AI-based intelligent grading and analysis methods and systems for optimizing teaching quality

Abstract

This invention relates to the field of AI data analysis technology, and in particular to a teaching quality optimization method and system based on AI-powered intelligent grading and analysis. The invention divides the homework explanation video stream into several grading and analysis periods based on several explanation and grading feature frames within the video stream. It determines the grading content tendency category of the homework explanation video stream based on the differences in content value coefficients among the various grading and analysis periods. Based on the grading content tendency category, it determines the evaluation factors of the homework explanation video stream. Based on the evaluation factors, it judges whether the allocation of teaching resources is reasonable and generates supplementary resource information recommended to students. Finally, it sends the generated supplementary resource information to the student's terminal. Thus, it realizes the construction of a dual-dimensional evaluation system of explanation and annotation, scientifically determines the rationality of teaching resource allocation and generates supplementary resource information, and realizes the transformation of the smart classroom from an experience-driven to a data-driven decision-making model.

Description

Technical Field

[0001] This invention relates to the field of AI data analysis technology, and in particular to a method and system for optimizing teaching quality based on AI-based intelligent grading and analysis. Background Technology

[0002] In the current education field, traditional teaching resource recommendations have significant limitations. With the rapid development of educational informatization, optimizing teaching quality remains a core concern. Traditional teaching quality assessments rely heavily on teacher experience and limited manual analysis, making it difficult to comprehensively and accurately evaluate the rationality of the teaching process and resource allocation. In homework correction and explanation, teachers often struggle to quantify the importance of different content, and the allocation of teaching resources lacks scientific data support, potentially leading to students not receiving timely and targeted supplementary learning resources. Furthermore, traditional methods are inadequate for systematic analysis of teaching content and cannot quickly identify problems such as uneven resource distribution. As AI technology is increasingly applied in education, using AI to achieve intelligent analysis and optimization of the teaching process has become a research hotspot. However, existing AI-based teaching quality optimization solutions still have significant room for improvement in areas such as in-depth analysis of homework explanation videos, accurate evaluation of teaching resource allocation, and personalized resource recommendations for students, in order to meet the urgent needs of modern education for improved teaching quality.

[0003] For example, Chinese Patent Publication No. CN119672739A discloses a homework correction method based on image recognition and large language models, relating to the teaching field. This method addresses the problem of insufficient practicality in existing homework correction methods. The method includes the following steps: Step S1: Acquire multiple homework texts and multiple answer texts to obtain preliminary homework data; Step S2: Divide the multiple homework texts into a first type of homework text and a second type of homework text, and correct the first type and second type of homework texts respectively to obtain homework correction data; Step S3: Analyze the correctness of answers submitted by the target student to obtain a first homework quality assessment coefficient and a second homework quality assessment coefficient, obtaining correctness analysis data; Step S4: Analyze the correctness analysis data to assess the completion quality of the target student's current batch of submitted homework. This invention expands the functionality of existing homework correction methods and further improves their practicality.

[0004] The following problems still exist in the existing technology:

[0005] Existing technologies do not consider quantitatively evaluating the matching degree between teachers' explanations and annotations. They cannot construct a dual-dimensional evaluation system for explanations and annotations, and lack in-depth analysis of the homework explanation process. This results in unreasonable allocation of teaching resources and students' inability to obtain targeted learning resources to fill knowledge gaps. Summary of the Invention

[0006] To address this, the present invention provides a teaching quality optimization method and system based on AI-powered intelligent grading and analysis, which overcomes the problem that existing technologies cannot construct a dual-dimensional evaluation system of explanation and annotation, and lack in-depth analysis of the homework explanation process.

[0007] To achieve the above objectives, this invention provides a teaching quality optimization method based on AI-powered intelligent grading and analysis, comprising:

[0008] The teacher obtains a video stream of homework explanation during the teaching process, identifies several explanation and correction feature frames in the homework explanation video stream, and divides the homework explanation video stream into several correction and analysis time periods based on the explanation and correction feature frames.

[0009] The content value coefficient of each grading and analysis period is determined based on the duration of the grading and analysis period, and the grading content tendency category of the homework explanation video stream is determined based on the differences in the content value coefficients.

[0010] The evaluation factors for the homework explanation video stream are determined based on the category of the grading content. The evaluation factors include the correlation between the first format information and the second format information within the characteristic grading and analysis period, and the correlation between the first correlation parameter and the second correlation parameter within the grading and analysis period.

[0011] The feature-based parsing period is determined based on the content value coefficient.

[0012] Based on the evaluation factors, it is determined whether the allocation of teaching resources is reasonable. In response to the determination that the allocation of teaching resources is unreasonable, supplementary resource information recommended to students is generated. The supplementary resource information is either first recommended supplementary information generated based on the first format information, or second recommended supplementary information generated based on the content of the heterogeneous correlation correction and analysis period.

[0013] The heterogeneous association parsing time period is determined based on the difference between the first association parameter and the second association parameter;

[0014] The generated supplementary resource information will be sent to the student's terminal.

[0015] Furthermore, the method for determining several explanation and correction feature frames is to acquire images of consecutive frames at preset intervals within the homework explanation video stream, and determine the time when the correction questions in the image are switched as the time when the explanation and correction feature frame is located.

[0016] Furthermore, the process of determining the content value coefficient includes:

[0017] The time interval between two adjacent explanation and correction feature frames is defined as the correction and parsing time interval;

[0018] The duration of each grading and analysis period is determined, and the ratio of the duration to the total duration of the assignment explanation video stream is determined as the content value coefficient.

[0019] Furthermore, the process of determining the categorization tendency of the grading content in the homework explanation video stream includes:

[0020] Calculate the standard deviation of the content value coefficient for each graded analysis period;

[0021] Based on the comparison results of the standard deviation satisfying the video category differentiation conditions, the grading content tendency category of the homework explanation video stream is determined to be the grading tendency video category;

[0022] Based on the comparison results where the standard deviation does not meet the video category distinction criteria, the grading content of the homework explanation video stream is determined to be a non-grading video category.

[0023] The video category differentiation condition is that the standard deviation exceeds a preset standard deviation comparison value.

[0024] Furthermore, the evaluation factors for the assignment explanation video stream are determined as follows:

[0025] If the homework explanation video stream is a grading-oriented video category, then the evaluation factor is determined to be the correlation between the first format information and the second format information within the characteristic grading analysis period;

[0026] If the homework explanation video stream is a non-biased grading video category, then the evaluation factor is determined to be the correlation coefficient between the first correlation parameter and the second correlation parameter during the grading and analysis period.

[0027] Furthermore, the process of determining the relevance between the first-format information and the second-format information includes:

[0028] Filter the grading and analysis time period corresponding to the maximum content value coefficient in the assignment explanation video stream;

[0029] Acquire the voice information within the correction and parsing period and determine the voice keywords based on semantic analysis technology, and determine the voice keywords as the first format information;

[0030] Based on the temporal relationship, the image corresponding to the explanation and correction feature frame of the correction and analysis period is obtained. Based on the text recognition technology, the keyword text of the newly added text in the image corresponding to the explanation and correction feature frame before and after the time sequence is determined, and the keyword text is determined as the second format information.

[0031] The Euclidean distance between the first format information and the second format information is determined as the correlation.

[0032] Furthermore, the process of determining the rationality of the allocation of teaching resources based on the relevance and generating supplementary resource information includes:

[0033] Based on the comparison results of the relevance exceeding the preset relevance threshold, it is determined that the allocation of teaching resources is unreasonable;

[0034] In response to the result that the configuration is unreasonable based on the relevance determination, the resource supplementary information to be sent to the student is the first recommended supplementary information generated based on the first format information;

[0035] Wherein, the content keywords of the first recommended supplementary information and the Euclidean distance in the first format information do not exceed the preset reference Euclidean distance.

[0036] Furthermore, the process of determining the association relationship between the first association parameter and the second association parameter includes:

[0037] Based on speech recognition, the number of words in the text converted from speech information within each batching and parsing period is determined, and the number of words is determined as the first association parameter;

[0038] Based on the temporal relationship, the images corresponding to the explanation and correction feature frames of each correction and analysis period are obtained. The number of text characters in the images corresponding to the explanation and correction feature frames of the preceding and following time periods are determined according to the text recognition technology. The number of text characters is then determined as the second association parameter.

[0039] The parsing time periods are sorted in ascending order of the first correlation parameter, and the second correlation parameters of each parsing time period are combined into a correlation relationship sequence according to the sorting.

[0040] Furthermore, the process of determining the rationality of the allocation of teaching resources based on correlations and generating supplementary resource information includes:

[0041] If the sequence of relationships is not an increasing sequence, then the allocation of teaching resources is deemed unreasonable.

[0042] In response to the result of unreasonable configuration determined based on the association relationship, the resource supplement information to be sent to students is the second recommended supplement information generated based on the content of the heterogeneous association grading and parsing period;

[0043] Based on the comparison results of the difference between the first associated parameter and the second associated parameter exceeding the difference comparison value within the batch parsing period, the batch parsing period is determined as the heterogeneous associated batch parsing period, and the difference comparison value is the average of the differences between the first associated parameter and the second associated parameter within several batch parsing periods;

[0044] Wherein, the Euclidean distance between the keywords of the second recommended supplementary information and the content of the heterogeneous correlation parsing period does not exceed the preset reference Euclidean distance.

[0045] Furthermore, the present invention also provides a teaching quality optimization system based on AI intelligent grading and analysis, including:

[0046] The information acquisition unit includes a video recorder for acquiring the homework explanation video stream and a microphone for acquiring audio information. The video recorder is also a video processor for dividing the homework explanation video stream into several grading and analysis time periods.

[0047] The feature extraction unit, which is connected to the information acquisition unit, is used to determine the content value coefficient of each grading and analysis period and to determine the grading content tendency category of the homework explanation video stream based on the differences in the content value coefficients.

[0048] An intelligent evaluation unit, which is connected to the information acquisition unit and the feature extraction unit respectively, is used to determine the evaluation factors of the homework explanation video stream based on the tendency category of the grading content.

[0049] The content recommendation generation unit, which is connected to the intelligent evaluation unit, is used to determine whether the allocation of teaching resources is reasonable based on the evaluation factors, and to generate supplementary resource information to recommend to students.

[0050] A transmission unit, connected to the content recommendation generation unit, is used to send the generated supplementary resource information to the student terminal.

[0051] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention divides the homework explanation video stream into several correction and analysis periods based on several explanation and correction feature frames within the video stream. It determines the correction content tendency category of the homework explanation video stream based on the differences in content value coefficients among the various correction and analysis periods. Based on the correction content tendency category, it determines the evaluation factors of the homework explanation video stream. Based on the evaluation factors, it judges whether the allocation of teaching resources is reasonable and generates supplementary resource information recommended to students. Finally, it sends the generated supplementary resource information to the student's terminal. Thus, it realizes the construction of a dual-dimensional evaluation system of explanation and annotation, scientifically determines the rationality of teaching resource allocation and generates supplementary resource information, and realizes the transformation of smart classrooms from experience-driven to data-driven decision-making models.

[0052] Furthermore, this invention acquires continuous frame images and determines the switching time of the grading questions as the explanation and grading feature frame. This method can clearly and accurately divide the homework explanation video stream into different grading and analysis periods. By determining the time period between adjacent explanation and grading feature frames as the grading and analysis period, and using the ratio of its duration to the total video duration as the content value coefficient, a quantitative assessment of the importance of each grading and analysis period in the entire homework grading process is achieved. By calculating the standard deviation of the content value coefficient of each grading and analysis period, and determining the grading content tendency category based on the relationship between the standard deviation and the preset standard deviation comparison value, it is helpful to adopt more targeted teaching resource allocation strategies for different types of explanation videos.

[0053] Furthermore, this invention, targeting the video category of correction tendency, selects the feature correction and analysis period corresponding to the maximum content value coefficient as the key to achieving teaching objectives. By focusing on this core period, it avoids wasting computing power by focusing on too much non-critical content. The speech keywords determined by semantic analysis technology during the correction and analysis period are used as the first format information, and the keyword text of the newly added text in the explanation and correction feature frame image obtained based on text recognition technology is used as the second format information. Key information is extracted from these two important dimensions: the teacher's explanation speech and the newly added text of the correction content update. The speech keywords reflect the key content conveyed orally by the teacher, while the keyword text of the newly added text reflects the key knowledge points presented in written form. This multi-dimensional extraction of key information comprehensively and deeply covers the important content carried by different carriers in the teaching process, realizing the construction of a dual-dimensional evaluation system of explanation and annotation, and realizing the transformation of smart classroom from experience-driven to data-driven decision-making mode.

[0054] Furthermore, after determining that the allocation of teaching resources is unreasonable, this invention generates first recommended supplementary information based on the first format information. This fully considers students' learning needs for the key content explained by the teacher. The first format information is key information obtained through semantic analysis of the teacher's lecture voice, representing the core knowledge conveyed orally by the teacher. The supplementary information generated based on this information can accurately supplement any shortcomings that students may have in understanding the teacher's explanation of assignments. Moreover, it requires that the Euclidean distance between the keywords in the first recommended supplementary information and the first format information does not exceed a preset reference Euclidean distance. This ensures the close relevance between the supplementary information and the teacher's key points, guarantees the consistency of the supplementary resources with the teaching focus in terms of content, enables students to efficiently obtain valuable learning resources, forms a dynamic and adaptive teaching resource ecosystem, and realizes the transformation of the smart classroom from an experience-driven to a data-driven decision-making model.

[0055] Furthermore, for non-biased video categories, this invention utilizes speech recognition and text recognition technologies to acquire relevant data when determining the correlation. After sorting the first correlation parameter, a correlation sequence is constructed, which can intuitively show the relationship between the amount of teacher explanation and the amount of text updates in homework correction, preventing the impact of configuration imbalance on teaching effectiveness. In terms of generating supplementary resource information, a second recommended supplementary information is generated based on the content of the heterogeneous correlation correction analysis period, accurately locating the periods of abnormal information output during the teaching process. These periods are determined by calculating the average difference of correlation parameters, which can effectively make up for weak links in teaching. At the same time, it is required that the Euclidean distance between the keywords of the supplementary information content and the content of the heterogeneous correlation correction analysis period meets the preset standard, ensuring that the supplementary information is closely related to the teaching content of the problem period, realizing the transformation of the smart classroom from an experience-driven to a data-driven decision-making model. Attached Figure Description

[0056] Figure 1 This is a flowchart illustrating the steps of the teaching quality optimization method based on AI-powered intelligent grading and analysis in an embodiment of the present invention.

[0057] Figure 2 A flowchart illustrating the logic of determining the grading content preference category of the assignment explanation video stream in an embodiment of the present invention;

[0058] Figure 3 This is a flowchart illustrating the steps of determining the correlation between first format information and second format information in an embodiment of the present invention.

[0059] Figure 4 A flowchart illustrating the steps for determining the association relationship between the first association parameter and the second association parameter in an embodiment of the present invention;

[0060] Figure 5 This is a system block diagram of the teaching quality optimization system based on AI intelligent grading and analysis, as described in an embodiment of the present invention. Detailed Implementation

[0061] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0062] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0063] It should be noted that in the description of this invention, the terms "upper," "lower," "inner," "outer," etc., which indicate the direction or positional relationship, are based on the direction or positional relationship shown in the drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0064] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "connected" and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0065] Please see Figure 1 The diagram illustrates the steps of an AI-based intelligent grading and analysis method for optimizing teaching quality according to an embodiment of the present invention. The AI-based intelligent grading and analysis method for optimizing teaching quality includes:

[0066] Step S100: Obtain the video stream of the teacher explaining the homework during the teaching process, determine several explanation and correction feature frames in the video stream, and divide the video stream into several correction and analysis time periods according to the explanation and correction feature frames.

[0067] Step S200: Determine the content value coefficient of each grading and analysis period based on the duration of the grading and analysis period, and determine the grading content tendency category of the homework explanation video stream based on the differences in the content value coefficients.

[0068] Step S300: Determine the evaluation factors of the homework explanation video stream based on the grading content tendency category. The evaluation factors include the correlation between the first format information and the second format information within the characteristic grading and parsing time period, and the correlation between the first correlation parameter and the second correlation parameter within the grading and parsing time period.

[0069] The feature-based parsing period is determined based on the content value coefficient.

[0070] Step S400: Determine whether the allocation of teaching resources is reasonable based on the evaluation factors. In response to the determination that the allocation of teaching resources is unreasonable, generate supplementary resource information recommended to students. The supplementary resource information is either first recommended supplementary information generated based on the first format information or second recommended supplementary information generated based on the content of the heterogeneous correlation correction and analysis period.

[0071] The heterogeneous association parsing time period is determined based on the difference between the first association parameter and the second association parameter;

[0072] Step S500: Send the generated supplementary resource information to the student terminal.

[0073] Specifically, the method for determining several explanation and correction feature frames is to acquire images of consecutive frames at preset intervals within the homework explanation video stream, and determine the time when the correction questions in the image are switched as the time when the explanation and correction feature frame is located.

[0074] Those skilled in the art will understand that in a video stream explaining assignments, the switching of grading questions usually signifies a change in teaching content. For example, transitioning from one grading question to another, acquiring continuous frame images at preset intervals is to systematically and comprehensively scan the video content. By utilizing the close connection between grading question switching and teaching content segmentation, the video stream can be reasonably divided according to the changes in actual teaching content.

[0075] In practice, the preset interval for acquiring consecutive frames of images can be set by technicians. To avoid missing the moment when the grading questions are switched due to an excessively long interval, and to avoid excessive image analysis due to an excessively short interval, the preset interval is set to [10, 30], with the unit being seconds. Preferably, the preset interval is set to 15 seconds.

[0076] In practice, the differences in the question text can be identified by using image fusion recognition (OCR) technology on the acquired consecutive frame images, thereby determining the time when the grading question is switched. Image fusion recognition (OCR) technology for identifying question text is an existing technology and will not be elaborated here.

[0077] Specifically, the process of determining the content value coefficient includes:

[0078] The time interval between two adjacent explanation and correction feature frames is defined as the correction and parsing time interval;

[0079] The duration of each grading and analysis period is determined, and the ratio of the duration to the total duration of the assignment explanation video stream is determined as the content value coefficient.

[0080] Understandably, the content value coefficient is determined by the ratio of the duration of the grading and analysis session to the total duration of the homework explanation video stream. The principle behind this is that the time allocation can, to some extent, reflect the relative importance of the teaching content within that session in the overall teaching process. If a grading and analysis session is relatively long, it indicates that the teacher has invested more time in explaining the content during that session, and the teaching content in that session is likely to make a significant contribution to achieving the overall teaching objectives. This quantification method transforms the time factor in the teaching process into a measurable numerical value.

[0081] Specifically, please refer to Figure 2 As shown, this is a flowchart illustrating the logic for determining the grading content tendency category of the homework explanation video stream. The process for determining the grading content tendency category of the homework explanation video stream includes:

[0082] Calculate the standard deviation of the content value coefficient for each graded analysis period;

[0083] Based on the comparison results of the standard deviation satisfying the video category differentiation conditions, the grading content tendency category of the homework explanation video stream is determined to be the grading tendency video category;

[0084] Based on the comparison results where the standard deviation does not meet the video category distinction criteria, the grading content of the homework explanation video stream is determined to be a non-grading video category.

[0085] The video category differentiation condition is that the standard deviation exceeds a preset standard deviation comparison value.

[0086] In practice, the standard deviation comparison value needs to be set to filter out the batch parsing period with a significantly longer duration. The standard deviation comparison value ranges from [0.07, 0.1]. Here, we provide an optimal value for the standard deviation comparison value, setting it to 0.08.

[0087] It is understandable that the larger the standard deviation of the content value coefficient of each grading and analysis period, the greater the difference in the content value coefficient of each grading and analysis period, that is, the more outstanding the content contribution of certain periods. Conversely, the smaller the standard deviation, the closer the content value coefficient of each grading and analysis period, which means that the teaching content is relatively balanced in terms of time distribution. By effectively classifying the explanation videos with different teaching characteristics, a basis is provided for the subsequent formulation of targeted teaching resource allocation strategies.

[0088] Specifically, this invention acquires continuous frame images at preset intervals and determines the switching time of the grading questions as the explanation and grading feature frame. This method can clearly and accurately divide the homework explanation video stream into different grading and analysis periods. By determining the time period between adjacent explanation and grading feature frames as the grading and analysis period, and using the ratio of its duration to the total video duration as the content value coefficient, a quantitative assessment of the importance of each grading and analysis period in the entire homework grading process is achieved. By calculating the standard deviation of the content value coefficient of each grading and analysis period, and determining the grading content tendency category based on the relationship between the standard deviation and the preset standard deviation, it is helpful to adopt more targeted teaching resource allocation strategies for different types of explanation videos.

[0089] Specifically, the evaluation factors for the assignment explanation video stream are determined as follows:

[0090] If the homework explanation video stream is a grading-oriented video category, then the evaluation factor is determined to be the correlation between the first format information and the second format information within the characteristic grading analysis period;

[0091] If the homework explanation video stream is a non-biased grading video category, then the evaluation factor is determined to be the correlation coefficient between the first correlation parameter and the second correlation parameter during the grading and analysis period.

[0092] Specifically, please refer to Figure 3 The diagram illustrates the steps of determining the correlation between first format information and second format information according to an embodiment of the present invention. The process of determining the correlation between the first format information and the second format information includes:

[0093] Step S301: Filter the grading and analysis time period corresponding to the maximum content value coefficient in the assignment explanation video stream;

[0094] Step S302: Obtain the voice information within the parsing period and determine the voice keywords based on semantic analysis technology, and determine the voice keywords as the first format information;

[0095] Step S303: Based on the temporal relationship, obtain the image corresponding to the explanation and correction feature frame of the correction and analysis period, determine the keyword text of the newly added text in the image corresponding to the explanation and correction feature frame before and after the time sequence according to the text recognition technology, and determine the keyword text as the second format information;

[0096] Step S304: Determine the Euclidean distance between the first format information and the second format information as the correlation.

[0097] Specifically, in the process of acquiring voice information during the grading and analysis period and determining voice keywords based on semantic analysis technology, audio acquisition equipment, such as high-quality microphones, can be deployed in appropriate locations in the classroom to ensure clear acquisition of the teacher's voice during the grading and analysis period. With the help of existing mature speech recognition engines, the stored audio files of the grading and analysis period are input into the speech recognition engine. The engine uses acoustic and language models to convert the voice signal into text. Using semantic analysis technology tools in natural language processing, the PageRank value of node words is calculated, and words with high PageRank values ​​are selected as voice keywords. These keywords can accurately reflect the core content of the teacher's explanation during the grading and analysis period.

[0098] Specifically, text recognition technology and tools can be used to sequentially input the images corresponding to each saved explanation and correction feature frame into the text recognition tool. The tool converts the text in the image into text form through steps such as image preprocessing, character segmentation, and character recognition. Based on the temporal relationship, the text corresponding to adjacent explanation and correction feature frames is compared, and a string matching algorithm is used to calculate the difference between the two texts. By calculating the similarity or difference of the texts, the newly added text content in the later explanation and correction feature frame image is found. Natural language processing is performed on the identified newly added text to extract keyword text that can accurately summarize its core content. This is existing technology and will not be elaborated here.

[0099] In practice, the first format information and the second format information are vectorized respectively. The two transformed vectors are then substituted into the Euclidean distance formula to calculate the Euclidean distance between the first format information and the second format information. The Euclidean distance formula is widely used in semantic recognition analysis, and will not be elaborated here.

[0100] For example, the audio information of a teacher explaining the concept of "quadratic functions" is acquired through a recording device. Assuming the audio content includes the keywords "quadratic function," "image opening direction," and "vertex coordinate formula," a word embedding model is used to map these keywords into a fixed-dimensional vector space. These keyword vectors are then weighted and averaged to obtain a comprehensive vector of the first format information. Newly added text content is extracted from the image of the graded questions using text recognition technology. If, during the explanation of "quadratic functions," the graded questions added the keywords "quadratic function expression," "axis of symmetry position," and "function monotonicity," the word embedding model is again used to map these keywords into a three-dimensional vector. These keyword vectors are then weighted and averaged to obtain a comprehensive vector of the second format information. The Euclidean distance formula is used to calculate the distance between the two vectors. The vectorization of text information is a prior art technique and will not be elaborated upon here.

[0101] Understandably, semantic analysis technology can understand natural language. By processing the speech information during the correction and analysis period, it converts the continuous speech stream into text and extracts keywords that reflect the core content to form the first format information, representing the key points of the teacher's oral explanation. For the explanation and correction feature frame image, text recognition technology can detect and recognize the text content in the image. Based on the temporal relationship, it obtains the explanation and correction feature frame images before and after, and extracts the keyword text of the newly added text as the second format information. This reflects the key points of updating written content in the teaching process.

[0102] In this invention, the first and second format information are treated as two vectors. Their relevance is quantified by calculating their Euclidean distance. From a teaching perspective, if the teacher's explanation is closely related to the written content, the corresponding audio keywords and newly added text keywords will be semantically and logically similar, resulting in a smaller Euclidean distance value, indicating high relevance. Conversely, if the two differ significantly, the Euclidean distance value will be large, indicating low relevance. This quantification method provides an intuitive way to judge whether the allocation of teaching resources is reasonable.

[0103] Specifically, this invention targets the video grading category and selects the grading analysis period corresponding to the maximum content value coefficient as the key to achieving teaching objectives. By focusing on this core period, it avoids wasting computing power by focusing on too much non-critical content. The invention uses the voice keywords determined by semantic analysis technology during the grading analysis period as the first format information, and the keyword text of newly added text in the explanation and grading feature frame image obtained based on text recognition technology as the second format information. Key information is extracted from these two important dimensions: the teacher's explanation voice and the newly added text of the updated grading content. The voice keywords reflect the key content conveyed orally by the teacher, while the keyword text of the newly added text reflects the key knowledge points presented in written form. This multi-dimensional extraction of key information comprehensively and deeply covers the important content carried by different carriers in the teaching process, realizing the construction of a dual-dimensional evaluation system of explanation and annotation, and achieving the transformation of smart classrooms from experience-driven to data-driven decision-making models.

[0104] Specifically, the process of determining the rationality of the allocation of teaching resources based on the relevance and generating supplementary resource information includes:

[0105] Based on the comparison results where the relevance exceeds a preset relevance threshold, the allocation of teaching resources is determined to be unreasonable; based on the comparison results where the relevance does not exceed a preset relevance threshold, the allocation of teaching resources is determined to be reasonable.

[0106] In response to the result that the configuration is unreasonable based on the relevance determination, the resource supplementary information to be sent to the student is the first recommended supplementary information generated based on the first format information;

[0107] Wherein, the content keywords of the first recommended supplementary information and the Euclidean distance in the first format information do not exceed the preset reference Euclidean distance.

[0108] In practice, the correlation threshold and reference Euclidean distance can be determined based on experimental data. The following are the Euclidean distance experimental data of the first format information and the second format information of 8 sets of experiments.

[0109]

[0110] To ensure that the correlation threshold setting can filter out the grading and parsing periods with smaller Euclidean distances between the first and second format information as periods with unreasonable allocation of teaching resources, the correlation threshold can be set as the average value of the Euclidean distances between the first and second format information in the experimental data, which is 0.64.

[0111] The preset reference Euclidean distance needs to ensure that the keywords of the first recommended supplementary information are highly relevant to the first format information in order to ensure the effectiveness of the supplementary content. Based on this, the value range of the reference Euclidean distance can be set to [0.4, 0.55]. Here, a preferred value of 0.5 for the reference Euclidean distance is provided.

[0112] Specifically, in daily teaching, teachers record homework explanations using multimedia devices, creating a video stream. The system extracts continuous frames at preset intervals, identifies feature frames indicating changes in grading questions, and divides the video into multiple grading and analysis periods. The system determines the content value coefficient by calculating the proportion of each period's duration. If the content value coefficient of a certain period is significantly higher than other periods and its standard deviation exceeds a preset value, it is identified as a grading-biased video category. The system focuses on this period and extracts audio keywords as the first format information, while simultaneously identifying newly added text keywords on the blackboard as the second format information, calculating the Euclidean distance between the two. If the distance exceeds a threshold, the system determines that the allocation of teaching resources is unreasonable.

[0113] Specifically, after determining that the allocation of teaching resources is unreasonable, this invention generates first recommended supplementary information based on first format information. This fully considers students' learning needs for the key content explained by the teacher. The first format information is key information obtained through semantic analysis of the teacher's lecture voice, representing the core knowledge conveyed orally by the teacher. The resource supplementary information generated based on this can accurately supplement any shortcomings that students may have in understanding the teacher's explanation of assignments. Furthermore, it requires that the Euclidean distance between the keywords in the first recommended supplementary information and the first format information does not exceed a preset reference Euclidean distance. This ensures the close relevance between the supplementary information and the teacher's key points, guarantees the consistency of the supplementary resources with the teaching focus in terms of content, enables students to efficiently obtain valuable learning resources, forms a dynamic and adaptive teaching resource ecosystem, and realizes the transformation of the smart classroom from an experience-driven to a data-driven decision-making model.

[0114] Specifically, please refer to Figure 4 The diagram illustrates the steps of determining the association relationship between the first association parameter and the second association parameter according to an embodiment of the present invention. The process of determining the association relationship between the first association parameter and the second association parameter includes:

[0115] Step S311: Based on speech recognition, determine the number of words in the text converted from speech information within each correction and parsing time period, and determine the number of words as the first association parameter;

[0116] Step S312: Based on the temporal relationship, obtain the images corresponding to the explanation and correction feature frames of each correction and analysis period, determine the number of text characters in the images corresponding to the explanation and correction feature frames of the preceding and following time periods according to text recognition technology, and determine the number of text characters as the second association parameter.

[0117] Step S313: Sort the batch parsing time periods in ascending order of the first correlation parameter, and combine the second correlation parameters of each batch parsing time period into a correlation relationship sequence according to the sorting.

[0118] Understandably, the process of converting teachers' speech information into text during each grading and analysis period using speech recognition technology employs acoustic and language models. The acoustic model extracts features and performs pattern recognition on the speech signal, converting it into corresponding phoneme sequences. The language model further converts the phoneme sequences into understandable text based on grammatical and semantic rules. The first correlation parameter is determined by counting the number of words in these converted texts. The principle is that the amount of content explained by the teacher can be quantified to some extent by the number of words in the converted text. The more explanations and analysis, the more words are in the converted text. Therefore, this parameter can intuitively reflect the amount of information conveyed orally by the teacher during each grading and analysis period. Text recognition technology in image recognition is used to detect, segment, and recognize the text in the image, converting the text information in the image into text form. The principle is that the update of grading questions during the teaching process often represents the presentation of new content. The number of newly added text words reflects the amount of change in the written teaching content. This method can provide another quantitative dimension for analyzing the teaching process from the perspective of updating written information. Corresponding to the first correlation parameter, together they constitute a comprehensive quantitative description of the information output of the teaching process.

[0119] Specifically, the process of determining the rationality of the allocation of teaching resources based on correlations and generating supplementary resource information includes:

[0120] If the correlation sequence is not an increasing sequence, the allocation of teaching resources is deemed unreasonable; if the correlation sequence is an increasing sequence, the allocation of teaching resources is deemed reasonable.

[0121] In response to the result of unreasonable configuration determined based on the association relationship, the resource supplement information to be sent to students is the second recommended supplement information generated based on the content of the heterogeneous association grading and parsing period;

[0122] Based on the comparison results of the difference between the first associated parameter and the second associated parameter exceeding the difference comparison value within the batch parsing period, the batch parsing period is determined as the heterogeneous associated batch parsing period, and the difference comparison value is the average of the differences between the first associated parameter and the second associated parameter within several batch parsing periods;

[0123] Wherein, the Euclidean distance between the keywords of the second recommended supplementary information and the content of the heterogeneous correlation parsing period does not exceed the preset reference Euclidean distance.

[0124] In practice, to ensure that the keywords of the second recommended supplementary information are highly relevant to the content of the heterogeneous correlation parsing period within the preset reference Euclidean distance, the reference Euclidean distance can be set to a range of [0.4, 0.55]. Here, a preferred value of 0.5 for the reference Euclidean distance is provided.

[0125] Understandably, a reasonable allocation of teaching resources should ensure that teachers' explanations and written content updates are coordinated. When the correlation sequence shows an increasing trend, it reflects that the oral and written information outputs are synchronized in the teaching process and the allocation of teaching resources is relatively reasonable. If the correlation sequence is not an increasing sequence, it indicates that there are problems in the coordination and allocation of teaching resources between oral and written information.

[0126] In this invention, after determining the heterogeneous association grading and parsing time periods, second recommended supplementary information is generated based on these time periods. The principle is that there are situations where teaching resources are not properly allocated during these time periods. By supplementing relevant learning resources, the deficiencies in the teaching process can be made up for. The Euclidean distance between the content keywords of the second recommended supplementary information and the content of the heterogeneous association grading and parsing time periods meets the preset reference Euclidean distance. The Euclidean distance is used to measure the similarity between the supplementary information keywords and the teaching content of the abnormal time periods in the feature space. When the Euclidean distance meets the preset standard, it indicates that the supplementary information can provide targeted and relevant learning materials to address the problems such as students' difficulty in understanding caused by the unreasonable allocation of teaching resources during that time period, helping students to better understand and master the teaching content of that time period, thereby optimizing the teaching effect.

[0127] Specifically, for grading non-biased video categories, this invention utilizes speech recognition and text recognition technologies to acquire relevant data when determining the correlation. After sorting the first correlation parameter, a correlation sequence is constructed, which can intuitively show the relationship between the amount of teacher explanation and the amount of text updates in homework grading, preventing the impact of configuration imbalance on teaching effectiveness. In terms of generating supplementary resource information, a second recommended supplementary information is generated based on the content of the heterogeneous correlation grading analysis period, accurately locating the periods of abnormal information output during the teaching process. These periods are determined by calculating the average difference of correlation parameters, which can effectively make up for weak links in teaching. At the same time, it requires that the Euclidean distance between the keywords of the supplementary information content and the content of the heterogeneous correlation grading analysis period meets the preset standard, ensuring that the supplementary information is closely related to the teaching content of the problem period, realizing the transformation of the smart classroom from an experience-driven to a data-driven decision-making model.

[0128] Please see Figure 5 The diagram shown is a system block diagram of a teaching quality optimization system based on AI intelligent grading and analysis according to an embodiment of the present invention. The present invention also provides a teaching quality optimization system based on AI intelligent grading and analysis, comprising:

[0129] The information acquisition unit includes a video recorder for acquiring the homework explanation video stream and a microphone for acquiring audio information. The video recorder is also a video processor for dividing the homework explanation video stream into several grading and analysis time periods.

[0130] Specifically, the present invention does not limit the specific structure of the video recorder, which can be a camera that acquires videos of teachers grading homework, and will not be described in detail here.

[0131] Specifically, the present invention does not limit the specific structure of the microphone, as long as it can realize the function of acquiring the teacher's voice information and converting it into a voice file, which will not be elaborated here.

[0132] The feature extraction unit, which is connected to the information acquisition unit, is used to determine the content value coefficient of each grading and analysis period and to determine the grading content tendency category of the homework explanation video stream based on the differences in the content value coefficients.

[0133] Specifically, the present invention does not limit the feature extraction unit. Preferably, the feature extraction unit can be a microcomputer processor, which will not be elaborated here.

[0134] An intelligent evaluation unit, which is connected to the information acquisition unit and the feature extraction unit respectively, is used to determine the evaluation factors of the homework explanation video stream based on the tendency category of the grading content.

[0135] Specifically, the intelligent evaluation unit in this invention can be a data storage device and a processor connected to the data storage device. The processor calls up the data in the data storage device according to the execution instructions, which will not be elaborated here.

[0136] The content recommendation generation unit, which is connected to the intelligent evaluation unit, is used to determine whether the allocation of teaching resources is reasonable based on the evaluation factors, and to generate supplementary resource information to recommend to students.

[0137] Specifically, the content recommendation generation unit in this invention can be an information processor connected to a network, which is used to connect to the network and filter online teaching resources or local teaching resources that meet the conditions according to the executed instructions, which will not be elaborated here.

[0138] A transmission unit, which is connected to the content recommendation generation unit, is used to send the generated supplementary resource information to the student terminal;

[0139] Specifically, the present invention does not limit the transmission unit. Preferably, it can be a data transmitter for sending resource supplement information to the terminal, which will not be elaborated here.

[0140] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0141] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. 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 teaching quality optimization method based on AI-powered intelligent grading and analysis, characterized in that, include: The teacher obtains a video stream of homework explanation during the teaching process, identifies several explanation and correction feature frames in the homework explanation video stream, and divides the homework explanation video stream into several correction and analysis time periods based on the explanation and correction feature frames. The content value coefficient of each grading and analysis period is determined based on the duration of the grading and analysis period, and the grading content tendency category of the homework explanation video stream is determined based on the differences in the content value coefficients. The evaluation factors for the homework explanation video stream are determined based on the category of the grading content. The evaluation factors include the correlation between the first format information and the second format information within the characteristic grading and analysis period, and the correlation between the first correlation parameter and the second correlation parameter within the grading and analysis period. The feature-based parsing period is determined based on the content value coefficient. Based on the evaluation factors, it is determined whether the allocation of teaching resources is reasonable. In response to the determination that the allocation of teaching resources is unreasonable, supplementary resource information recommended to students is generated. The supplementary resource information is either first recommended supplementary information generated based on the first format information, or second recommended supplementary information generated based on the content of the heterogeneous correlation correction and analysis period. The heterogeneous association parsing time period is determined based on the difference between the first association parameter and the second association parameter; The generated supplementary resource information will be sent to the student's terminal.

2. The teaching quality optimization method based on AI intelligent grading and analysis according to claim 1, characterized in that, The method for determining several explanation and correction feature frames is to acquire images of consecutive frames at preset intervals within the homework explanation video stream, and determine the time when the correction questions in the image are switched as the time when the explanation and correction feature frame is located.

3. The teaching quality optimization method based on AI intelligent grading and analysis according to claim 2, characterized in that, The process of determining the content value coefficient includes: The time interval between two adjacent explanation and correction feature frames is defined as the correction and parsing time interval; The duration of each grading and analysis period is determined, and the ratio of the duration to the total duration of the homework explanation video stream is determined as the content value coefficient.

4. The teaching quality optimization method based on AI intelligent grading and analysis according to claim 3, characterized in that, The process of determining the categorization of the grading content in the assignment explanation video stream includes: Calculate the standard deviation of the content value coefficient for each graded analysis period; Based on the comparison results of the standard deviation satisfying the video category differentiation conditions, the grading content tendency category of the homework explanation video stream is determined to be the grading tendency video category; Based on the comparison results where the standard deviation does not meet the video category distinction criteria, the grading content of the homework explanation video stream is determined to be a grading non-biased video category. The video category differentiation condition is that the standard deviation exceeds a preset standard deviation comparison value.

5. The teaching quality optimization method based on AI intelligent grading and analysis according to claim 4, characterized in that, The evaluation factors for the assignment explanation video stream are determined as follows: If the homework explanation video stream is a grading-oriented video category, then the evaluation factor is determined to be the correlation between the first format information and the second format information within the feature grading analysis period; If the homework explanation video stream is a non-biased grading video category, then the evaluation factor is determined to be the correlation coefficient between the first correlation parameter and the second correlation parameter during the grading and analysis period.

6. The teaching quality optimization method based on AI intelligent grading and analysis according to claim 5, characterized in that, The process of determining the relevance between the first-format information and the second-format information includes: Filter the grading and analysis time period corresponding to the maximum content value coefficient in the assignment explanation video stream; Acquire the voice information within the correction and parsing period and determine the voice keywords based on semantic analysis technology, and determine the voice keywords as the first format information; Based on the temporal relationship, the image corresponding to the explanation and correction feature frame of the correction and analysis period is obtained. Based on the text recognition technology, the keyword text of the newly added text in the image corresponding to the explanation and correction feature frame before and after the time sequence is determined, and the keyword text is determined as the second format information. The Euclidean distance between the first format information and the second format information is determined as the correlation.

7. The teaching quality optimization method based on AI intelligent grading and analysis according to claim 6, characterized in that, The process of determining the rationality of the allocation of teaching resources based on the relevance and generating supplementary resource information includes: Based on the comparison results of the relevance exceeding the preset relevance threshold, it is determined that the allocation of teaching resources is unreasonable; In response to the result that the configuration is unreasonable based on the relevance determination, the resource supplementary information to be sent to the student is the first recommended supplementary information generated based on the first format information; Wherein, the content keywords of the first recommended supplementary information and the Euclidean distance in the first format information do not exceed the preset reference Euclidean distance.

8. The teaching quality optimization method based on AI intelligent grading and analysis according to claim 5, characterized in that, The process of determining the association relationship between the first association parameter and the second association parameter includes: Based on speech recognition, the number of words in the text converted from speech information within each batching and parsing period is determined, and the number of words is determined as the first association parameter; Based on the temporal relationship, the images corresponding to the explanation and correction feature frames of each correction and analysis period are obtained. The number of text characters in the images corresponding to the explanation and correction feature frames of the preceding and following time periods are determined according to the text recognition technology. The number of text characters is then determined as the second association parameter. The parsing time periods are sorted in ascending order of the first correlation parameter, and the second correlation parameters of each parsing time period are combined into a correlation relationship sequence according to the sorting.

9. The teaching quality optimization method based on AI intelligent grading and analysis according to claim 8, characterized in that, The process of determining the rationality of the allocation of teaching resources based on correlation and generating supplementary resource information includes: If the sequence of relationships is not an increasing sequence, then the allocation of teaching resources is deemed unreasonable. In response to the result of unreasonable configuration determined based on the association relationship, the resource supplement information to be sent to students is the second recommended supplement information generated based on the content of the heterogeneous association grading and parsing period; Based on the comparison results of the difference between the first associated parameter and the second associated parameter exceeding the difference comparison value within the batch parsing period, the batch parsing period is determined as the heterogeneous associated batch parsing period, and the difference comparison value is the average of the differences between the first associated parameter and the second associated parameter within several batch parsing periods; Wherein, the Euclidean distance between the keywords of the second recommended supplementary information and the content of the heterogeneous correlation parsing period does not exceed the preset reference Euclidean distance.

10. A teaching quality optimization system based on AI intelligent grading and analysis, used to execute the teaching quality optimization method based on AI intelligent grading and analysis as described in any one of claims 1-9, characterized in that, include: The information acquisition unit includes a video recorder for acquiring the homework explanation video stream and a microphone for acquiring audio information. The video recorder is also a video processor for dividing the homework explanation video stream into several grading and analysis time periods. The feature extraction unit, which is connected to the information acquisition unit, is used to determine the content value coefficient of each grading and analysis period and to determine the grading content tendency category of the homework explanation video stream based on the differences in the content value coefficients. An intelligent evaluation unit, which is connected to the information acquisition unit and the feature extraction unit respectively, is used to determine the evaluation factors of the homework explanation video stream based on the tendency category of the grading content. The content recommendation generation unit, which is connected to the intelligent evaluation unit, is used to determine whether the allocation of teaching resources is reasonable based on the evaluation factors, and to generate supplementary resource information to recommend to students. A transmission unit, connected to the content recommendation generation unit, is used to send the generated supplementary resource information to the student terminal.

Images