An online education level assessment system and method based on cloud server

Abstract

The present invention discloses an online education level evaluation system and method based on a cloud server, including: obtaining the course learning monitoring information of the target user, analyzing the user's learning progress according to the course learning monitoring information, and obtaining the learning progress analysis information; setting the evaluation index, evaluating the learning status of the target user, and evaluating the user's course mastery; obtaining the user's learning monitoring information, analyzing the user's emotional state during learning according to the user's learning monitoring information, and obtaining the emotional state analysis information; obtaining the course content corresponding to the user's doubt state according to the emotional state analysis information, and analyzing the user's doubt reasons in combination with the learning status evaluation information; analyzing whether the teaching content needs to be adjusted according to the user's doubt reason analysis information, and formulating an optimization plan. Comprehensively understand the reasons for students' doubts, analyze whether the teaching difficulty of the teaching content is too great, so as to teach students in accordance with their aptitude, conduct targeted teaching, and improve the teaching quality and teaching level.

Description

Technical Field

[0001] The present invention relates to the technical field of education level assessment, and in particular to an online education level assessment system and method based on a cloud server. Background Art

[0002] With the rapid development of Internet technology, online education has become one of the important development directions in the field of education. Online education platforms provide students with a more flexible and convenient way of learning, allowing learners to study according to their personal needs and schedules. The emergence of cloud servers provides strong technical support for online education. Through cloud computing technology, online education platforms can achieve centralized storage and efficient distribution of teaching resources, ensuring the stability and availability of teaching. At the same time, cloud computing technology also facilitates the collection, processing and analysis of data, enabling online education platforms to better understand students' learning needs and behavior patterns, and providing the possibility for personalized teaching.

[0003] Traditional education assessment methods mainly rely on classroom examinations, homework assessments, etc. Although they can evaluate students to a certain extent, they cannot provide a deeper understanding of the reasons for students' weak knowledge points, and cannot determine whether the current teaching content is too in-depth or the current teaching difficulty of the class is not suitable for students, which leads to a mismatch between students' actual learning situation and teaching content, resulting in students not being able to keep up with the learning progress, resulting in low teaching quality. Therefore, how to better understand the relationship between students and teaching content is an important issue. Summary of the invention

[0004] The present invention overcomes the defects of the prior art and provides an online education level evaluation system and method based on a cloud server, the important purpose of which is to improve the teaching quality and improve and optimize online education.

[0005] To achieve the above objectives, the present invention provides a first aspect of a cloud server-based online education level assessment method, comprising:

[0006] Obtaining the target user's course learning monitoring information, and performing user learning progress analysis based on the course learning monitoring information to obtain learning progress analysis information;

[0007] Set evaluation indicators, evaluate the learning status of target users, evaluate the user's course mastery, and obtain learning status evaluation information;

[0008] Acquire user learning monitoring information, analyze the user's emotional state during learning according to the user learning monitoring information, and obtain emotional state analysis information;

[0009] Acquire the course content corresponding to the user's doubt state according to the emotional state analysis information, analyze the cause of the user's doubt in combination with the learning status evaluation information, and obtain the user's doubt cause analysis information;

[0010] Based on the user's doubt cause analysis information, it is determined whether the teaching content needs to be adjusted and an optimization plan is developed.

[0011] In this solution, the acquisition of the target user's course learning monitoring information and the analysis of the user's learning progress based on the course learning monitoring information specifically include:

[0012] Obtaining the target user's course learning monitoring information, performing data cleaning on the course learning monitoring information, extracting the category features corresponding to each data, and combining the clustering algorithm to sort the data to obtain preprocessing information;

[0013] Performing feature extraction on the preprocessed information to extract the course progress features of the target user, including course category, course name, course viewing status, and homework completion status, to obtain course progress feature information;

[0014] Perform course learning situation analysis based on the course progress characteristic information, and analyze the target user's learning time, course viewing progress, and viewing times of each course to obtain course learning situation analysis information;

[0015] Extract the course assignments corresponding to the courses that the target user has studied, calculate the target user's course assignment completion rate based on the course feature information, perform course assignment completion analysis, and obtain course assignment completion analysis information;

[0016] Extracting the learning completion time of each course of the user according to the course progress feature information, and extracting the expected completion time of the user's current learning stage based on the course learning situation analysis information, performing a learning cycle analysis on the target user, judging whether it meets the expected consumption time, and obtaining the user's learning cycle analysis information;

[0017] The learning progress evaluation weight is preset, and a weighted evaluation is performed in combination with the course learning situation analysis information, the course homework completion situation analysis information and the user learning cycle analysis information to obtain the learning progress analysis information.

[0018] In this solution, the evaluation indicators are set to evaluate the learning status of the target user, evaluate the user's course mastery, and obtain learning status evaluation information, which specifically includes:

[0019] Based on big data retrieval, learning status evaluation indicators are obtained, the Pearson correlation coefficient between each evaluation indicator and the learning status is calculated to represent the degree of correlation, and the evaluation indicators are screened by judging with the preset threshold, and the Pearson correlation coefficient matrix is ​​constructed according to the screening results;

[0020] Calculate the Euclidean distance between the selected evaluation indicators to form a distance matrix, convert the distance matrix into an adjacency matrix using a Gaussian kernel function, and calculate the vertex matrix;

[0021] Subtracting the vertex matrix from the adjacency matrix to obtain a Laplacian matrix, performing eigendecomposition on the Laplacian matrix, presetting an expected number of clusters M, and performing clustering using a K-means algorithm;

[0022] The evaluation indicators are set according to the clustering results, and the evaluation indicators are graded according to the Pearson correlation coefficients corresponding to the evaluation indicators in each cluster to obtain the evaluation indicator information;

[0023] The entropy weight method is used to calculate the information entropy of each evaluation indicator, and the information gain rate of the corresponding evaluation indicator is calculated by the calculated information entropy. The weight of each evaluation indicator is assigned to obtain the evaluation weight information;

[0024] A learning status evaluation model is constructed based on the evaluation indicator information and the evaluation weight information, and course learning monitoring information and learning progress analysis information are obtained and input into the learning status evaluation model for evaluation to obtain learning status evaluation information.

[0025] In this solution, the acquiring of user learning monitoring information, analyzing the user's emotional state during learning according to the user learning monitoring information, and obtaining emotional state analysis information specifically includes:

[0026] Based on big data retrieval, image information corresponding to various emotional states during learning is obtained, image preprocessing is performed, and classification is performed according to emotional states to form an instance data set;

[0027] Performing feature extraction on the image information corresponding to each emotional state in the example data set, extracting facial features and body features, and obtaining first feature information;

[0028] Extract the occurrence frequency of each feature in each image in the corresponding emotional state according to the first feature information as a correlation index, make a judgment with a preset threshold, select features greater than the preset threshold as strong correlation features, and obtain correlation degree analysis information;

[0029] Constructing an emotional state analysis model based on CNN, constructing a training data set according to the correlation analysis information and the example data set to perform deep learning and training on the emotional state analysis model, and obtaining an emotional state analysis model that meets expectations;

[0030] Acquiring user learning monitoring information, the user learning monitoring information including face monitoring information and body monitoring information, performing wavelet transform on the user learning monitoring information, and decomposing the target image into sub-bands of different frequency bands;

[0031] Preset a processing threshold, perform threshold processing on each sub-band, set the wavelet coefficients smaller than the processing threshold to zero, extract high-frequency wavelet coefficients and perform linear enhancement, merge coefficients according to the enhanced high-frequency wavelet coefficients, perform inverse transformation on the merged wavelet coefficients, and obtain enhanced image information;

[0032] The enhanced image information is input into the emotional state analysis model for analysis, the emotional changes of the user during learning are analyzed, and the information is associated with the corresponding time to obtain emotional state analysis information.

[0033] In this solution, obtaining the course content corresponding to the user's doubt state according to the emotional state analysis information and analyzing the cause of the user's doubt in combination with the learning status evaluation information specifically includes:

[0034] Acquire emotional state analysis information, construct an emotional change trend graph according to the emotional state analysis information, and extract a corresponding doubtful emotional time period according to the emotional change trend graph;

[0035] Extract the emotional state of neighboring time periods of the doubtful emotional time period according to the emotional change trend graph, perform user emotional transition analysis, and obtain emotional transition analysis information;

[0036] Based on the extracted doubt emotion time period, the user viewing content of the corresponding time period is extracted and marked as doubt course content, thereby obtaining doubt course content information;

[0037] Acquire learning status evaluation information, extract knowledge point features of target course content according to the doubtful course content information, acquire user mastery of corresponding knowledge point features through the learning status evaluation information, and obtain user mastery analysis information;

[0038] The user learning frequency, score and knowledge point error rate of the target knowledge point are extracted based on the user mastery analysis information, and the user doubt cause analysis is performed in combination with the emotion change analysis information to obtain user doubt cause analysis information.

[0039] In this solution, the analysis of whether the teaching content needs to be adjusted based on the user's doubt analysis information and the formulation of an optimization plan specifically include:

[0040] Obtain the analysis information of the reasons for user doubts, extract the corresponding doubt knowledge point information, count the number of target users in the current teaching class who have doubts about the target knowledge point, conduct a universal analysis of the reasons for doubts, and obtain universal analysis information of the reasons for doubts;

[0041] According to the universal analysis information of the doubt reasons, the course content adaptation is determined, and the adaptability of the teaching method of the current course content and the current teaching class is analyzed to obtain adaptability analysis information;

[0042] If the adaptability analysis information determines that the teaching method of the current course content is not suitable for the users of the current teaching class, it means that the current teaching content needs to be adjusted;

[0043] Based on the universality analysis information of the doubt reasons, the number of users who doubt the target course content is extracted, and compared with the preset threshold, the influence of the current course content on the users in the target teaching class is analyzed to obtain the influence analysis information;

[0044] Retrieve the teaching plan for the corresponding knowledge point based on the current education platform, extract the teaching difficulty of each teaching plan, match it with the impact analysis information, formulate an adjustment plan based on the matching result, and adjust the teaching content of the current teaching class;

[0045] If the adaptability analysis information determines that the teaching method of the current course content is suitable for the users in the current teaching class, then an adaptability evaluation is performed on the target users;

[0046] Preset user adaptation evaluation weights, obtain learning status evaluation information, extract course learning features corresponding to doubtful knowledge points as completion features and learning cycle features, conduct adaptation evaluation on target users, and obtain user adaptation evaluation information;

[0047] Based on the user adaptation evaluation information, it is determined whether the user meets the teaching difficulty of the current teaching class. If the user meets the teaching difficulty, other teaching plans are obtained according to the target user's doubtful knowledge points as optimization plans and pushed to the user for knowledge point reinforcement;

[0048] If it does not match, it means that the teaching difficulty of the current teaching class is not suitable for the target user, then the teaching classes of other teaching difficulties are obtained to generate a learning suggestion report and push it to the target user.

[0049] A second aspect of the present invention provides an online education level assessment system based on a cloud server, the system comprising: a memory, a processor, the memory containing an online education level assessment method program based on a cloud server, the online education level assessment method program based on a cloud server when executed by the processor implements the following steps:

[0050] Obtaining the target user's course learning monitoring information, and performing user learning progress analysis based on the course learning monitoring information to obtain learning progress analysis information;

[0051] Set evaluation indicators, evaluate the learning status of target users, evaluate the user's course mastery, and obtain learning status evaluation information;

[0052] Acquire user learning monitoring information, analyze the user's emotional state during learning according to the user learning monitoring information, and obtain emotional state analysis information;

[0053] Acquire the course content corresponding to the user's doubt state according to the emotional state analysis information, analyze the cause of the user's doubt in combination with the learning status evaluation information, and obtain the user's doubt cause analysis information;

[0054] Based on the user's doubt cause analysis information, it is determined whether the teaching content needs to be adjusted and an optimization plan is developed.

[0055] The present invention discloses an online education level evaluation system and method based on a cloud server, including: obtaining the course learning monitoring information of the target user, analyzing the user's learning progress according to the course learning monitoring information, and obtaining the learning progress analysis information; setting the evaluation index, evaluating the learning status of the target user, and evaluating the user's course mastery; obtaining the user's learning monitoring information, analyzing the user's emotional state during learning according to the user's learning monitoring information, and obtaining the emotional state analysis information; obtaining the course content corresponding to the user's doubt state according to the emotional state analysis information, and analyzing the user's doubt reasons in combination with the learning status evaluation information; analyzing whether the teaching content needs to be adjusted according to the user's doubt reason analysis information, and formulating an optimization plan. Comprehensively understand the reasons for students' doubts, analyze whether the teaching difficulty of the teaching content is too great, so as to teach students in accordance with their aptitude, conduct targeted teaching, and improve the teaching quality and teaching level. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions in the embodiments or exemplary embodiments of the present invention, the drawings required for use in the embodiments or exemplary descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained according to the drawings without paying creative work.

[0057] Figure 1 A flow chart of a cloud server-based online education level assessment method provided in one embodiment of the present invention;

[0058] Figure 2 A flow chart of teaching content optimization provided by an embodiment of the present invention;

[0059] Figure 3 A block diagram of an online education level assessment system based on a cloud server provided in one embodiment of the present invention;

[0060] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0061] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0062] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0063] Figure 1 A flow chart of a cloud server-based online education level assessment method provided in one embodiment of the present invention;

[0064] like Figure 1 As shown, the present invention provides a flow chart of an online education level assessment method based on a cloud server, comprising:

[0065] S102, obtaining the course learning monitoring information of the target user, and performing the user learning progress analysis according to the course learning monitoring information to obtain the learning progress analysis information;

[0066] S104, setting evaluation indicators, evaluating the learning status of the target user, evaluating the user's course mastery, and obtaining learning status evaluation information;

[0067] S106, obtaining user learning monitoring information, and analyzing the user's emotional state during learning according to the user learning monitoring information to obtain emotional state analysis information;

[0068] S108, obtaining course content corresponding to the user's doubt state according to the emotional state analysis information, analyzing the cause of the user's doubt in combination with the learning status evaluation information, and obtaining analysis information of the cause of the user's doubt;

[0069] S110, analyzing whether it is necessary to adjust the teaching content based on the user's doubt cause analysis information, and formulating an optimization plan.

[0070] It should be noted that the present invention provides an online education evaluation system and method based on a cloud server. First, the user's learning progress is analyzed through the user's information during the course learning process, including learning time, viewing content, completed exercises, etc., to determine whether it meets the expected progress. At the same time, by setting evaluation indicators, the user's course mastery is comprehensively evaluated, such as answering speed, accuracy, and knowledge point understanding, so as to understand the user's knowledge point mastery in detail. In addition, according to the user's emotional state during the learning process, especially when the user shows doubts, the course content corresponding to the doubt state is analyzed, and combined with the user's learning status evaluation information, the cause of the doubt is deeply excavated, and it is determined whether it is a problem with the teaching content, and further whether the teaching content needs to be adjusted, so as to adjust the teaching content, including the adjustment of the course structure, the re-explanation of knowledge points, the addition of auxiliary materials, etc., so as to teach students in accordance with their aptitude, improve the user's learning experience, and optimize the quality and level of online education.

[0071] Furthermore, in a preferred embodiment of the present invention, the step of obtaining the target user's course learning monitoring information and analyzing the user's learning progress according to the course learning monitoring information specifically includes:

[0072] Obtaining the target user's course learning monitoring information, performing data cleaning on the course learning monitoring information, extracting the category features corresponding to each data, and combining the clustering algorithm to sort the data to obtain preprocessing information;

[0073] Performing feature extraction on the preprocessed information to extract the course progress features of the target user, including course category, course name, course viewing status, and homework completion status, to obtain course progress feature information;

[0074] Perform course learning situation analysis based on the course progress characteristic information, and analyze the target user's learning time, course viewing progress, and viewing times of each course to obtain course learning situation analysis information;

[0075] Extract the course assignments corresponding to the courses that the target user has studied, calculate the target user's course assignment completion rate based on the course feature information, perform course assignment completion analysis, and obtain course assignment completion analysis information;

[0076] Extracting the learning completion time of each course of the user according to the course progress feature information, and extracting the expected completion time of the user's current learning stage based on the course learning situation analysis information, performing a learning cycle analysis on the target user, judging whether it meets the expected consumption time, and obtaining the user's learning cycle analysis information;

[0077] The learning progress evaluation weight is preset, and a weighted evaluation is performed in combination with the course learning situation analysis information, the course homework completion situation analysis information and the user learning cycle analysis information to obtain the learning progress analysis information.

[0078] It should be noted that when conducting an educational level assessment, the user's learning progress reflects the user's learning of the teaching content. By obtaining the user's course learning monitoring information, including learning time, number of learning times and grades, etc., the user can be analyzed in detail. The user's learning of the course is analyzed based on the user's learning time, number of learning times and viewing progress of the course. The user's homework completion status is analyzed based on the user's homework completion rate. The length of time the user spends studying the course is analyzed to determine whether the user has completed the required courses within the expected time, thereby determining whether the user's learning progress is normal. It not only helps users understand their own learning progress and status, but also provides educators and platform parties with an important basis for optimizing teaching content and improving learning experience.

[0079] Further, in a preferred embodiment of the present invention, the setting of evaluation indicators, conducting a learning status evaluation on the target user, evaluating the user's course mastery, and obtaining learning status evaluation information specifically includes:

[0080] Based on big data retrieval, learning status evaluation indicators are obtained, the Pearson correlation coefficient between each evaluation indicator and the learning status is calculated to represent the degree of correlation, and the evaluation indicators are screened by judging with the preset threshold, and the Pearson correlation coefficient matrix is ​​constructed according to the screening results;

[0081] Calculate the Euclidean distance between the selected evaluation indicators to form a distance matrix, convert the distance matrix into an adjacency matrix using a Gaussian kernel function, and calculate the vertex matrix;

[0082] Subtracting the vertex matrix from the adjacency matrix to obtain a Laplacian matrix, performing eigendecomposition on the Laplacian matrix, presetting an expected number of clusters M, and performing clustering using a K-means algorithm;

[0083] The evaluation indicators are set according to the clustering results, and the evaluation indicators are graded according to the Pearson correlation coefficients corresponding to the evaluation indicators in each cluster to obtain the evaluation indicator information;

[0084] The entropy weight method is used to calculate the information entropy of each evaluation indicator, and the information gain rate of the corresponding evaluation indicator is calculated by the calculated information entropy. The weight of each evaluation indicator is assigned to obtain the evaluation weight information;

[0085] A learning status evaluation model is constructed based on the evaluation indicator information and the evaluation weight information, and course learning monitoring information and learning progress analysis information are obtained and input into the learning status evaluation model for evaluation to obtain learning status evaluation information.

[0086] It should be noted that the learning status assessment is a detailed analysis of the target users, an in-depth analysis of the users' mastery of the teaching content, so as to have a deeper understanding of the users' situation. First, the evaluation indicators related to the learning status are obtained through big data retrieval technology. Including the user's learning time, the correct rate of answering questions, the course completion and other aspects. Then, the Pearson correlation coefficient between each evaluation indicator and the learning status is calculated to quantify the degree of correlation between the evaluation indicator and the learning status. By presetting the threshold, the evaluation indicators that are highly correlated with the learning status are screened out, and the Pearson correlation coefficient matrix is ​​constructed accordingly. Then, the Euclidean distance between the selected evaluation indicators is calculated to form a distance matrix. Using the Gaussian kernel function, the distance matrix is ​​converted into an adjacency matrix, and the vertex matrix is ​​calculated. Through the method of graph theory, the internal connection and mutual influence between the evaluation indicators are further revealed. Next, the vertex matrix is ​​subtracted from the adjacency matrix to obtain the Laplace matrix. The evaluation indicators are clustered by performing eigendecomposition on the Laplace matrix using the K-means algorithm. According to the Pearson correlation coefficient corresponding to the evaluation indicators in each cluster, each evaluation indicator is graded. Based on the correlation of each evaluation indicator, the level of each evaluation indicator is judged and divided into levels, such as primary indicators, secondary indicators, etc. Then, the entropy weight method is used to assign weights. Information entropy reflects the amount of information contained in the evaluation indicator, while the information gain rate measures the contribution of the evaluation indicator to the model. Finally, based on the evaluation indicator information and evaluation weight information, a learning status evaluation model is constructed. A comprehensive and objective evaluation of the user's learning status is performed.

[0087] Further, in a preferred embodiment of the present invention, the acquiring of user learning monitoring information, analyzing the user's emotional state during learning according to the user learning monitoring information, and obtaining emotional state analysis information specifically includes:

[0088] Based on big data retrieval, image information corresponding to various emotional states during learning is obtained, image preprocessing is performed, and classification is performed according to emotional states to form an instance data set;

[0089] Performing feature extraction on the image information corresponding to each emotional state in the example data set, extracting facial features and body features, and obtaining first feature information;

[0090] Extract the occurrence frequency of each feature in each image in the corresponding emotional state according to the first feature information as a correlation index, make a judgment with a preset threshold, select features greater than the preset threshold as strong correlation features, and obtain correlation degree analysis information;

[0091] Constructing an emotional state analysis model based on CNN, constructing a training data set according to the correlation analysis information and the example data set to perform deep learning and training on the emotional state analysis model, and obtaining an emotional state analysis model that meets expectations;

[0092] Acquiring user learning monitoring information, the user learning monitoring information including face monitoring information and body monitoring information, performing wavelet transform on the user learning monitoring information, and decomposing the target image into sub-bands of different frequency bands;

[0093] Preset a processing threshold, perform threshold processing on each sub-band, set the wavelet coefficients smaller than the processing threshold to zero, extract high-frequency wavelet coefficients and perform linear enhancement, merge coefficients according to the enhanced high-frequency wavelet coefficients, perform inverse transformation on the merged wavelet coefficients, and obtain enhanced image information;

[0094] The enhanced image information is input into the emotional state analysis model for analysis, the emotional changes of the user during learning are analyzed, and the information is associated with the corresponding time to obtain emotional state analysis information.

[0095] It should be noted that for the course content that the user is confused about, the changes in the user's facial expressions during class are monitored, so as to judge the different doubts of the user about the course content in a targeted manner, not only judging the course content through course grades, etc., so that the analysis of the user is more comprehensive. Before the emotional state analysis, the image information corresponding to various emotional states during learning is obtained through big data retrieval, and various images are sorted according to the emotional categories to form an instance data set. Then, since the expression of emotional state is not only through facial expression changes, but also the body movements can express the inner emotions to a certain extent, the correlation between the features in various emotional state images and the corresponding emotional states is further analyzed based on these characteristics. For some features, such as frowning, scratching the head, etc., they represent the emotion of doubt to a certain extent. Therefore, according to the frequency of each feature appearing in the corresponding emotional state image, a strong correlation feature is defined, so that when performing real-time analysis, the analysis speed can be accelerated and the analysis accuracy can be improved. Subsequently, an emotional state analysis model is constructed based on CNN, and a training data set is constructed according to the correlation analysis information and the instance data set to perform deep learning and training on the emotional state analysis model. Finally, based on the acquired learning monitoring information, the emotional changes of target users during course learning are analyzed through the emotional state analysis model, so as to better understand the content of users' doubts.

[0096] In addition, in the online education level assessment method based on a cloud server provided by the present invention, the following steps are also included:

[0097] Acquire real-time learning monitoring information, extract the distance between the target user and the screen according to the real-time learning monitoring information, form a distance change trend graph, analyze whether the user is too close to the screen, and obtain first analysis information;

[0098] According to the eyeball features of the target user in each frame of the real-time learning monitoring information, user attention analysis is performed according to the eyeball features of the user, the eyeball features of the target user are used as gaze points, and the eyeball feature change speeds of the k image and the k+1 image are calculated;

[0099] Preset a speed threshold, compare the calculated eye feature change speed with the preset threshold, mark the corresponding feature according to the judgment result, generate a gaze point trajectory and draw a gaze point density map, analyze whether the user's attention is focused through the gaze point density map, and obtain attention analysis information;

[0100] Inputting the real-time learning monitoring information into an emotional state analysis model for analysis to determine whether the user is experiencing fatigue, thereby obtaining second analysis information;

[0101] Combining the first analysis information, the second analysis information and the attention analysis information to determine the abnormal learning state, analyzing whether the user's current learning state is abnormal, and obtaining abnormal learning state analysis information;

[0102] The target user is reminded according to the abnormal learning status analysis information.

[0103] It should be noted that when users are learning online, they are prone to fatigue, distraction or being too close to the screen. By real-time monitoring and analysis of users, reminders can be given when users are in abnormal learning states to improve the quality of their learning.

[0104] Further, in a preferred embodiment of the present invention, obtaining the course content corresponding to the user's doubt state according to the emotional state analysis information and analyzing the cause of the user's doubt in combination with the learning status evaluation information specifically includes:

[0105] Acquire emotional state analysis information, construct an emotional change trend graph according to the emotional state analysis information, and extract a corresponding doubtful emotional time period according to the emotional change trend graph;

[0106] Extract the emotional state of neighboring time periods of the doubtful emotional time period according to the emotional change trend graph, perform user emotional transition analysis, and obtain emotional transition analysis information;

[0107] Based on the extracted doubt emotion time period, the user viewing content of the corresponding time period is extracted and marked as doubt course content, thereby obtaining doubt course content information;

[0108] Acquire learning status evaluation information, extract knowledge point features of target course content according to the doubtful course content information, acquire user mastery of corresponding knowledge point features through the learning status evaluation information, and obtain user mastery analysis information;

[0109] The user learning frequency, score and knowledge point error rate of the target knowledge point are extracted based on the user mastery analysis information, and the user doubt cause analysis is performed in combination with the emotion change analysis information to obtain user doubt cause analysis information.

[0110] It should be noted that by analyzing the emotional state of the target user, the emotional state of the target user during the learning process is obtained, thereby forming an emotional change trend graph, extracting the course content corresponding to the doubtful emotion, marking the course content as doubtful course content, and then extracting the knowledge point features of the doubtful course content, analyzing the user's mastery of the knowledge point, and thus analyzing whether the user has mastered the doubtful course content, and judging whether the doubtful course content is the cause of the user's doubt. If the target user has doubts during learning, but has mastered it later, it means that the corresponding course content is not doubtful content. If the target user fails to master it, it appears that the corresponding doubtful course content is the cause of the user's doubt.

[0111] Furthermore, in a preferred embodiment of the present invention, analyzing whether the teaching content needs to be adjusted based on the user's doubt reason analysis information and formulating an optimization plan specifically includes:

[0112] Obtain the analysis information of the reasons for user doubts, extract the corresponding doubt knowledge point information, count the number of target users in the current teaching class who have doubts about the target knowledge point, conduct a universal analysis of the reasons for doubts, and obtain universal analysis information of the reasons for doubts;

[0113] According to the universal analysis information of the doubt reasons, the course content adaptation is determined, and the adaptability of the teaching method of the current course content and the current teaching class is analyzed to obtain adaptability analysis information;

[0114] If the adaptability analysis information determines that the teaching method of the current course content is not suitable for the users of the current teaching class, it means that the current teaching content needs to be adjusted;

[0115] Based on the universality analysis information of the doubt reasons, the number of users who doubt the target course content is extracted, and compared with the preset threshold, the influence of the current course content on the users in the target teaching class is analyzed to obtain the influence analysis information;

[0116] Retrieve the teaching plan for the corresponding knowledge point based on the current education platform, extract the teaching difficulty of each teaching plan, match it with the impact analysis information, formulate an adjustment plan based on the matching result, and adjust the teaching content of the current teaching class;

[0117] If the adaptability analysis information determines that the teaching method of the current course content is suitable for the users in the current teaching class, then an adaptability evaluation is performed on the target users;

[0118] Preset user adaptation evaluation weights, obtain learning status evaluation information, extract course learning features corresponding to doubtful knowledge points as completion features and learning cycle features, conduct adaptation evaluation on target users, and obtain user adaptation evaluation information;

[0119] Based on the user adaptation evaluation information, it is determined whether the user meets the teaching difficulty of the current teaching class. If the user meets the teaching difficulty, other teaching plans are obtained according to the target user's doubtful knowledge points as optimization plans and pushed to the user for knowledge point reinforcement;

[0120] If it does not match, it means that the teaching difficulty of the current teaching class is not suitable for the target user, then the teaching classes of other teaching difficulties are obtained to generate a learning suggestion report and push it to the target user.

[0121] It should be noted that after analyzing the reasons for the user's doubts, the reasons for the user's doubts are obtained as corresponding doubt courses. At this time, it is necessary to analyze the universality of the target doubt course, and to obtain the universality of the doubt content by analyzing the doubt situation in the teaching class where the target user is located. If the universality is very large, it means that the teaching content of the doubt content is not suitable for the current class, and the teaching plan needs to be adjusted. If the universality is not large, it means that the doubt content is the personal doubt of the target user, and then the target user is evaluated for adaptation to evaluate whether the target user is suitable for studying in the current teaching class. By analyzing the user's learning characteristics, homework completion characteristics and learning cycle, the degree of adaptation between the user and the current class is judged, and then decisions are made based on the evaluation results, and the target user's knowledge points are reinforced, or the target user is recommended to go to other classes of teaching difficulty to study, so as to teach students in accordance with their aptitude and improve the user's learning experience and learning quality.

[0122] Figure 2 A flow chart of teaching content optimization provided by an embodiment of the present invention;

[0123] like Figure 2 As shown, the present invention provides a teaching content optimization flow chart, including:

[0124] S202, obtaining the target user's course learning monitoring information, and performing learning progress analysis and learning status analysis;

[0125] S204, obtaining user learning monitoring information, analyzing the user's emotional state during learning according to the user learning monitoring information, extracting course content corresponding to the user's doubt state and analyzing the cause of the user's doubt;

[0126] S206, judging whether it is necessary to adjust the teaching content according to the user's doubt reason analysis information, and analyzing the universality of the doubt reason;

[0127] S208, analyzing the information based on the universality of the doubt reasons to determine whether the doubt is a problem with the teaching content or a personal doubt of the user;

[0128] S210: If the question is a personal question of the user, an adaptation assessment is performed on the target user to determine whether to reinforce the knowledge points or reduce the teaching difficulty.

[0129] In addition, the present invention provides a cloud server-based online education level assessment method, which further includes the following steps:

[0130] Obtain user adaptation evaluation information, and determine whether the user meets the teaching difficulty of the current teaching class based on the user adaptation evaluation information. If the user meets the teaching difficulty, an optimization plan is formulated;

[0131] Obtain the target user's doubtful course content, search for knowledge points through the target education platform, and obtain the teaching courses and exercise content corresponding to the doubtful course content;

[0132] Preset a difficulty level classification threshold, extract the difficulty characteristics of each teaching course and exercise content, sort the difficulty, and classify them according to the difficulty level classification threshold to obtain level classification result information;

[0133] Obtain the learning status assessment information of the target user, analyze the target user's mastery of the doubtful course content, conduct a grade assessment, and obtain the user grade assessment information;

[0134] Match the user level assessment information and the level division result information, analyze the target user's knowledge point reinforcement level, extract the teaching courses and exercise contents of the corresponding level, and formulate an optimization plan to reinforce the target user's knowledge points;

[0135] Obtain test result information on target users’ knowledge points, evaluate the user’s mastery level, and adjust the optimization plan based on the evaluation results.

[0136] It should be noted that when the target users only have questions about certain course content, it may be that the doubts are caused by the teaching content of the current class. By changing the teaching courses and exercises, the target users can be provided with more detailed and low-difficulty knowledge points, and the difficulty level of the corresponding exercises can be adjusted according to the user's test results, so that users can better reinforce their knowledge points, improve their learning quality, and enhance the platform's education level.

[0137] Figure 3An online education level assessment system 3 based on a cloud server is provided in one embodiment of the present invention. The system comprises: a memory 31 and a processor 32. The memory 31 contains an online education level assessment method program based on a cloud server. When the online education level assessment method program based on a cloud server is executed by the processor 32, the following steps are implemented:

[0138] Obtaining the target user's course learning monitoring information, and performing user learning progress analysis based on the course learning monitoring information to obtain learning progress analysis information;

[0139] Set evaluation indicators, evaluate the learning status of target users, evaluate the user's course mastery, and obtain learning status evaluation information;

[0140] Acquire user learning monitoring information, analyze the user's emotional state during learning according to the user learning monitoring information, and obtain emotional state analysis information;

[0141] Acquire the course content corresponding to the user's doubt state according to the emotional state analysis information, analyze the cause of the user's doubt in combination with the learning status evaluation information, and obtain the user's doubt cause analysis information;

[0142] Based on the user's doubt cause analysis information, it is determined whether the teaching content needs to be adjusted and an optimization plan is developed.

[0143] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0144] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0145] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0146] A person of ordinary skill in the art can understand that: all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: a mobile storage device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0147] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0148] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A cloud server-based online education level assessment method, characterized in that: include: Obtaining the target user's course learning monitoring information, and performing user learning progress analysis based on the course learning monitoring information to obtain learning progress analysis information; Set evaluation indicators, evaluate the learning status of target users, evaluate the user's course mastery, and obtain learning status evaluation information; Acquire user learning monitoring information, analyze the user's emotional state during learning according to the user learning monitoring information, and obtain emotional state analysis information; Acquire the course content corresponding to the user's doubt state according to the emotional state analysis information, analyze the cause of the user's doubt in combination with the learning status evaluation information, and obtain the user's doubt cause analysis information; Analyze whether the teaching content needs to be adjusted based on the user's doubt analysis information and formulate an optimization plan; The obtaining of the target user's course learning monitoring information and performing user learning progress analysis based on the course learning monitoring information specifically includes: Obtaining the target user's course learning monitoring information, performing data cleaning on the course learning monitoring information, extracting the category features corresponding to each data, and combining the clustering algorithm to sort the data to obtain preprocessing information; Performing feature extraction on the preprocessed information to extract the course progress features of the target user, including course category, course name, course viewing status, and homework completion status, to obtain course progress feature information; Perform course learning situation analysis based on the course progress characteristic information, and analyze the target user's learning time, course viewing progress, and viewing times of each course to obtain course learning situation analysis information; Extract the course assignments corresponding to the courses that the target user has studied, calculate the target user's course assignment completion rate based on the course feature information, perform course assignment completion analysis, and obtain course assignment completion analysis information; Extracting the learning completion time of each course of the user according to the course progress feature information, and extracting the expected completion time of the user's current learning stage based on the course learning situation analysis information, performing a learning cycle analysis on the target user, judging whether it meets the expected consumption time, and obtaining the user's learning cycle analysis information; Preset learning progress assessment weights, combine the course learning situation analysis information, course homework completion situation analysis information and user learning cycle analysis information to perform weighted assessment, and obtain learning progress analysis information; The setting of evaluation indicators, evaluating the learning status of the target user, evaluating the user's course mastery, and obtaining learning status evaluation information specifically include: Based on big data retrieval, learning status evaluation indicators are obtained, the Pearson correlation coefficient between each evaluation indicator and the learning status is calculated to represent the degree of correlation, and the evaluation indicators are screened by judging with the preset threshold, and the Pearson correlation coefficient matrix is ​​constructed according to the screening results; Calculate the Euclidean distance between the selected evaluation indicators to form a distance matrix, convert the distance matrix into an adjacency matrix using a Gaussian kernel function, and calculate the vertex matrix; Subtracting the vertex matrix from the adjacency matrix to obtain a Laplacian matrix, performing eigendecomposition on the Laplacian matrix, presetting an expected number of clusters M, and performing clustering using a K-means algorithm; The evaluation indicators are set according to the clustering results, and the evaluation indicators are graded according to the Pearson correlation coefficients corresponding to the evaluation indicators in each cluster to obtain the evaluation indicator information; The entropy weight method is used to calculate the information entropy of each evaluation indicator, and the information gain rate of the corresponding evaluation indicator is calculated by the calculated information entropy. The weight of each evaluation indicator is assigned to obtain the evaluation weight information; Building a learning status evaluation model based on the evaluation index information and the evaluation weight information, obtaining course learning monitoring information and learning progress analysis information, inputting them into the learning status evaluation model for evaluation, and obtaining learning status evaluation information; The acquiring of user learning monitoring information, and analyzing the user's emotional state during learning according to the user learning monitoring information to obtain emotional state analysis information specifically includes: Based on big data retrieval, image information corresponding to various emotional states during learning is obtained, image preprocessing is performed, and classification is performed according to emotional states to form an instance data set; Performing feature extraction on the image information corresponding to each emotional state in the example data set, extracting facial features and body features, and obtaining first feature information; Extract the occurrence frequency of each feature in each image in the corresponding emotional state according to the first feature information as a correlation index, make a judgment with a preset threshold, select features greater than the preset threshold as strong correlation features, and obtain correlation degree analysis information; Constructing an emotional state analysis model based on CNN, constructing a training data set according to the correlation analysis information and the example data set to perform deep learning and training on the emotional state analysis model, and obtaining an emotional state analysis model that meets expectations; Acquiring user learning monitoring information, the user learning monitoring information including face monitoring information and body monitoring information, performing wavelet transform on the user learning monitoring information, and decomposing the target image into sub-bands of different frequency bands; Preset a processing threshold, perform threshold processing on each sub-band, set the wavelet coefficients smaller than the processing threshold to zero, extract high-frequency wavelet coefficients and perform linear enhancement, merge coefficients according to the enhanced high-frequency wavelet coefficients, perform inverse transformation on the merged wavelet coefficients, and obtain enhanced image information; The enhanced image information is input into the emotional state analysis model for analysis, the emotional changes of the user during learning are analyzed, and the information is associated with the corresponding time to obtain emotional state analysis information.

2. According to the cloud server-based online education level assessment method of claim 1, it is characterized in that: The acquiring of course content corresponding to the user's doubt state according to the emotional state analysis information, and analyzing the cause of the user's doubt in combination with the learning status evaluation information specifically includes: Acquire emotional state analysis information, construct an emotional change trend graph according to the emotional state analysis information, and extract a corresponding doubtful emotional time period according to the emotional change trend graph; Extract the emotional state of neighboring time periods of the doubtful emotional time period according to the emotional change trend graph, perform user emotional transition analysis, and obtain emotional transition analysis information; Based on the extracted doubt emotion time period, the user viewing content of the corresponding time period is extracted and marked as doubt course content, thereby obtaining doubt course content information; Acquire learning status evaluation information, extract knowledge point features of target course content according to the doubtful course content information, acquire user mastery of corresponding knowledge point features through the learning status evaluation information, and obtain user mastery analysis information; The user learning frequency, score and knowledge point error rate of the target knowledge point are extracted based on the user mastery analysis information, and the user doubt cause analysis is performed in combination with the emotion change analysis information to obtain user doubt cause analysis information.

3. According to the cloud server-based online education level assessment method of claim 1, it is characterized in that: The analyzing whether the teaching content needs to be adjusted according to the user's doubt reason analysis information and formulating an optimization plan specifically includes: Obtain the analysis information of the reasons for user doubts, extract the corresponding doubt knowledge point information, count the number of target users in the current teaching class who have doubts about the target knowledge point, conduct a universal analysis of the reasons for doubts, and obtain universal analysis information of the reasons for doubts; According to the universal analysis information of the doubt reasons, the course content adaptation is determined, and the adaptability of the teaching method of the current course content and the current teaching class is analyzed to obtain adaptability analysis information; If the adaptability analysis information determines that the teaching method of the current course content is not suitable for the users of the current teaching class, it means that the current teaching content needs to be adjusted; Based on the universality analysis information of the doubt reasons, the number of users who doubt the target course content is extracted, and compared with the preset threshold, the influence of the current course content on the users in the target teaching class is analyzed to obtain the influence analysis information; Retrieve the teaching plan for the corresponding knowledge point based on the current education platform, extract the teaching difficulty of each teaching plan, match it with the impact analysis information, formulate an adjustment plan based on the matching result, and adjust the teaching content of the current teaching class; If the adaptability analysis information determines that the teaching method of the current course content is suitable for the users in the current teaching class, then an adaptability evaluation is performed on the target users; Preset user adaptation evaluation weights, obtain learning status evaluation information, extract course learning features corresponding to doubtful knowledge points as completion features and learning cycle features, conduct adaptation evaluation on target users, and obtain user adaptation evaluation information; Based on the user adaptation evaluation information, it is determined whether the user meets the teaching difficulty of the current teaching class. If the user meets the teaching difficulty, other teaching plans are obtained according to the target user's doubtful knowledge points as optimization plans and pushed to the user for knowledge point reinforcement; If it does not match, it means that the teaching difficulty of the current teaching class is not suitable for the target user, then the teaching classes of other teaching difficulties are obtained to generate a learning suggestion report and push it to the target user.

4. An online education level assessment system based on a cloud server, characterized in that: The system includes: a memory and a processor, wherein the memory contains a cloud server-based online education level assessment method program, and when the cloud server-based online education level assessment method program is executed by the processor, the following steps are implemented: Obtaining the target user's course learning monitoring information, and performing user learning progress analysis based on the course learning monitoring information to obtain learning progress analysis information; Set evaluation indicators, evaluate the learning status of target users, evaluate the user's course mastery, and obtain learning status evaluation information; Acquire user learning monitoring information, analyze the user's emotional state during learning according to the user learning monitoring information, and obtain emotional state analysis information; Acquire the course content corresponding to the user's doubt state according to the emotional state analysis information, analyze the cause of the user's doubt in combination with the learning status evaluation information, and obtain the user's doubt cause analysis information; Analyze whether the teaching content needs to be adjusted based on the user's doubt analysis information and formulate an optimization plan; The obtaining of the target user's course learning monitoring information and performing user learning progress analysis based on the course learning monitoring information specifically includes: Obtaining the target user's course learning monitoring information, performing data cleaning on the course learning monitoring information, extracting the category features corresponding to each data, and combining the clustering algorithm to sort the data to obtain preprocessing information; Performing feature extraction on the preprocessed information to extract the course progress features of the target user, including course category, course name, course viewing status, and homework completion status, to obtain course progress feature information; Perform course learning situation analysis based on the course progress characteristic information, and analyze the target user's learning time, course viewing progress, and viewing times of each course to obtain course learning situation analysis information; Extract the course assignments corresponding to the courses that the target user has studied, calculate the target user's course assignment completion rate based on the course feature information, perform course assignment completion analysis, and obtain course assignment completion analysis information; Extracting the learning completion time of each course of the user according to the course progress feature information, and extracting the expected completion time of the user's current learning stage based on the course learning situation analysis information, performing a learning cycle analysis on the target user, judging whether it meets the expected consumption time, and obtaining the user's learning cycle analysis information; Preset learning progress assessment weights, combine the course learning situation analysis information, course homework completion situation analysis information and user learning cycle analysis information to perform weighted assessment, and obtain learning progress analysis information; The setting of evaluation indicators, evaluating the learning status of the target user, evaluating the user's course mastery, and obtaining learning status evaluation information specifically include: Based on big data retrieval, learning status evaluation indicators are obtained, the Pearson correlation coefficient between each evaluation indicator and the learning status is calculated to represent the degree of correlation, and the evaluation indicators are screened by judging with the preset threshold, and the Pearson correlation coefficient matrix is ​​constructed according to the screening results; Calculate the Euclidean distance between the selected evaluation indicators to form a distance matrix, convert the distance matrix into an adjacency matrix using a Gaussian kernel function, and calculate the vertex matrix; Subtracting the vertex matrix from the adjacency matrix to obtain a Laplacian matrix, performing eigendecomposition on the Laplacian matrix, presetting an expected number of clusters M, and performing clustering using a K-means algorithm; The evaluation indicators are set according to the clustering results, and the evaluation indicators are graded according to the Pearson correlation coefficients corresponding to the evaluation indicators in each cluster to obtain the evaluation indicator information; The entropy weight method is used to calculate the information entropy of each evaluation indicator, and the information gain rate of the corresponding evaluation indicator is calculated by the calculated information entropy. The weight of each evaluation indicator is assigned to obtain the evaluation weight information; Building a learning status evaluation model based on the evaluation index information and the evaluation weight information, obtaining course learning monitoring information and learning progress analysis information, inputting them into the learning status evaluation model for evaluation, and obtaining learning status evaluation information; The acquiring of user learning monitoring information, and analyzing the user's emotional state during learning according to the user learning monitoring information to obtain emotional state analysis information specifically includes: Based on big data retrieval, image information corresponding to various emotional states during learning is obtained, image preprocessing is performed, and classification is performed according to emotional states to form an instance data set; Performing feature extraction on the image information corresponding to each emotional state in the example data set, extracting facial features and body features, and obtaining first feature information; Extract the occurrence frequency of each feature in each image in the corresponding emotional state according to the first feature information as a correlation index, make a judgment with a preset threshold, select features greater than the preset threshold as strong correlation features, and obtain correlation degree analysis information; Constructing an emotional state analysis model based on CNN, constructing a training data set according to the correlation analysis information and the example data set to perform deep learning and training on the emotional state analysis model, and obtaining an emotional state analysis model that meets expectations; Acquiring user learning monitoring information, the user learning monitoring information including face monitoring information and body monitoring information, performing wavelet transform on the user learning monitoring information, and decomposing the target image into sub-bands of different frequency bands; Preset a processing threshold, perform threshold processing on each sub-band, set the wavelet coefficients smaller than the processing threshold to zero, extract high-frequency wavelet coefficients and perform linear enhancement, merge coefficients according to the enhanced high-frequency wavelet coefficients, perform inverse transformation on the merged wavelet coefficients, and obtain enhanced image information; The enhanced image information is input into the emotional state analysis model for analysis, the emotional changes of the user during learning are analyzed, and the information is associated with the corresponding time to obtain emotional state analysis information.

Images