Online training student management system and method

Abstract

The invention discloses an online training student management system and method, and the system comprises a data collection module, a data processing module and an authority verification module, and relates to the technical field of teaching and training management. According to the online training student management system and method, learning behavior details of students can be more completely captured by collecting text interaction data, operation track data and audio and video data, and the accuracy of collected data is improved by preprocessing behavior data, so that the learning efficiency is improved. Reliable data input is provided for generation of a structured intention report and calculation of course priorities, then the credibility of decision basis is improved, dynamic adjustment of a learning path is performed through a timing path dynamic adjustment mechanism, the learning efficiency and learning quality of students are improved, and in cooperation with calculation of a multi-dimensional evaluation report and a business value contribution value, the learning efficiency and learning quality of the students are improved. An enterprise can clearly judge the contribution of training to a business target, a data basis is provided for scientific distribution of training budget, and the situation of resource waste can be effectively reduced.

Description

Technical Field

[0001] This invention relates to the field of education and training management technology, specifically to an online training student management system and method. Background Technology

[0002] With the increasing prevalence of online education and training in corporate training, professional certification, and other scenarios, online training management has become a core tool for improving training efficiency. However, current online training management methods suffer from the following shortcomings:

[0003] 1) Limited data collection dimensions: Current online training management methods often only collect superficial data such as students' course viewing time and assignment submission rate, which cannot provide high-quality data support for decision-making;

[0004] 2) Static and fixed learning paths: The learning paths in the current online training management method are only generated according to the difficulty of the course or the preset order, and cannot be dynamically adjusted based on the actual learning status of the learners. This results in a low degree of matching between the learning paths and the learners' needs and low learning efficiency.

[0005] 3) The evaluation of training effectiveness has limitations: The current online training management method focuses only on internal training indicators such as the accuracy rate of knowledge assessment and the score of practical skills, and lacks the connection with external business data such as order conversion rate and customer satisfaction. As a result, enterprises cannot judge the actual contribution of training to business objectives, the allocation of enterprise training budget lacks data basis, and resources are wasted.

[0006] In view of this, an online training student management system and method are proposed. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides an online training student management system and method, which solves the problems of current online training management methods, such as single data collection dimensions, static and fixed learning paths, and limited training effect evaluation.

[0008] To achieve the above objectives, a first aspect of the present invention provides an online training student management system, comprising:

[0009] The data acquisition module is used to collect trainees' behavioral data, preprocess the behavioral data, and synchronize external business data; the behavioral data includes text interaction data, operation trajectory data, and audio and video data.

[0010] The data processing module generates a structured intent report based on the preprocessed behavioral data, quantifies course priorities, and generates and dynamically adjusts personalized learning paths.

[0011] The permission verification module is used to verify the identities of different roles and provide differentiated services to different roles; the roles include students, administrators and lecturers.

[0012] A second aspect of the present invention provides a method for managing online training participants, specifically including the following steps:

[0013] S1. Start the data acquisition unit to collect text interaction data, operation trajectory data and audio and video data through the learning terminal used by the students, preprocess the three behavioral data, and synchronize external business data.

[0014] S2. Generate a structured intent report based on preprocessed text interaction data, operation trajectory data, and audio / video data;

[0015] S3. Based on the assessment and answer data, business scenario simulation operation data and external business data during the course learning process, evaluate the assessment accuracy, practical task score and the change rate of business indicators before and after training, generate a multi-dimensional assessment report, calculate the course priority score, obtain personalized learning paths based on the course priority score in descending order, and dynamically adjust the personalized learning paths based on the timed path dynamic adjustment mechanism.

[0016] S4. Calculate the business value contribution value and the return on investment in training as a value mapping relationship between training and business.

[0017] This invention provides an online training student management system and method. It has the following beneficial effects:

[0018] This invention, by collecting text interaction data, operation trajectory data, and audio-visual data, can more comprehensively capture the details of learners' learning behavior. Preprocessing of the behavioral data improves the accuracy of the collected data, providing reliable data input for the generation of structured intent reports and the calculation of course priorities, thereby enhancing the credibility of decision-making. Furthermore, a timed dynamic path adjustment mechanism dynamically adjusts the learning path, improving learners' learning efficiency and quality. Combined with multi-dimensional evaluation reports and the calculation of business value contribution, this allows enterprises to clearly determine the contribution of training to business objectives, providing a data foundation for the scientific allocation of training budgets and effectively reducing resource waste. Attached Figure Description

[0019] Figure 1 This is a system principle block diagram of the present invention. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0021] Please see Figure 1The first aspect of this invention provides an online training student management system, including a data acquisition module, a data processing module, and an access verification module connected in sequence. The data acquisition module is used to collect student behavior data, preprocess the behavior data, and synchronize external business data. The behavior data includes text interaction data, operation trajectory data, and audio / video data. The data processing module generates a structured intent report based on the preprocessed behavior data, quantifies course priorities, and generates and dynamically adjusts personalized learning paths. The access verification module is used to verify the identities of different roles and provide differentiated services to different roles. Roles include students, administrators, and instructors.

[0022] In one exemplary embodiment, the permission verification module includes an identity recognition and verification unit, a role permission allocation unit, and a role service distribution unit;

[0023] The identity recognition and verification unit is used to collect user login information and match and verify it with pre-stored role information to identify the user's actual role;

[0024] The role permission allocation unit is used to assign preset permissions according to different roles;

[0025] The role service distribution unit is used to provide corresponding service functions to roles based on their permissions.

[0026] In one exemplary embodiment, the data acquisition module includes a terminal data acquisition unit and a data preprocessing unit.

[0027] To further explain, the terminal data acquisition unit is used to collect trainees' behavioral data and synchronize external business data. The behavioral data includes:

[0028] Text interaction data includes keywords used in student questions, notes modification records, and comments.

[0029] Operation trajectory data, which includes click coordinates, page dwell time, and navigation path;

[0030] The audio and video data includes the volume of the speaker's voice during the live class and facial expression data based on 68 feature points.

[0031] To further explain, the data preprocessing unit removes outliers from behavioral data based on the 3σ principle, fills missing text values ​​with unknown labels, fills missing numerical values ​​with linear interpolation, and retains only the first record of duplicate values ​​per unit time.

[0032] In an exemplary embodiment, the data processing module includes a behavioral semantic analysis unit, a multi-dimensional evaluation unit, an intelligent decision-making unit, a dynamic path generation unit, and a mapping processing unit.

[0033] To further explain, the behavioral semantic analysis unit generates a structured intent report based on behavioral data; the structured intent report includes the type of comprehension obstacle, interest tendencies, and learning status.

[0034] As explained in detail, the methods for determining interests include:

[0035] Based on the coordinates in the operation trajectory data, obtain the students' click records on the courses, and count and calculate the ratio of the total number of times students click on case study courses to the total number of times students click on all courses within a specified period, such as 12 hours. This ratio is recorded as the click percentage of case study courses.

[0036] A case keyword dictionary is constructed based on the case themes of the training courses. Text interaction data, including keywords in student questions, notes modification records, and comments, are obtained based on the case keyword dictionary. These include, but are not limited to, customer cases, transaction cases, and case breakdowns. The ratio of the total number of occurrences of case-related keywords to the total number of occurrences of all keywords in the text is recorded as the case keyword frequency ratio.

[0037] Based on student comments on the course from text interaction data, a Naive Bayes model is used to determine the sentiment tendency of student ratings. If useful and easy-to-understand vocabulary appears, the sentiment tendency is judged as positive; if difficult or useless vocabulary appears, the sentiment tendency is judged as negative; if average vocabulary appears, the sentiment tendency is judged as neutral. A rating is assigned based on the sentiment tendency.

[0038]

[0039] The formula for determining interest inclination is as follows:

[0040]

[0041] In the formula, Rate your interests The percentage of clicks for case study courses. Weighting based on the percentage of clicks for case study courses. For the frequency ratio of case keywords, The frequency ratio weight of case keywords Rate the review Weighting of review ratings;

[0042] when At that time, the current course was determined to be a high-interest course;

[0043] As explained in detail, the methods for determining learning status include:

[0044] Facial images of trainees are collected and facial feature point expression data of 68 points are extracted using the Dlib+68-point model. Cosine similarity is calculated between these data and the 68 feature point expression data of a pre-stored standard confused expression. The closer the result is to 1, the more confused the trainee is.

[0045] The volume of students speaking during live classes is collected through the microphone of the student learning terminal, and the audio energy is extracted using the audio processing tool FFmpeg. The volume energy range is 0-1, and the lower the energy, the more fatigued the student is.

[0046] The formula for determining the learning state is:

[0047]

[0048]

[0049] In the formula, Rate focus level The similarity between the current emoji and the standard confused emoji. Click frequency, This represents the fatigue energy value.

[0050] Regarding focus rating, if a focus rating remains below 0.5 for 5 minutes when a high-interest course is identified, it is considered a misjudgment. The priority score of that high-interest course will be reduced by a set percentage, such as 20%, to ensure that the course priority is more in line with the student's actual learning status.

[0051] To further explain, the multi-dimensional assessment unit evaluates the accuracy of assessments, scores of practical tasks, and changes in business indicators before and after training based on test answer data, business scenario simulation operation data, and external business data, and obtains a multi-dimensional assessment report; external business data includes enterprise HR system data, enterprise CRM / ERP system data, and certification supervision system data;

[0052] To further explain, the intelligent decision-making unit quantifies course priority scores based on structured intent reports and multi-dimensional evaluation reports.

[0053] As explained in detail, the methods for obtaining quantitative course priority scores include:

[0054] The knowledge matching degree is the ratio of the coverage of course content with the knowledge points that hinder understanding. For example, for a course on product pricing formulas, students may have two knowledge points that hinder understanding: pricing formulas and demand classification. However, if the course content only covers one knowledge point that hinders understanding: pricing formulas, then the knowledge matching degree is only 0.5.

[0055] Obtain the practical task score from the multi-dimensional assessment report. When the practical task score is lower than the preset threshold, mark the course skill content corresponding to the skill of the practical task as a skill deficiency. Calculate the ratio of the number of skill content corresponding to the skill deficiency in the current course to the total number of skill content corresponding to the skill deficiency in the multi-dimensional assessment report, and record it as the skill matching degree.

[0056] when At that time, the currently clicked course type is determined to be a high-interest course type. Based on the proportion of high-interest courses in the learning curriculum, segmented values ​​are assigned, recorded as the interest matching degree:

[0057]

[0058] Based on external business data, student performance targets are obtained. Based on operation trajectory data, changes in relevant business indicators after students learn courses are obtained. The ratio of the change in indicators of course contribution to performance targets is calculated and recorded as the target matching degree. It should be noted that the upper limit of the target matching degree is 1.0.

[0059] The formula for calculating the quantitative course priority score is:

[0060]

[0061] In the formula, Score based on course priority. Knowledge matching degree, which refers to the proportion of knowledge points that cause comprehension difficulties covered in the course. As the weight for knowledge matching degree, Skill matching degree, which refers to the percentage of skill gaps covered. As a weight for skill matching, Interest matching score, which is the ratio of course type to interest inclination score. Weighted by interest matching degree The target alignment rate refers to the proportion of the course's contribution to performance objectives in external business data. The target matching degree weight.

[0062] To further explain, the dynamic path generation unit obtains personalized learning paths based on the descending order of course priority scores, and constructs a timed dynamic path adjustment mechanism:

[0063] If the time spent on the page exceeds the set time and the error rate exceeds the set percentage, it is determined that there is a point of difficulty in understanding. At this time, several related breakdown courses will be added.

[0064] When highly engaging courses appear, supplement them with practical business tasks.

[0065] When the accuracy rate of the assessment exceeds the preset value, the knowledge point is judged as mastered and the subsequent repeated lessons are skipped.

[0066] If the focus score falls below the set threshold and remains below the set threshold for a set duration, such as If the student's attention is found to be distracted after 10 minutes, a short break is inserted: a 3-5 minute countdown break is initiated, and the current lesson is paused.

[0067] To further explain, the mapping processing unit establishes a value mapping relationship between training and business based on external business data and multi-dimensional evaluation reports. The value mapping relationship between training and business includes the business value contribution value and the return on investment in training.

[0068] As explained in detail, the calculation method for business value contribution includes:

[0069] The ratio of completed courses in the personalized learning path to the total number of courses in the personalized learning path is calculated and recorded as the course completion rate.

[0070] The ratio of total training investment to the total number of trainees in the external business data is recorded as the average training cost per person.

[0071] The formula for calculating business value contribution is:

[0072]

[0073]

[0074] In the formula, Contribution to business value The knowledge score is calculated based on the accuracy of the assessment. As for the weighting of knowledge scores, we assign 0.3 for sales positions and 0.5 for technical positions. The score is for skills, specifically for practical tasks. As a weighting factor for skills scores, a weight of 0.5 is used for sales positions and 0.3 for technical positions. To achieve the course completion rate, As a weight for the course completion rate, it is set to 0.1. For the average training cost per person, The weight for per capita training cost is set to 0.1;

[0075] The formula for calculating the return on investment in training is:

[0076]

[0077] In the formula, Return on investment in training.

[0078] A second aspect of this invention provides an online training student management method, applied to the aforementioned online training student management system, the method specifically including the following steps:

[0079] S1. Start the data acquisition unit to collect text interaction data, operation trajectory data and audio and video data through the learning terminal used by the students, preprocess the three behavioral data, and synchronize external business data.

[0080] S2. Generate a structured intent report based on preprocessed text interaction data, operation trajectory data, and audio / video data;

[0081] S3. Based on the assessment and answer data, business scenario simulation operation data and external business data during the course learning process, evaluate the assessment accuracy, practical task score and the change rate of business indicators before and after training, generate a multi-dimensional assessment report, calculate the course priority score, obtain personalized learning paths based on the course priority score in descending order, and dynamically adjust the personalized learning paths based on the timed path dynamic adjustment mechanism.

[0082] S4. Calculate the business value contribution value and the return on training investment as a value mapping relationship between training and business.

[0083] S5, the authorization verification module identifies and determines specific roles based on identity information, and provides differentiated services to different roles:

[0084] For learners, personalized learning paths are pushed to them, and their personal knowledge and skills scores and progress towards achieving their goals are displayed.

[0085] For instructors, this involves showing them student interaction data for the courses they teach.

[0086] For administrators, the system pushes alerts to courses that pass the preset threshold and generates heatmaps of training return on investment for different courses.

[0087] In conjunction with the second aspect of the embodiment, from the perspective of the learner, the specific steps of their personalized learning are as follows:

[0088] STEP 1: Enter the system terminal, enter the student ID or mobile phone number and mobile verification code as identity information. The identity recognition and verification unit verifies the identity information through the pre-stored role identity database. After successful verification, you will enter the student's exclusive interface.

[0089] STEP2: The student-exclusive homepage generates unit outputs and displays the learning path through dynamic paths, with the learning paths arranged in descending order of course priority scores.

[0090] STEP 3: Click to enter the course interface. The terminal data acquisition unit collects learning behavior data in real time.

[0091] Text-based interactive data: Enter keywords for questions, edit notes, and submit course comments after completing the course;

[0092] Operation trajectory data: records page dwell time, click coordinates, and navigation path;

[0093] Audio and video data: Collect student speaking volume and facial expression data from 68 feature points;

[0094] STEP 4: Participants view the multi-dimensional assessment report through the multi-dimensional assessment unit:

[0095] Knowledge score, which is the accuracy rate of knowledge points;

[0096] Skill score, i.e., score for practical tasks;

[0097] Progress towards achieving the target: The remaining percentage to achieve the business target, such as the current conversion rate of 12%, which is 2 percentage points away from the quarterly target of 14%;

[0098] STEP 5: Automatically update the learning path every 12 hours.

[0099] If a course is paused for more than 30 seconds and the error rate exceeds 40%, it will be marked as an obstacle knowledge point, and 2-3 breakdown courses will be automatically added to the path.

[0100] If the interest score is ≥0.6, it is recorded as a high interest point, and a similar business practice task is added;

[0101] If the accuracy rate of the assessment is ≥90%, it means that the knowledge points have been mastered and subsequent repeated courses can be skipped.

[0102] In conjunction with the second aspect of the embodiment, the usage steps for a lecturer are as follows:

[0103] STEP 1: After entering their employee ID and completing facial recognition, the instructor logs in and enters their personal workbench, which displays courses to be taught and courses taught in the past.

[0104] STEP 2: Click on the historical courses to view the multi-dimensional evaluation results and behavioral data of the students in those courses:

[0105] Question data: The frequency of questions is counted by keyword and sorted from highest to lowest frequency. Click on a keyword to view the specific questions asked by students.

[0106] Comment data: After being categorized by sentiment, the commenter's ID is labeled.

[0107] Instructors can improve their teaching based on the multi-dimensional evaluation results temporarily displayed in the multi-dimensional evaluation unit: for example, the error rate for the knowledge point on handling price objections is 45%, so it is recommended to add negotiation cases in the home appliance industry;

[0108] In conjunction with the second aspect of the embodiment, the usage steps for an administrator are as follows:

[0109] STEP 1: The administrator logs in by entering their username, password, and Ukey, grants full data access, and sets a course pass rate warning threshold, such as 50%-80%.

[0110] STEP 2: Real-time push notifications of alerts, sorted by urgency level: red = urgent, yellow = need attention. Course alerts: a yellow alert is issued when the course pass rate is below 80%, and a red alert is issued when the course pass rate is below 50%.

[0111] STEP 3: Generate a heatmap based on the return on investment (ROI) of the course. The higher the ROI, the darker the color.

[0112] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An online training student management system, characterized by, include: The data acquisition module is used to collect students' behavioral data, preprocess the behavioral data, and synchronize external business data. The behavioral data includes text interaction data, operation trajectory data, and audio and video data; The data processing module generates a structured intent report based on the preprocessed behavioral data, quantifies course priorities, and generates and dynamically adjusts personalized learning paths. The permission verification module is used to verify the identities of different roles and provide differentiated services to different roles; the roles include students, administrators and lecturers.

2. The online training management system of claim 1, wherein, The permission verification module includes an identity recognition and verification unit, a role permission allocation unit, and a role service distribution unit; The identity recognition and verification unit is used to collect user login information and match and verify it with pre-stored role information to identify the user's actual role; The role permission allocation unit is used to assign preset permissions according to different roles; The role service distribution unit is used to provide corresponding service functions to roles according to their permissions.

3. The online training management system of claim 1, wherein, The data acquisition module includes a terminal data acquisition unit and a data preprocessing unit; The terminal data acquisition unit is used to collect trainees' behavioral data and synchronize external business data. The behavioral data includes: Text interaction data, which includes keywords from student questions, notes modification records, and comments; Operation trajectory data, which includes click coordinates, page dwell time, and jump path; The audio and video data includes the volume of the speaker's voice during the live class and facial expression data based on 68 feature points. The data preprocessing unit removes outliers from behavioral data based on the 3σ principle, fills missing text values ​​with unknown labels, fills missing numerical values ​​with linear interpolation, and retains only the first record of duplicate values ​​per unit time.

4. The online training management system of claim 3, wherein, The data processing module includes a behavioral semantic analysis unit, a multi-dimensional evaluation unit, an intelligent decision-making unit, a dynamic path generation unit, and a mapping processing unit. The behavioral semantic analysis unit generates a structured intent report based on behavioral data; the structured intent report includes the type of comprehension obstacle, interest tendency, and learning status; The multi-dimensional evaluation unit assesses the accuracy of the assessment, the score of the practical task, and the rate of change in business indicators before and after training based on the test answer data, business scenario simulation operation data, and external business data, and obtains a multi-dimensional evaluation report; the external business data includes enterprise HR system data, enterprise CRM / ERP system data, and certification supervision system data. The intelligent decision-making unit quantifies course priority scores based on structured intent reports and multi-dimensional evaluation reports; The dynamic path generation unit obtains personalized learning paths based on the course priority scores in descending order, and constructs a timed dynamic path adjustment mechanism. The mapping processing unit establishes a value mapping relationship between training and business based on external business data and multi-dimensional evaluation reports. The value mapping relationship between training and business includes the business value contribution value and the return on investment in training.

5. The online training management system of claim 4, wherein, The formula for determining the interest tendency is: ； In the formula, is the interest tendency score, is the case course click ratio, is the case course click ratio weight, is the case keyword frequency ratio, is the case keyword frequency ratio weight, is the comment score, is the comment score weight. When the current course is determined to be a high-interest course; The formula for determining the learning state is: ； ； In the formula, is the concentration score, is the similarity of the current expression to the standard confusion expression, is the click frequency, is the fatigue energy value.

6. The online training management system of claim 5, wherein, The calculation formula of the quantified course priority score is: ； In the formula, is the course priority score, is the knowledge matching degree, i.e., the proportion of course coverage of knowledge points of understanding barriers, is the knowledge matching degree weight, is the skill matching degree, i.e., the proportion of coverage of skill short boards, is the skill matching degree weight, is the interest matching degree, i.e., the matching proportion of course type and interest tendency score, is the interest matching degree weight, is the target matching degree, i.e., the contribution proportion of the course to performance targets in external business data, is the target matching degree weight.

7. The online training management system of claim 6, wherein, The timing path dynamic adjustment mechanism comprises: When the page stay duration exceeds the set duration and the error question rate exceeds the set percentage, it is determined that there is an understanding obstacle point, and at this time, a number of related disassembly courses are added; When a high-interest course appears, a business practice task is supplemented; When the evaluation accuracy rate exceeds the preset value, it is determined that the knowledge point is a mastered knowledge point, and the subsequent repeated course is skipped; After the concentration score is lower than the set threshold value and lasts for a set duration, it is determined that there is a state of attention dispersion, and a short rest node is inserted.

8. The online training management system of claim 7, wherein, The calculation formula of the business value contribution value is: ； ； wherein, is a business value contribution value, is a knowledge score, is a knowledge score weight, is a skill score, is a skill score weight, is a course completion rate, is a course completion rate weight, is a cost per training per person, is a cost per training per person weight; The calculation formula of the training investment return rate is: ； In the formula, Return on investment for training.

9. An online training student management method applied to the online training student management system of claim 8, characterized in that, The method specifically comprises the following steps: S1, starting the data acquisition unit, collecting text interaction data, operation trajectory data and audio and video data through the learning terminal used by the student, preprocessing the three behavior data, and synchronizing external business data; S2, generating a structured intent report based on the preprocessed text interaction data, operation trajectory data and audio and video data; S3, evaluating the evaluation accuracy rate, the real operation task score and the business index change rate before and after training based on the evaluation answer data, the business scene simulation operation data and the external business data in the course learning process, generating a multi-dimensional evaluation report, calculating the course priority score, obtaining the personalized learning path based on the descending order of the course priority score, and dynamically adjusting the personalized learning path by relying on the timing path dynamic adjustment mechanism; S4, calculating the business value contribution value and the training investment return rate as the value mapping relationship of training-business.

10. The method of claim 9, wherein, The permission verification module identifies and determines the specific role based on the identity information, and provides differentiated services to different roles: For the student role, the student is pushed with a personalized learning path, and the personal knowledge, skill score and target achievement progress are displayed; For the lecturer role, the student interaction data of the course is displayed to the lecturer; For the administrator role, the administrator is pushed with a course warning that the course passing rate is lower than the preset threshold value, and a training investment return rate heat map of different courses is generated.

Images