementation. Figure 1 , which is a flow chart of the correction timeline determination and handwriting analysis method in the embodiment of the present application.

[0031] S101, obtaining teacher's correction record and student's question handwriting data, wherein the teacher's correction record includes the teacher's correction time information for the student's question, and the student's question handwriting data includes the student's handwriting and writing time for the question; Among them, the teacher's correction record refers to the data record generated when the teacher corrects the student's questions, which is used to indicate the teacher's correction time information and correction content information for the student's questions. The correction time information refers to the specific time when the teacher corrects each question; the student's question handwriting data refers to the handwriting trajectory data generated when the student completes the question, including the writing handwriting and the writing time; the writing handwriting refers to the trajectory, pressure, inclination and other characteristic data when the student writes on the question; the writing time refers to the specific time point when each handwriting is generated, accurate to the millisecond level.

[0032] Specifically, first, the server connects to the teacher's correction database to obtain the teacher's correction record containing correction timestamp, correction content, correction object, etc. At the same time, the server connects to the student homework database to extract all the handwriting produced by the students on the questions, including the spatial coordinate sequence of each stroke, pen pressure data, writing speed and corresponding writing time, including the handwriting and writing time of the students on the questions.

[0033] S102, based on the teacher's correction record, detecting whether each student has a class correction time and / or a student correction time, the class correction time refers to the time when the teacher corrects the first question of the first student, and the student correction time refers to the time when the teacher corrects the first question of each student; Among them, the class correction time refers to the starting time when the teacher starts to correct the class assignments, specifically the time when the teacher corrects the first question of the first student; the student correction time refers to the starting time when the teacher starts to correct a specific student's assignments, specifically the time when the teacher corrects the first question of the student.

[0034] Specifically, first, the server determines the class grading time and / or the student grading time from the teacher's grading record. If there is a class grading time and a part of the students' grading time, it means that the teacher has started grading the class assignments and has graded the assignments of some students. If there is a class grading time and a part of the students' grading time, it means that the teacher has started grading the class assignments and has graded the assignments of all students. If there is no class grading time and a part of the students' grading time, it means that the teacher has not started grading the class assignments. Then, the server searches for each student whether there is a class grading time and / or a student grading time. The above method takes into account different modes of teacher grading assignments, including both unified grading and decentralized grading.

[0035] S103: If the first student has a class correction time and a student correction time, then the correction timeline of the first student is set to the student correction time; Among them, the first student refers to the student who is corrected first in the teacher's correction sequence; the correction timeline refers to the time dividing point used to distinguish the original answer content and the corrected content.

[0036] Specifically, first, the server verifies the validity of the first student's class correction time and the student correction time, ensuring that both time points exist and are meaningful. Then, the server uses the first student's student correction time as the first student's correction timeline. The reason for choosing the student correction time instead of the class correction time is that the student correction time more accurately reflects the actual time point when the first student receives the correction feedback, and can more accurately define the start time of the first student's correction behavior.

[0037] S104, if the second student has a class correction time and does not have a student correction time, setting the second student's correction timeline to the class correction time; Among them, the second student refers to a student who has a class correction time but no independent student correction time in the teacher's correction sequence, that is, a student who has not been corrected yet; the existence of a class correction time means that the class where the second student is located has started correction, that is, the teacher has corrected the first question of the first student in the class; the absence of a student correction time means that the teacher has not yet corrected the second student's homework.

[0038] Specifically, first, the server verifies the existence and validity of the class correction time, confirming that the class correction time is indeed earlier than the current system time and within a reasonable time range. Then, the server uses the second student's class correction time as the second student's correction timeline. This processing method takes into account the different correction strategies that teachers may adopt when correcting homework. For example, teachers may give unified comments in class but do not correct each student's homework one by one.

[0039] S105. If the third student does not have a class review time and a student review time, obtain the time when the third student first submits the question; Among them, the third student refers to a student who has neither class grading time nor student grading time; no grading time means that the teacher has not performed any grading operations on the homework of the third student's class or the third student personally; the first submission time of the question refers to the time point when the third student first submitted the completed question.

[0040] Specifically, first, the server confirms that the third student and the class the third student is in do not have any correction records. Then, the server queries the third student's homework submission records and finds the earliest submission time as the first submission time of the question. To ensure data accuracy, the server will verify the integrity of the homework submission records and exclude abnormal submission records caused by network problems or system failures.

[0041] S106, based on the student's question handwriting data, detecting whether there is new handwriting of the completed question within a preset time period after the question is first submitted; Among them, the preset time duration refers to a predefined period of time used to limit the time range for detecting new handwriting; new handwriting refers to the additional writing content that appears after the student submits the question; and the completed questions refer to the questions that the student has completed the first answer.

[0042] Specifically, first, the server determines the start and end time points of the preset time, which is usually calculated from the time when the question is first submitted. The preset time may be 30 minutes or other pre-defined reasonable time values. Then, the server extracts all handwriting data within the preset time after the first submission of the question from the student's handwriting data, and compares it with the handwriting at the time of the first submission. The server identifies whether there is new handwriting by comparing the spatial position, timestamp and content features of the handwriting.

[0043] S107, if it does not exist, determine the deadline for question submission as the third student's revision timeline based on the first submission time of the question and the preset duration; Specifically, if the server does not detect any additional writing or modification operations within the preset time after the first submission of the question, it is deemed that there is no new handwriting of the completed question within the preset time after the first submission of the question. The server uses the first submission of the question as the reference time, plus the preset time (for example, 30 minutes), to calculate the deadline for question submission. During the calculation process, the server will consider time factors such as time zone conversion and daylight saving time to ensure the accuracy of time calculation. The calculated deadline for question submission will be set as the correction timeline of the third student.

[0044] S108, marking the handwriting of the student's question handwriting data whose writing time is later than the correction time line as the correction handwriting, and determining the correction statistical data according to the correction handwriting.

[0045] Specifically, first, the server obtains all the student question handwriting data, and compares the writing time of each handwriting in the student question handwriting data with the determined correction timeline. For handwriting that is written later than the correction timeline, the server will add correction handwriting for it and record relevant context information, such as the subject, knowledge point classification, etc. Then, the server performs multi-dimensional statistical analysis on these marked correction handwritings, including but not limited to: counting the number of corrections by subject, counting the correction distribution by knowledge point category, calculating the correction time distribution characteristics, analyzing the correctness of the correction content, etc., to obtain correction statistical data.

[0046] By adopting the above technical solution, the server adopts different correction timeline determination strategies for the three situations: the existence of class correction time and student correction time, the existence of only class correction time but no student correction time, and the existence of neither class correction time nor student correction time. The correction timeline is defined in combination with the handwriting changes within a preset time after the first submission of the question, which effectively solves the limitation of traditional electronic homework systems that only rely on homework submission timestamps and homework correction timestamps to distinguish between original answer content and corrected content. This multi-dimensional correction timeline determination method can more accurately identify students' correction behavior, reduce correction statistical deviations caused by inconsistent teacher correction progress or multiple submissions by students, and improve the accuracy of correction behavior analysis.

[0047] The following is a more detailed description of the process of the method provided by this implementation. Figure 2 , is another flow chart of the correction timeline determination and handwriting analysis method in the embodiment of the present application.

[0048] After step S106, the following steps may be performed: S201, if it exists, setting the appearance time of the newly added handwriting as the correction timeline of the third student; Specifically, if the server detects additional writing or modification within a preset time after the first submission of the question, it is considered that there is additional handwriting of the completed question within the preset time after the first submission of the question. The server sets the earliest appearance time of the additional handwriting as the third student's correction timeline.

[0049] S202: If the teacher's correction record contains the class's correction time and the student's correction time, then the correction timeline of the third student is set to the student's correction time; Specifically, the server checks in real time whether there is a class correction time and / or a student correction time in the teacher correction record. If there is a class correction time and a student correction time in the teacher correction record, the server selects the student correction time as the correction timeline of the third student.

[0050] S203: If the teacher's correction record contains the class correction time but does not contain the student's correction time, then the correction timeline of the third student is set to the class correction time; Specifically, if there is only the class correction time in the teacher's correction record, the server sets the class correction time as the third student's correction timeline, and no longer considers the third student's actual behavior, and the teacher's correction situation prevails.

[0051] S204, determining the handwriting in the student's question handwriting data whose writing time is later than the correction time line as the correction handwriting; Specifically, the server obtains all student question handwriting data, compares the writing time of each handwriting in the student question handwriting data with the determined correction timeline, and marks the handwriting whose writing time is later than the correction timeline as the correction handwriting.

[0052] S205, counting the questions corresponding to the corrected handwriting to determine the number of corrected questions; Specifically, the server counts the questions corrected by the students and counts them to obtain the number of corrected questions. During the counting process, the server will exclude duplicate counting to ensure that even if the same question has multiple corrections, it is only counted once. For corrections across questions, the server will accurately classify them according to the spatial position of the handwriting and the contextual relationship.

[0053] S206, classifying the questions corresponding to the corrected handwriting according to a preset dimension, and determining the number of corrections in each category; Among them, preset dimensions refer to pre-defined classification standards, such as knowledge points, question types, difficulty, etc.; classification refers to the process of classifying questions containing correction handwriting according to preset dimensions; categories refer to specific groupings after division according to preset dimensions; the number of corrections refers to the frequency of correction behavior in each category.

[0054] Specifically, first, the server classifies the questions containing the correction handwriting in multiple dimensions according to preset dimensions (such as classification by knowledge points, classification by question types, classification by difficulty level, etc.). For each classification dimension, the server counts the number of corrections in the category and records detailed correction feature information.

[0055] S207, performing time distribution statistics according to the correction time corresponding to the correction handwriting to determine the correction period; Among them, time distribution statistics refers to the analysis of the distribution characteristics of correction behavior in the time dimension; the correction period refers to the time interval when correction behaviors occur in a concentrated manner.

[0056] Specifically, first, the server collects the correction times corresponding to all the correction handwriting and arranges the correction times in the order of the day. Then, the server uses the time series analysis method to identify the time distribution characteristics of the correction behavior, such as the start time, end time, and high-frequency time period of the correction. The server will aggregate and analyze the time distribution data according to different time granularities (such as hours, days, weeks, etc.) to find out the regular characteristics of the correction behavior.

[0057] S208, determining the student's weak knowledge areas based on the number of corrections in each category; Specifically, first, the server normalizes the number of corrections for each category to eliminate the impact of the difference in the number of questions in different categories. Then, the server sets a threshold for the number of corrections (such as 1.5 times the average number of corrections) and marks the categories that exceed the threshold as weak knowledge areas, where weak knowledge areas refer to the knowledge ranges with high correction frequencies by students. During the judgment process, the server will consider factors such as the difficulty coefficient of the question and the distribution of correction time, and obtain a more accurate assessment of the degree of weakness through weighted calculation.

[0058] S209, determining the student's weak knowledge graph according to the association relationship between the weak knowledge areas; Among them, weak knowledge areas refer to the range of knowledge that students have difficulty in learning; association relationships refer to the logical connections and dependencies between different knowledge areas; and weak knowledge graphs are used to represent the association status between weak knowledge areas displayed in a network structure.

[0059] Specifically, first, the server builds a network of association relationships between weak knowledge areas based on the preset knowledge system structure. The server will consider multiple association types such as pre-relationship, parallel relationship, and inclusion relationship between weak knowledge points to build a weak knowledge graph with direction and weight. During the construction process, the server will combine the preset knowledge point association rules and refer to the student's actual correction behavior pattern to dynamically adjust the association strength.

[0060] S210, determining a weak knowledge point association strength matrix according to the weak knowledge graph, wherein the matrix element values ​​in the weak knowledge point association strength matrix are used to represent the association degree between different weak knowledge points; Among them, the weak knowledge graph refers to the network structure of the association relationship between weak knowledge points; the weak knowledge point association strength matrix refers to a two-dimensional array that expresses the degree of association between weak knowledge points in numerical form.

[0061] Specifically, the server converts the association relationship in the weak knowledge graph into a matrix form, where the rows and columns of the matrix correspond to different weak knowledge points. The server calculates the association strength between weak knowledge points through multiple dimensions, including direct dependency, co-occurrence frequency, and correction sequence relationship between weak knowledge points. The calculated association strength value will be standardized to the [0, 1] interval, where 0 indicates no association and 1 indicates the strongest association. For asymmetric association relationships, the server will generate an asymmetric matrix to record the bidirectional association strength between weak knowledge points.

[0062] S211, based on the weak knowledge point association strength matrix, hierarchical clustering is performed on the weak knowledge points to obtain a weak knowledge point hierarchy tree; Among them, hierarchical clustering refers to the process of combining similar weak knowledge points layer by layer to form a hierarchical structure; the weak knowledge point hierarchy tree refers to a model that displays the clustering relationship of weak knowledge points in a tree structure.

[0063] Specifically, the server uses a hierarchical clustering algorithm (such as UPGMA or Ward's method) to perform cluster analysis on weak knowledge points. First, the server converts the association strength into a distance metric and calculates the similarity between weak knowledge points. Then, the server merges the closest weak knowledge points or weak knowledge point groups in an iterative manner to form higher-level categories. During each merging process, the server updates the inter-class distance and records the merged hierarchical relationship. The resulting weak knowledge point hierarchy tree will clearly display the hierarchical organizational structure between weak knowledge points, including trunk knowledge points, branch knowledge points, and the hierarchical relationship between them.

[0064] S212, determining the learning priority of the weak knowledge points according to the weak knowledge point hierarchy tree; Specifically, first, the server analyzes the structural characteristics of the weak knowledge point hierarchy tree, including the hierarchical position, branch relationship and node importance of the weak knowledge points. Then, the server calculates the priority score of each weak knowledge point through a comprehensive evaluation algorithm, that is, the learning priority of the weak knowledge point, where the learning priority of the weak knowledge point refers to the importance and order of different weak knowledge points in the learning sequence, and the factors considered include: the basic nature of the weak knowledge point (the more basic, the higher the priority), the pre-dependency relationship (the priority is increased by other weak knowledge points with high dependency), the correction frequency (the priority of weak knowledge points with high frequency correction is increased), the learning difficulty (the difficulty is matched with the current level of the students) and so on. The server will assign a priority score of 0-100 to each weak knowledge point, and divide the weak knowledge points into several priority levels according to the priority score, such as "highest priority", "priority", "second priority", etc.

[0065] S213, generating a learning path and multiple stage learning objectives based on the learning priority of the weak knowledge point; Among them, the learning path refers to the recommended sequence and advanced route for learning weak knowledge points; the stage learning goal refers to the specific ability requirements that need to be achieved at different stages of the learning process.

[0066] Specifically, first, the server constructs the optimal learning path based on the learning priority of weak knowledge points and the dependencies between weak knowledge points. The server divides the learning process into multiple stages, and sets learning goals for each stage, including knowledge mastery, problem-solving ability requirements, application ability level, etc. When formulating a learning path, the server will consider the following factors: learning time allocation (estimating the required learning time based on the difficulty of the knowledge points and the student's ability level), knowledge point connection (ensuring a smooth transition between adjacent knowledge points), stage evaluation points (setting appropriate checkpoints to evaluate learning effects), and alternative paths (providing alternatives for possible learning difficulties).

[0067] S214, based on the learning path, selecting weak knowledge courses and personalized exercises that match the learning objectives of each stage from the preset knowledge question bank; Among them, the preset knowledge question bank refers to a collection of pre-stored teaching resources; weak knowledge courses refer to special learning content targeting specific weak knowledge points; and personalized exercises refer to practice questions customized according to students' characteristics.

[0068] Specifically, first, the server performs multi-dimensional matching and screening in the preset knowledge question bank according to the specific requirements of the learning objectives at each stage. The screening process considers the following factors: knowledge point coverage (ensuring that the selected resources fully cover the target knowledge points), difficulty gradient (the difficulty of the exercises should be adapted to the learning stage and present a reasonable progressive relationship), question type diversity (including different types of questions to cultivate comprehensive abilities), and personalized adaptation (selecting highly targeted questions based on students' learning characteristics and error types). For each learning stage, the server will generate a complete learning package that includes theoretical explanations, typical examples, practice questions, and ability-enhancing questions. The server will also set up an adaptive difficulty adjustment mechanism to dynamically adjust the difficulty level of subsequent courses and exercises based on students' actual learning performance.

[0069] S215, determining the revised statistical data based on the number of revised topics, the number of revisions in each category, and the revision period; Among them, the number of corrected questions refers to the total number of questions that students have corrected; the number of corrections in each category refers to the frequency of correction behavior under different classification dimensions; the correction period refers to the time interval when correction behavior occurs in a concentrated manner; correction statistics refers to a multi-dimensional data set that comprehensively reflects the characteristics of students' correction behavior.

[0070] Specifically, the server integrates the number of corrected questions, the number of corrections in each category, and the correction period to form multi-dimensional correction statistics. The server calculates the correlation between the correction statistics of each dimension and analyzes the overall characteristics of the correction behavior. For example, the correction concentration of questions in a specific category in a specific period of time is analyzed to identify the pattern characteristics of students' correction behavior. The server also generates visual expressions of the data, such as statistical charts, time distribution charts, etc., to facilitate subsequent teaching analysis and feedback.

[0071] S216, obtaining the student's original handwriting before the revision timeline; Specifically, the server will follow the correction timeline and then trace back to obtain the original handwriting when the question was first completed. The server will completely extract all the handwriting information of the original answer, including the spatial coordinate sequence of each stroke, the pressure change curve, and timestamp information. To ensure data integrity, the server also obtains auxiliary features such as pen lift information and handwriting direction changes during the writing process. For multiple answer records that may exist, the server will compare the timestamps to ensure that the original handwriting data of the first answer is obtained.

[0072] S217, comparing the handwriting features of the corrected handwriting with the original handwriting to obtain a comparison result, the handwriting features including stroke sequence, stroke strength, writing speed and handwriting continuity; Among them, stroke order refers to the order of strokes when writing; stroke force refers to the amount of pressure applied when writing; writing speed refers to the speed of movement when handwriting is formed; handwriting continuity refers to the smoothness of the writing process; comparison results refer to the difference analysis data of the handwriting characteristics of two times; handwriting characteristics refer to the various parameter indicators that can characterize the characteristics of writing.

[0073] Specifically, the server uses a handwriting feature analysis algorithm to perform a multi-dimensional comparison between the corrected handwriting and the original handwriting. In terms of stroke order, it analyzes whether the writing order of the strokes conforms to the Chinese character standard; in terms of stroke strength, it analyzes the changing characteristics of pen strength through pressure sensor data; in terms of writing speed, it calculates the instantaneous speed and average speed of the pen tip movement; in terms of handwriting continuity, it evaluates the number and duration of pauses during the writing process. The server will generate a detailed comparison report, including the difference values ​​of various handwriting features, change trend charts, and abnormal point annotations.

[0074] S218, determining a correction quality score based on the comparison result, wherein the higher the correction quality score is, the higher the correction quality is; Specifically, the server calculates the correction quality score through a comprehensive scoring model. The comprehensive scoring model considers multiple dimensions: handwriting standardization (including stroke integrity and character shape accuracy), writing seriousness (assessed by the stability of stroke strength and speed), correction process involvement (based on correction time and handwriting continuity analysis), error correction accuracy (assessing the correctness of the correction content), etc. The server sets weights for each dimension and uses a weighted summation method to calculate the final score, i.e., the correction quality score, which ranges from 0 to 100 points. The correction quality score refers to a score that expresses the degree of correction seriousness in numerical form.

[0075] S219. Classify the student's correction behavior according to the correction quality score to obtain a correction behavior classification result, which includes careful correction, mechanical correction and perfunctory correction.

[0076] Specifically, the server sets classification thresholds based on the correction quality score and divides the correction behavior into three categories. For correction behaviors with a correction quality score of 80 points or above, they are classified as serious corrections, characterized by standard handwriting, moderate speed, accurate corrections, and traces of serious thinking; for correction behaviors with scores between 50-80 points, they are classified as mechanical corrections, characterized by basically standard handwriting but lack of changes, correct correction content but lack of thinking annotations; for correction behaviors with scores below 50 points, they are classified as perfunctory corrections, characterized by sloppy handwriting, too fast speed, and incomplete correction of errors. The server will regularly count the distribution of students' correction behaviors and generate a correction attitude analysis report to provide a basis for teachers to understand students' correction status.

[0077] The following describes the server in the embodiment of the present invention from the perspective of hardware processing. Figure 3 , is a schematic diagram of a physical device structure of a server in an embodiment of the present application.

[0078] It should be noted that Figure 3 The structure of the server shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0079] like Figure 3 As shown, the server includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 to the random access memory (RAM) 303, such as executing the method described in the above embodiment. In RAM 303, various programs and data required for system operation are also stored. CPU 301, ROM 302 and RAM 303 are connected to each other through bus 304. Input / output (I / O) interface 305 is also connected to bus 304.

[0080] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD) and an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read therefrom is installed into the storage section 308 as needed.

[0081] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 309, and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are performed.

[0082] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0083] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings.

[0084] Specifically, the server of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, the correction timeline determination and handwriting analysis method provided in the above embodiment is implemented.

[0085] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the server described in the above embodiment; or may exist independently without being assembled into the server. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the server, the server implements the correction timeline determination and handwriting analysis method provided in the above embodiment.

[0086] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements 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 application.

[0087] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0088] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A method for determining a correction timeline and analyzing handwriting, characterized in that: Applied to a server, the method comprises: Acquire the teacher's correction record and the student's question handwriting data, wherein the teacher's correction record includes the teacher's correction time information for the student's question, and the student's question handwriting data includes the student's handwriting and writing time for the question; Based on the teacher's correction record, detecting whether each student has a class correction time and / or a student correction time, wherein the class correction time refers to the time when the teacher corrects the first question of the first student, and the student correction time refers to the time when the teacher corrects the first question of each student; If the first student has the class grading time and the student grading time, then the first student's revision timeline is set to the student grading time; If the second student has the class grading time and does not have the student grading time, then the second student's revision timeline is set to the class grading time; If the third student does not have the class grading time and the student grading time, then obtain the first submission time of the third student's question; Based on the student's question handwriting data, detect whether there is new handwriting of the completed question within a preset time period after the question is first submitted; If it does not exist, determining the deadline for question submission as the correction timeline for the third student according to the first submission time of the question and the preset duration; The handwriting in the student's question handwriting data whose writing time is later than the correction timeline is marked as corrected handwriting, and the correction statistical data is determined according to the corrected handwriting.

2. The method according to claim 1, characterized in that After the step of detecting whether there is newly added handwriting of the completed question within a preset time period after the first submission of the question based on the student's question handwriting data, the method further includes: If so, the appearance time of the newly added handwriting is set as the correction timeline of the third student.

3. The method according to claim 1, characterized in that After the step of obtaining the time when the third student first submitted the question if the third student does not have the class grading time and the student grading time, the method further includes: If the teacher's correction record contains the class correction time and the student correction time, setting the third student's correction timeline to the student correction time; If the class marking time exists in the teacher's marking record but the student marking time does not exist, the correction timeline of the third student is set to the class marking time.

4. The method according to claim 1, characterized in that After the step of marking the handwriting in the student's question handwriting data whose writing time is later than the correction timeline as the correction handwriting, and determining the correction statistical data according to the correction handwriting, the method further includes: Obtaining the student's original handwriting prior to the said revised timeline; Comparing the handwriting features of the corrected handwriting with the original handwriting to obtain a comparison result, wherein the handwriting features include stroke sequence, stroke strength, writing speed and handwriting continuity; Based on the comparison result, determining a correction quality score, wherein the higher the correction quality score is, the higher the correction quality is; The students' correction behaviors are classified according to the correction quality scores to obtain correction behavior classification results, which include serious correction, mechanical correction and perfunctory correction.

5. The method according to claim 1, characterized in that The step of marking the handwriting in the student's question handwriting data that is written later than the correction timeline as the correction handwriting, and determining the correction statistical data according to the correction handwriting, specifically includes: Determine the handwriting in the student's question handwriting data whose writing time is later than the correction timeline as the correction handwriting; Counting the questions corresponding to the corrected handwriting to determine the number of corrected questions; Classifying the questions corresponding to the corrected handwriting according to preset dimensions, and determining the number of corrections in each category; Perform time distribution statistics according to the correction time corresponding to the correction handwriting to determine the correction period; The revision statistics are determined based on the number of revised topics, the number of revisions in each category, and the revision period.

6. The method according to claim 5, characterized in that After the step of classifying the questions corresponding to the corrected handwriting according to the preset dimensions and determining the number of corrections in each category, the method further includes: Identify students' weak knowledge areas based on the number of corrections in each of the categories described; Determine the student's weak knowledge graph according to the correlation between the weak knowledge areas; The weak knowledge courses and personalized exercises of the student are generated according to the weak knowledge graph, and the personalized exercises include exercises corresponding to different weak knowledge points respectively.

7. The method according to claim 6, characterized in that The generating of the weak knowledge courses and personalized exercises of the student according to the weak knowledge graph, wherein the personalized exercises include exercises corresponding to different weak knowledge points, specifically including: Determine a weak knowledge point association strength matrix according to the weak knowledge graph, wherein the matrix element values ​​in the weak knowledge point association strength matrix are used to represent the association degree between different weak knowledge points; Based on the weak knowledge point association strength matrix, the weak knowledge points are hierarchically clustered to obtain a weak knowledge point hierarchy tree; Determine the learning priority of weak knowledge points according to the weak knowledge point hierarchy tree; Based on the learning priority of the weak knowledge points, generate a learning path and multi-stage learning objectives; According to the learning path, weak knowledge courses and personalized exercises that match the learning objectives of each stage are selected from the preset knowledge question bank.

8. A server, characterized in that: The server includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the server to execute the method described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a server, the server is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product is run on a server, the server is caused to execute the method according to any one of claims 1 to 7.

Images