A method, device and equipment for managing offline homework check-in in an online education training camp

Abstract

This invention discloses a method, device, and equipment for managing offline assignments in online education training camps, relating to the field of artificial intelligence technology. The method first extracts the answer content and answer duration for each test question from the target student's offline assignments. For each test question, it obtains the text similarity between the answer content and the standard answer, and based on the corresponding in-class monitoring video segments, it obtains the confidence level that the target student's listening status for the corresponding knowledge point is attentive. Then, it imports the answer duration, text similarity, and confidence level corresponding to each test question into an assignment plagiarism classification model, outputting another confidence level indicating that the target student's offline assignments contain plagiarism. Finally, based on the comparison result of this other confidence level with a preset threshold, it determines whether the target student has plagiarized their offline assignments. If so, a new offline assignment is pushed to the target student; otherwise, the offline assignment is accepted and the student completes the check-in.

Description

Technical Field

[0001] This invention belongs to the field of artificial intelligence technology, specifically relating to a method, device, and equipment for managing offline assignments in online education training camps. Background Technology

[0002] It is well known that the learning effect of a group is far greater than that of an individual. The online education training camps offered by educational institutions represent a new online blended training model that combines "online knowledge input + community services." By using online communities as a platform, students with similar learning needs are brought together, enabling them to complete learning tasks efficiently and systematically, while also creating more opportunities for exchange and sharing, greatly enhancing the learning experience.

[0003] Currently, training camp courses typically assign offline homework, requiring students to submit their completed assignments within a specified deadline. However, the lack of oversight of these assignments makes it difficult to identify plagiarism, which in turn affects the accuracy of training institutions' assessment of student progress. Summary of the Invention

[0004] The purpose of this invention is to provide a method, device, computer equipment, computer-readable storage medium, and computer program product for managing offline assignments in online education and training camps, in order to solve the problem that existing educational institutions have limited accuracy in grasping students' learning progress because they cannot distinguish whether students' offline assignments are plagiarized.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] Firstly, a method for managing offline assignments and check-ins in online education training camps is provided, including:

[0007] After receiving the first online education training camp offline assignments submitted by the target students, the first answer content and answer time of each test question are extracted from the first online education training camp offline assignments.

[0008] For each test question, the first text similarity between the corresponding standard answer and the corresponding first answer is calculated based on the standard answer content.

[0009] For each test question, according to the corresponding knowledge point, a segment of the in-class monitoring video of the target student is extracted from the in-class monitoring video. The video segment is collected during the same time period as the display time of the handout page of the knowledge point. The in-class monitoring video segment is then imported into a pre-trained class listening status classification model based on the first machine learning algorithm. The first confidence score of the target student’s class listening status for the corresponding knowledge point is that the student is listening attentively.

[0010] The answering time, the first text similarity and the first confidence level corresponding to each test question are imported into a homework plagiarism classification model based on the second machine learning algorithm that has been pre-trained, and the second confidence level of the first online education training camp offline homework being plagiarized is output.

[0011] Determine whether the second confidence level exceeds a preset first confidence threshold;

[0012] If so, then for the knowledge points to which each test question belongs, new test questions will be extracted and all of the new test questions will be pushed to the target student as new offline assignments. Otherwise, the submission timestamp of the first online education training camp offline assignment will be used as the offline assignment check-in timestamp of the target student.

[0013] Based on the above-mentioned invention, a new solution for managing students' offline homework check-in based on in-class monitoring videos and answer data is provided. First, the answer content and duration of each test question are extracted from the target student's offline homework. For each test question, the text similarity between the answer content and the standard answer is obtained. Then, based on the in-class monitoring video segmentation where the collection time coincides with the display time of the corresponding knowledge point's lecture notes page, the confidence level of the target student's attentive listening status for that knowledge point is obtained. Next, the answer duration, text similarity, and confidence level corresponding to each test question are imported into a homework plagiarism classification model, outputting another confidence level indicating plagiarism in the target student's offline homework. Finally, based on the comparison result of this other confidence level with a preset threshold, it is determined whether the target student has plagiarized their offline homework. If so, a new offline homework is pushed to the target student; otherwise, the offline homework is accepted and the student completes the check-in. This automatically identifies whether students' offline homework is plagiarized, improving the accuracy of educational institutions' grasp of students' learning progress and facilitating practical application and promotion.

[0014] In one possible design, the submission timestamp of the offline assignments of the first online education training camp is used as the offline assignment check-in timestamp for the target student, including:

[0015] For each student who has already checked in, determine whether the target student has copied the corresponding student's offline assignments;

[0016] If it is determined that the target student has copied the offline assignment of any student who has checked in earlier, then new test questions will be extracted for the knowledge points to which each test question belongs, and all of the new test questions will be pushed to the target student as new offline assignments. Otherwise, the submission timestamp of the first online education training camp offline assignment will be used as the offline assignment check-in timestamp of the target student.

[0017] In one possible design, for each student who has previously checked in, it is determined whether the target student has copied the corresponding student's offline assignments, including:

[0018] The second answer content for each test question is extracted from the offline assignment of the second online education training camp submitted by a student who had previously checked in.

[0019] For each test question, the second text similarity between the second answer and the corresponding first answer is calculated based on the second answer content.

[0020] The answering time, the second text similarity, and the first confidence level corresponding to each test question are imported into the homework plagiarism classification model, and the third confidence level of the offline homework of the first online education training camp is output to show that there is plagiarism.

[0021] Determine whether the third confidence level exceeds a preset second confidence threshold;

[0022] If so, it is determined that the target student copied the offline assignment of a student who had previously checked in; otherwise, it is determined that the target student did not copy the offline assignment of a student who had previously checked in.

[0023] In one possible design, for each test question, the first text similarity between the corresponding standard answer and the corresponding first answer is calculated, including:

[0024] The first response to a test question is segmented into words to obtain a first set of words, and the standard answer to the test question is also segmented into words to obtain a second set of words.

[0025] The first text similarity T between the standard answer to a certain test question and the first answer to that test question is calculated using the following formula. sim,1 :

[0026] T sim,1 =max(T) TF-IDF,sim ×T JSC,sim ,T MED,sim ×T JSC,sim )

[0027] In the formula, T TF-IDF,sim T represents the similarity between the first word set and the second word set, calculated based on Term Frequency-Inverse Document Frequency (TF-IDF). MED,sim T represents the degree of difference between the first word set and the second word set, calculated based on edit distance (MED). JSC,sim This represents the similarity between the first word set and the second word set, calculated based on the Jaccard similarity coefficient, and max() represents the maximum value function.

[0028] In one possible design, the first machine learning algorithm employs an artificial intelligence algorithm based on support vector machines, decision trees, or random forests, while the second machine learning algorithm employs an artificial intelligence algorithm based on K-nearest neighbor, stochastic gradient descent, multilayer perceptron, backpropagation neural network, or radial basis function network.

[0029] In one possible design, new test questions are extracted for each of the knowledge points to which the test questions belong, including:

[0030] For each test question, based on the first confidence level that the target learner is in a state of attentive listening at the corresponding knowledge point, the difficulty level of the corresponding question is determined and is negatively correlated with the first confidence level. New test questions are then extracted from the corresponding question bank based on the difficulty level.

[0031] Secondly, an online education training camp offline homework check-in management device is provided, which includes a response data extraction unit, a text similarity calculation unit, a class listening status classification unit, a homework plagiarism classification unit, a judgment unit, and a judgment response unit.

[0032] The answer data extraction unit is used to extract the first answer content and answer duration of each test question from the first online education training camp offline assignment submitted by the target student after receiving the first online education training camp offline assignment.

[0033] The text similarity calculation unit is communicatively connected to the answer data extraction unit, and is used to calculate the first text similarity between the standard answer content and the corresponding first answer content for each test question based on the corresponding standard answer content.

[0034] The listening status classification unit is communicatively connected to the answer data extraction unit. For each test question, according to the corresponding knowledge point, it extracts a segment of the listening monitoring video from the target student's in-class listening monitoring video. The video segment is collected during the same time period as the display time of the handout page of the knowledge point. The segment is then imported into a listening status classification model based on the first machine learning algorithm and pre-trained. The model outputs the first confidence score that the target student's listening status for the corresponding knowledge point is attentive listening.

[0035] The homework plagiarism classification unit is communicatively connected to the answer data extraction unit, the text similarity calculation unit, and the class listening status classification unit. It is used to import the answer duration, the first text similarity, and the first confidence level corresponding to each test question into the homework plagiarism classification model based on the second machine learning algorithm and which has been pre-trained, and output the second confidence level that the offline homework of the first online education training camp has plagiarism behavior.

[0036] The judgment unit is communicatively connected to the homework plagiarism classification unit and is used to determine whether the second confidence level exceeds a preset first confidence threshold.

[0037] The determination response unit, communicatively connected to the determination unit, is used to, when determining that the second confidence level exceeds the first confidence level threshold, re-extract corresponding new test questions for the knowledge points to which each test question belongs, and push all the new test questions as new offline assignments to the target student; and when determining that the second confidence level does not exceed the first confidence level threshold, use the submission timestamp of the first online education training camp offline assignment as the offline assignment check-in timestamp of the target student.

[0038] Thirdly, the present invention provides a computer device comprising a memory, a processor, and a transceiver connected in sequence for communication, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the online education training camp offline homework check-in management method as described in the first aspect or any possible design in the first aspect.

[0039] Fourthly, the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, perform the online education training camp offline assignment check-in management method as described in the first aspect or any possible design in the first aspect.

[0040] Fifthly, the present invention provides a computer program product, including a computer program or instructions, wherein when the computer program or instructions are executed by a computer, they implement the online education training camp offline homework check-in management method as described in the first aspect or any possible design in the first aspect.

[0041] The beneficial effects of the above scheme are:

[0042] (1) This invention creatively provides a new solution for managing students' offline homework check-in based on in-class monitoring videos and answer data. First, the answer content and answer duration of each test question are extracted from the offline homework of the target students. For each test question, the text similarity between the answer content and the standard answer is obtained. Then, based on the in-class monitoring video segmentation that is consistent with the display time of the corresponding knowledge point's lecture handout page, the confidence level of the target students' listening status on the corresponding knowledge point is obtained. Then, the answer duration, text similarity and confidence level corresponding to each test question are imported into the homework plagiarism classification model. Another confidence level is output to show that the target students' offline homework has plagiarism behavior. Finally, based on the comparison result of this other confidence level with the preset threshold, it is determined whether the target students have plagiarized offline homework. If so, a new offline homework is pushed to the target students. Otherwise, the offline homework is accepted and the check-in is completed. In this way, it can automatically identify whether students' offline homework has plagiarism, improve the accuracy of the training institution's grasp of students' learning status, and facilitate practical application and promotion. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 A flowchart illustrating the offline homework check-in management method for online education training camps provided in this embodiment of the application.

[0045] Figure 2 A schematic diagram of the structure of the online education training camp offline homework check-in management device provided in the embodiments of this application.

[0046] Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in conjunction with the accompanying drawings and descriptions of the embodiments or the prior art. Obviously, the following description of the structure of the accompanying drawings is only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained based on these embodiments without creative effort. It should be noted that the description of these embodiments is for the purpose of helping to understand the present invention, but does not constitute a limitation of the present invention.

[0048] It should be understood that although the terms "first" and "second", etc., may be used herein to describe various objects, these objects should not be limited by these terms. These terms are only used to distinguish one object from another. For example, the first object may be referred to as the second object, and similarly, the second object may be referred to as the first object, without departing from the scope of the exemplary embodiments of the invention.

[0049] It should be understood that the term "and / or" that may appear in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, or A and B exist simultaneously. Another example is A, B and / or C, which can mean that any one of A, B, and C or any combination thereof exists. The term " / and" that may appear in this document describes another relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone or A and B exist simultaneously. In addition, the character " / " that may appear in this document generally indicates that the related objects before and after it are in an "or" relationship.

[0050] Example:

[0051] like Figure 1 As shown, the online education training camp offline homework check-in management method provided in the first aspect of this embodiment can be executed, but is not limited to, by computer devices with certain computing resources, such as platform servers, personal computers (PCs, referring to a type of multi-purpose computer suitable for personal use in terms of size, price, and performance; desktops, laptops, mini-laptops, tablets, and ultrabooks are all considered personal computers), smartphones, personal digital assistants (PDAs), or wearable devices. Figure 1 As shown, the online education training camp offline assignment check-in management method may include, but is not limited to, the following steps S1 to S6.

[0052] S1. After receiving the first online education training camp offline assignments submitted by the target students, extract the first answer content and answer time of each test question from the first online education training camp offline assignments.

[0053] In step S1, the target student is the object to be managed; the first online education training camp offline assignment is the target student's answer to the offline assignment given after completing the training camp course. Specifically, the assignment distribution, answering, and submission can be implemented by referring to the operation details of existing examination software. The first answer content is the target student's answer result data for the corresponding test questions, which can be extracted conventionally based on the preset answer position, such as extracting multiple-choice options, fill-in-the-blank input characters, or true / false options from parentheses that were originally empty parentheses; the answering time is the data of the target student's answering process for the corresponding test questions. The display time of the test question page for the corresponding test question recorded by the examination software can be used as the answering time (existing examination software will automatically or manually jump to the answer display page of the next question after completing a question, so the display time of the test question page can be used as the answering time).

[0054] S2. For each test question, calculate the first text similarity between the corresponding standard answer and the corresponding first answer based on the standard answer content.

[0055] In step S2, the standard answer content for each test question is generated synchronously when the corresponding test question is generated, and therefore can be read normally. Preferably, for each test question, the first text similarity between the standard answer content and the corresponding first answer content is calculated based on the corresponding standard answer content, including but not limited to the following steps S21 to S22.

[0056] S21. Perform word segmentation on the first answer to a test question to obtain a first word set, and perform the same word segmentation on the standard answer to the test question to obtain a second word set.

[0057] In step S21, the word segmentation process can be implemented, but is not limited to, using the jieba word segmentation tool (which is a widely used Chinese word segmentation tool that supports multiple segmentation modes, including precise mode, full mode, and search engine mode. The precise mode aims to segment sentences most accurately and is suitable for text analysis; the full mode scans all possible word units in the sentence, which is faster but does not resolve ambiguity; the search engine mode further segments long words based on the precise mode to improve recall rate, and is particularly suitable for search engine applications).

[0058] S22. Calculate the first text similarity T between the standard answer content of a certain test question and the first answer content of the certain test question according to the following formula. sim,1 :

[0059] T sim,1 =max(T) TF-IDF,sim ×T JSC,sim ,T MED,sim ×T JSC,sim )

[0060] In the formula, T TF-IDF,sim T represents the similarity between the first word set and the second word set, calculated based on TF-IDF (Term Frequency–Inverse Document Frequency, a statistical method used to assess the importance of a word to a document in a document set or corpus). MED,sim T represents the degree of difference between the first word set and the second word set, calculated based on the minimum edit distance (MED, a quantitative measure of the difference between two strings, determined by the minimum number of processing steps required to transform one string into another). JSC,sim This indicates the similarity between the first word set and the second word set, calculated based on the Jaccard Similarity Coefficient (which measures the similarity between two sets, defined as the number of elements in the intersection of the two sets divided by the number of elements in the union). max() represents the maximum value function.

[0061] In step S22, the aforementioned intermediate parameter T TF-IDF,sim T MED,sim and T JSC,sim The specific calculation formulas are all based on existing technical methods. Although T TF-IDF,sim In the term frequency-inverse document frequency dimension, it would be a cosine value, T. MED,sim In terms of edit distance dimension and T JSC,sim Normalization is applied to the Jaccard similarity coefficients so that they all take values ​​in the range [0,1]. However, it's important to note that the values ​​given in each dimension describe different degrees of similarity / difference. Even if the same value is given in each dimension, the degree of similarity / difference will differ. How to aggregate these three intermediate parameters under different circumstances is an important issue. Consider T... TF-IDF,simThe representation of meaning is affected by word frequency. That is, when the standard answer or the first response is short, the frequency of each word may be 1, making it difficult to distinguish the importance of each word. Therefore, T... TF-IDF,sim This would be detrimental to calculating the similarity between very short texts (i.e., between the standard answer and the first response). And T MED,sim With T TF-IDF,sim In contrast, it is better suited for evaluating differences between shorter texts because shorter texts have a limited number of words, requiring fewer computational steps to find the mapping of the second word (i.e., words in the second word set) from the first word set. T... JSC,sim This will be an indicator that is insensitive to text length; that is, regardless of the length of the text, it always determines the difference in sets by the size of the intersection. Therefore, using the above formula, T can be emphasized when the text is short. MED,sim ×T JSC,sim (At this time, T is generally) MED,sim Greater than T TF-IDF,sim This is used to comprehensively measure the similarity / difference between the first and second word sets, while for longer texts, T is emphasized. TF-IDF,sim ×T JSC,sim (At this time, T is generally) TF-IDF,sim Greater than T MED,sim This is used to comprehensively measure the similarity / difference between the first and second word sets, ensuring that regardless of text length, the aggregation of metrics (i.e., the first text similarity T) is guaranteed. sim,1 The accuracy of ).

[0062] S3. For each test question, extract a segment of the in-class monitoring video from the target student's in-class monitoring video, based on the corresponding knowledge point, and extract the in-class monitoring video segment whose collection time is consistent with the display time of the lecture handout page of the knowledge point. Then, import the in-class monitoring video segment into a pre-trained listening state classification model based on the first machine learning algorithm, and output the first confidence score that the target student's listening state on the corresponding knowledge point is a state of attentive listening.

[0063] In step S3, the correspondence between each test question and the knowledge point can be automatically or manually established when the corresponding test questions are generated. For example, for a certain knowledge point, AI (Artificial Intelligence) technology can be used to automatically generate multiple test questions covering different question types. The in-class monitoring video is a routinely recorded monitoring video of the target student's performance in the online classroom corresponding to the offline assignments of the first online education training camp (it can be routinely recorded by the student's camera and uploaded locally). For example, if the lecture notes page for a certain knowledge point is displayed in the online classroom from 10:00 to 10:05, then for that certain knowledge point, it is necessary to extract the in-class monitoring video segment between 10:00 and 10:05 from the target student's in-class monitoring video. Since the in-class monitoring video segments will display the micro-expression features of the target students during the time shown on the handout page, such as focus or distraction, a classification model for the class attendance status can be pre-trained based on a certain amount of in-class monitoring video segment samples and sample labels (e.g., attentive or inattentive status). This can be achieved through conventional calibration and verification modeling methods (the specific process includes model calibration and verification, i.e., first comparing the model simulation results with the measured data, and then adjusting the model parameters based on the comparison results to make the simulation results match the actual situation). Furthermore, specifically, the first machine learning algorithm uses artificial intelligence algorithms based on support vector machines, decision trees, or random forests (which are core artificial intelligence algorithms that specifically study how computers simulate or realize human learning behavior to acquire new knowledge or skills, reorganize existing knowledge structures to continuously improve their performance, and are the fundamental way to make computers intelligent).

[0064] S4. The answering time, the first text similarity and the first confidence level corresponding to each test question are imported into the homework plagiarism classification model based on the second machine learning algorithm and which has been pre-trained, and the second confidence level of the first online education training camp offline homework plagiarism behavior is output.

[0065] In step S4, since the answering time and the first text similarity can reflect the target student's first level of mastery of the knowledge point corresponding to the test question from the perspective of answering questions (for example, the faster and more accurately the target student answers, the higher the level of mastery, and vice versa), and the first confidence level can reflect the target student's second level of mastery of the knowledge point corresponding to the test question from the perspective of listening to lectures (for example, the more attentively the target student listens to lectures on a certain knowledge point (at which point the first confidence level is higher), it can reflect that the target student has a higher level of mastery of that knowledge point, and vice versa), plagiarism can be identified through big data analysis based on the matching results of the first level of mastery and the second level of mastery: if they match, there is no plagiarism; if they do not match, plagiarism may exist. Therefore, after pre-training the homework plagiarism classification model using conventional calibration and verification modeling methods based on a certain amount of sample data (i.e., sample response time, sample similarity, and sample confidence, etc.) and label data (e.g., plagiarism or non-plagiarism), the second confidence level of plagiarism in the offline homework of the first online education training camp can be estimated using this homework plagiarism classification model. Furthermore, specifically, the second machine learning algorithm employs artificial intelligence algorithms based on K-nearest neighbor method, stochastic gradient descent method, multilayer perceptron, backpropagation neural network, or radial basis function network.

[0066] S5. Determine whether the second confidence level exceeds the preset first confidence level threshold.

[0067] In step S5, the first confidence threshold can be, for example, 62%.

[0068] S6. If yes, then for the knowledge points to which each test question belongs, extract the corresponding new test questions and push all the new test questions as new offline assignments to the target student; otherwise, use the submission timestamp of the first online education training camp offline assignment as the offline assignment check-in timestamp of the target student.

[0069] In step S6, if the second confidence level exceeds the first confidence level threshold, it can be considered that the offline assignments submitted by the target student in the first online education training camp contain plagiarism. The assignments need to be resubmitted to ensure the target student's mastery of the knowledge points covered by each test question, enabling them to keep up with subsequent online education training camp courses and improving the effectiveness of the online education training camp activities. Preferably, for each knowledge point covered by the test questions, new test questions are re-selected, including but not limited to: for each knowledge point covered by the test questions, based on the first confidence level that the target student's listening status on the corresponding knowledge point is attentive, determining the corresponding question difficulty negatively correlated with the first confidence level, and selecting new test questions from the corresponding question bank based on the question difficulty. The purpose of establishing a negative correlation between the first confidence level and the difficulty of the questions is as follows: if a target student listens attentively to a particular knowledge point (i.e., the higher the first confidence level), it reflects that the target student has a higher level of mastery of that knowledge point. Therefore, easier questions can be provided for routine offline training. Conversely, if a target student does not listen attentively to a particular knowledge point (i.e., the lower the first confidence level), it reflects that the target student has a lower level of mastery of that knowledge point. Therefore, more difficult questions can be provided for focused offline training to improve the target student's mastery of that knowledge point. In this targeted question-setting method, the mastery of knowledge points by all students can be balanced, which further helps them keep up with the subsequent online education training camp courses.

[0070] In step S6, in order to further distinguish whether there is plagiarism in the offline assignments of the students, preferably, the submission timestamp of the offline assignment of the first online education training camp is used as the offline assignment check-in timestamp of the target student, including but not limited to the following steps S61 to S62.

[0071] S61. For each student who has checked in beforehand, determine whether the target student has copied the corresponding student's offline assignment.

[0072] In step S61, the "previously checked-in student" refers to other target students who successfully completed the homework plagiarism identification process based on the aforementioned steps S1-S6 and successfully checked in (i.e., determined the offline homework check-in timestamp) because no homework plagiarism was found. Specifically, for each previously checked-in student, it is determined whether the target student plagiarized the corresponding student's offline homework, including but not limited to the following steps S611-S615.

[0073] S611. Extract the second answers to each test question from the second online education training camp offline assignments submitted by a student who previously checked in.

[0074] In step S611, the second online education training camp offline assignment refers to the answer result of a previously checked-in student on the offline assignment given after completing the training camp course. Specifically, the assignment distribution, answering, and submission can be implemented by referring to the operational details of existing examination software. The second answer content is the answer result data of the previously checked-in student on the corresponding test questions, which can also be routinely extracted based on preset answer positions, such as extracting multiple-choice options, fill-in-the-blank input characters, or true / false options from parentheses that were originally empty.

[0075] S612. For each test question, calculate the second text similarity between the second answer and the corresponding first answer based on the second answer content.

[0076] S613. The answering time, the second text similarity, and the first confidence level corresponding to each test question are imported into the homework plagiarism classification model, and the third confidence level of plagiarism behavior in the offline homework of the first online education training camp is output.

[0077] S614. Determine whether the third confidence level exceeds the preset second confidence level threshold.

[0078] In step S614, the second confidence threshold can also be exemplified as 62%. Furthermore, the specific technical details of the aforementioned steps S612 to S614 can be derived by referring to the conventional derivation of the aforementioned steps S2 and S4 to S5, and will not be repeated here.

[0079] S615. If yes, then it is determined that there is a situation where the target student copied the offline assignment of a student who had previously checked in; otherwise, it is determined that there is no situation where the target student copied the offline assignment of a student who had previously checked in.

[0080] S62. If it is determined that the target student has copied the offline assignment of any student who has checked in earlier, then for the knowledge points to which each test question belongs, new test questions are re-extracted and all the new test questions are pushed to the target student as new offline assignments. Otherwise, the submission timestamp of the first online education training camp offline assignment is used as the offline assignment check-in timestamp of the target student.

[0081] Therefore, based on the online education training camp offline homework check-in management method described in steps S1 to S6 above, a new solution is provided for managing students' offline homework check-in based on in-class monitoring videos and answer data. This involves first extracting the answer content and answer duration for each test question from the target student's offline homework; then, for each test question, obtaining the text similarity between the answer content and the standard answer; and finally, segmenting the in-class monitoring video according to the time period of collection and the display time of the corresponding knowledge point's lecture notes page to obtain the target student's listening status for the corresponding knowledge point. The system first establishes a confidence level for genuine student engagement in class. Then, it imports the response time, text similarity, and confidence level corresponding to each test question into a homework plagiarism classification model. This model outputs another confidence level indicating whether the target student's offline homework contains plagiarism. Finally, based on the comparison between this other confidence level and a preset threshold, it determines whether the target student has plagiarized their offline homework. If so, a new offline homework assignment is sent to the target student; otherwise, the student completes the offline homework and checks in. This method can automatically identify whether students' offline homework contains plagiarism, improving the accuracy of educational institutions' understanding of students' learning progress and facilitating practical application and promotion.

[0082] like Figure 2 As shown, the second aspect of this embodiment provides a virtual device for implementing the online education training camp offline homework check-in management method described in the first aspect, including a response data extraction unit, a text similarity calculation unit, a class attendance status classification unit, a homework plagiarism classification unit, a judgment unit, and a judgment response unit;

[0083] The answer data extraction unit is used to extract the first answer content and answer duration of each test question from the first online education training camp offline assignment submitted by the target student after receiving the first online education training camp offline assignment.

[0084] The text similarity calculation unit is communicatively connected to the answer data extraction unit, and is used to calculate the first text similarity between the standard answer content and the corresponding first answer content for each test question based on the corresponding standard answer content.

[0085] The listening status classification unit is communicatively connected to the answer data extraction unit. For each test question, according to the corresponding knowledge point, it extracts a segment of the listening monitoring video from the target student's in-class listening monitoring video. The video segment is collected during the same time period as the display time of the handout page of the knowledge point. The segment is then imported into a listening status classification model based on the first machine learning algorithm and pre-trained. The model outputs the first confidence score that the target student's listening status for the corresponding knowledge point is attentive listening.

[0086] The homework plagiarism classification unit is communicatively connected to the answer data extraction unit, the text similarity calculation unit, and the class listening status classification unit. It is used to import the answer duration, the first text similarity, and the first confidence level corresponding to each test question into the homework plagiarism classification model based on the second machine learning algorithm and which has been pre-trained, and output the second confidence level that the offline homework of the first online education training camp has plagiarism behavior.

[0087] The judgment unit is communicatively connected to the homework plagiarism classification unit and is used to determine whether the second confidence level exceeds a preset first confidence threshold.

[0088] The determination response unit, communicatively connected to the determination unit, is used to, when determining that the second confidence level exceeds the first confidence level threshold, re-extract corresponding new test questions for the knowledge points to which each test question belongs, and push all the new test questions as new offline assignments to the target student; and when determining that the second confidence level does not exceed the first confidence level threshold, use the submission timestamp of the first online education training camp offline assignment as the offline assignment check-in timestamp of the target student.

[0089] The working process, working details and technical effects of the aforementioned device provided in the second aspect of this embodiment can be found in the online education training camp offline homework check-in management method described in the first aspect, and will not be repeated here.

[0090] like Figure 3 As shown, the third aspect of this embodiment provides a computer device for executing the online education training camp offline assignment check-in management method as described in the first aspect. The device includes a memory, a processor, and a transceiver connected in sequence. The memory stores a computer program, the transceiver sends and receives messages, and the processor reads the computer program to execute the online education training camp offline assignment check-in management method as described in the first aspect. Specifically, the memory may include, but is not limited to, random-access memory (RAM), read-only memory (ROM), flash memory, first-in-first-out (FIFO) memory, and / or first-in-last-out (FILO) memory, etc.; the processor may include, but is not limited to, a microprocessor of the STM32F105 series. Furthermore, the computer device may also include, but is not limited to, a power module, a display screen, and other necessary components.

[0091] The working process, working details and technical effects of the aforementioned computer equipment provided in the third aspect of this embodiment can be found in the offline homework check-in management method for online education training camps described in the first aspect, and will not be repeated here.

[0092] This fourth aspect of the embodiment provides a computer-readable storage medium storing instructions comprising the online education training camp offline assignment check-in management method as described in the first aspect. Specifically, the computer-readable storage medium stores instructions that, when executed on a computer, perform the online education training camp offline assignment check-in management method as described in the first aspect. The computer-readable storage medium refers to a data storage medium, which may include, but is not limited to, floppy disks, optical disks, hard disks, flash memory, USB flash drives, and / or Memory Sticks. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0093] The working process, working details and technical effects of the aforementioned computer-readable storage medium provided in the fourth aspect of this embodiment can be found in the online education training camp offline homework check-in management method described in the first aspect, and will not be repeated here.

[0094] This fifth aspect of the embodiment provides a computer program product, including a computer program or instructions, which, when executed by a computer, implements the online education training camp offline assignment check-in management method as described in the first aspect. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0095] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for managing offline assignment check-ins in online education training camps, characterized in that, include: After receiving the first online education training camp offline assignments submitted by the target students, the first answer content and answer time of each test question are extracted from the first online education training camp offline assignments. For each test question, the first text similarity between the corresponding standard answer and the corresponding first answer is calculated based on the standard answer content. Specifically, this includes: performing word segmentation on the first answer of a test question to obtain a first word set, and performing the same word segmentation on the standard answer of the test question to obtain a second word set; the first text similarity between the standard answer of a test question and the first answer of the test question is calculated using the following formula. : In the formula, This represents the similarity between the first word set and the second word set, calculated based on Term Frequency-Inverse Document Frequency (TF-IDF). This represents the degree of difference between the first word set and the second word set, calculated based on the edit distance (MED). This indicates the similarity between the first word set and the second word set, calculated based on the Jaccard similarity coefficient. This represents the function that takes the maximum value. For each test question, according to the corresponding knowledge point, a segment of the in-class monitoring video of the target student is extracted from the in-class monitoring video. The video segment is collected during the same time period as the display time of the handout page of the knowledge point. The in-class monitoring video segment is then imported into a pre-trained class listening status classification model based on the first machine learning algorithm. The first confidence score of the target student’s class listening status for the corresponding knowledge point is that the student is listening attentively. The answering time, the first text similarity and the first confidence level corresponding to each test question are imported into a homework plagiarism classification model based on the second machine learning algorithm that has been pre-trained, and the second confidence level of the first online education training camp offline homework being plagiarized is output. Determine whether the second confidence level exceeds a preset first confidence threshold; If so, then for the knowledge points to which each test question belongs, new test questions are extracted and all of these new test questions are pushed to the target student as new offline assignments. Otherwise, the submission timestamp of the first online education training camp's offline assignment is used as the target student's offline assignment check-in timestamp. Specifically, this includes: for each student who has checked in earlier, determining whether the target student has copied the offline assignment of the corresponding student; if it is determined that the target student has copied the offline assignment of any student who has checked in earlier, then for the knowledge points to which each test question belongs, new test questions are extracted and all of these new test questions are pushed to the target student as new offline assignments. Otherwise, the submission timestamp of the first online education training camp's offline assignment is used as the target student's offline assignment check-in timestamp.

2. The method for managing offline assignments in online education training camps according to claim 1, characterized in that, For each student who has already checked in, determine whether the target student plagiarized the corresponding student's offline assignments, including: The second answer content for each test question is extracted from the offline assignment of the second online education training camp submitted by a student who had previously checked in. For each test question, the second text similarity between the second answer and the corresponding first answer is calculated based on the second answer content. The answering time, the second text similarity, and the first confidence level corresponding to each test question are imported into the homework plagiarism classification model, and the third confidence level of the offline homework of the first online education training camp is output to show that there is plagiarism. Determine whether the third confidence level exceeds a preset second confidence threshold; If so, it is determined that the target student copied the offline assignment of a student who had previously checked in; otherwise, it is determined that the target student did not copy the offline assignment of a student who had previously checked in.

3. The method for managing offline assignments in online education training camps according to claim 1, characterized in that, The first machine learning algorithm uses an artificial intelligence algorithm based on support vector machine, decision tree or random forest, and the second machine learning algorithm uses an artificial intelligence algorithm based on K nearest neighbor method, stochastic gradient descent method, multilayer perceptron, backpropagation neural network or radial basis function network.

4. The method for managing offline assignments in online education training camps according to claim 1, characterized in that, For each of the aforementioned test questions, new test questions will be drawn to correspond to the relevant knowledge points, including: For each test question, based on the first confidence level that the target learner is in a state of attentive listening at the corresponding knowledge point, the difficulty level of the corresponding question is determined and is negatively correlated with the first confidence level. New test questions are then extracted from the corresponding question bank based on the difficulty level.

5. A device for managing offline assignments in an online education training camp, characterized in that, It includes a response data extraction unit, a text similarity calculation unit, a class attendance status classification unit, a homework plagiarism classification unit, a judgment unit, and a judgment response unit; The answer data extraction unit is used to extract the first answer content and answer duration of each test question from the first online education training camp offline assignment submitted by the target student after receiving the first online education training camp offline assignment. The text similarity calculation unit, communicatively connected to the answer data extraction unit, is used to calculate, for each test question, a first text similarity between the corresponding standard answer content and the corresponding first answer content, specifically including: performing word segmentation on the first answer content of a test question to obtain a first word set, and performing the word segmentation on the standard answer content of the test question to obtain a second word set; and calculating the first text similarity between the standard answer content of a test question and the first answer content of the test question according to the following formula. : In the formula, This represents the similarity between the first word set and the second word set, calculated based on Term Frequency-Inverse Document Frequency (TF-IDF). This represents the degree of difference between the first word set and the second word set, calculated based on the edit distance (MED). This indicates the similarity between the first word set and the second word set, calculated based on the Jaccard similarity coefficient. This represents the function that takes the maximum value. The listening status classification unit is communicatively connected to the answer data extraction unit. For each test question, according to the corresponding knowledge point, it extracts a segment of the listening monitoring video from the target student's in-class listening monitoring video. The video segment is collected during the same time period as the display time of the handout page of the knowledge point. The segment is then imported into a listening status classification model based on the first machine learning algorithm and pre-trained. The model outputs the first confidence score that the target student's listening status for the corresponding knowledge point is attentive listening. The homework plagiarism classification unit is communicatively connected to the answer data extraction unit, the text similarity calculation unit, and the class listening status classification unit. It is used to import the answer duration, the first text similarity, and the first confidence level corresponding to each test question into the homework plagiarism classification model based on the second machine learning algorithm and which has been pre-trained, and output the second confidence level that the offline homework of the first online education training camp has plagiarism behavior. The judgment unit is communicatively connected to the homework plagiarism classification unit and is used to determine whether the second confidence level exceeds a preset first confidence threshold. The determination response unit, communicatively connected to the determination unit, is used to, when determining that the second confidence level exceeds the first confidence level threshold, re-extract corresponding new test questions for the knowledge points to which each test question belongs, and push all the new test questions as new offline assignments to the target student; and when determining that the second confidence level does not exceed the first confidence level threshold, use the submission timestamp of the first online education training camp offline assignment as the offline assignment check-in timestamp of the target student. Specifically, this includes: for each previously checked-in student, determining whether the target student has copied the offline assignment of the corresponding student; if it is determined that the target student has copied the offline assignment of any previously checked-in student, then re-extract corresponding new test questions for the knowledge points to which each test question belongs, and push all the new test questions as new offline assignments to the target student; otherwise, use the submission timestamp of the first online education training camp offline assignment as the offline assignment check-in timestamp of the target student.

6. A computer device, characterized in that, The device includes a memory, a processor, and a transceiver connected in sequence for communication. The memory is used to store computer programs, the transceiver is used to send and receive messages, and the processor is used to read the computer programs and execute the online education training camp offline homework check-in management method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that... The computer-readable storage medium stores instructions that, when executed on a computer, perform the online education training camp offline homework check-in management method as described in any one of claims 1 to 4.

8. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or the instructions are executed by the computer, they implement the online education training camp offline homework check-in management method as described in any one of claims 1 to 4.

Images