Adult online education learning quality evaluation method

Abstract

The invention discloses an adult online education learning quality evaluation method, and relates to the technical field of online education, and the method comprises the following steps: obtaining teaching video text content; dividing the text content into a plurality of teaching units carrying time range marks, function type marks and knowledge point marks according to teaching functions; acquiring a concentration index of the student when playing each teaching unit; obtaining the playing integrity of each teaching unit of the student; calculating a basic score for each teaching unit according to the playing integrity and concentration indexes; the method has the beneficial effects that the attention state of the student in the learning process can be effectively identified, and misjudgment caused by only depending on the watching duration or the test result is avoided; the test question can be dynamically generated according to the knowledge points of the teaching unit and the basic score of the student, the basic score is corrected through the answer result, and compared with a traditional single scoring mechanism, the real learning quality of the student can be more comprehensively measured.

Description

Technical Field

[0001] The present invention relates to the technical field of online education, and particularly to a method for evaluating the learning quality of adult online education. Background Art

[0002] With the rapid development of Internet technology, online education has become one of the important forms of adult continuing education. Compared with the traditional offline education mode, online education has the advantages of flexibility, convenience, rich resources, personalized learning, etc., and can meet the needs of different learners. However, in the online education environment, it is difficult to comprehensively perceive and quantitatively evaluate the learning behaviors of learners. The evaluation of teaching quality mainly relies on traditional methods such as the self-feedback of students, exam scores, or learning duration. These methods have certain limitations and are difficult to comprehensively and objectively reflect the actual learning situation and learning quality of students.

[0003] Existing online learning quality evaluation methods usually evaluate the learning effect of students based on viewing duration, course completion rate, or quiz scores. However, these indicators can only measure the learning situation of students from a partial perspective and cannot truly reflect the degree of concentration, knowledge mastery, and learning behavior patterns of students during the entire learning process. Moreover, the difficulty and content of the test questions used in quizzes often lack an effective correlation with the actual learning situation of students. For example, traditional tests may provide the same difficulty level of questions for all students and cannot dynamically adjust the question difficulty according to the learning performance of students, resulting in some students being unable to obtain an accurate evaluation due to the questions being too difficult or too easy.

[0004] Therefore, a method for evaluating the learning quality of adult online education is proposed. Summary of the Invention

[0005] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions shall not be used to limit the scope of the present invention.

[0006] In view of the above problems, the present invention is proposed.

[0007] To solve the above technical problems, the present invention provides the following technical solution: A method for evaluating the learning quality of adult online education, comprising the following steps:

[0008] Obtain the text content of the teaching video;

[0009] Divide the text content into multiple teaching units carrying time range marks, function type identifiers, and knowledge point identifiers according to teaching functions;

[0010] Obtain the concentration index of the learner when playing each of the teaching units;

[0011] Obtain the playback completeness of each teaching unit of the learner;

[0012] Calculate a basic score for each teaching unit based on the playback completeness and concentration index;

[0013] Judge whether to generate test questions according to the function type identifier. If so, generate a set of test questions according to the knowledge point identifier and the basic score, obtain the answer results to correct the basic score, and obtain the final score of each teaching unit. Otherwise, use the basic score as the final score;

[0014] Perform weighted calculation on the final scores of all teaching units to obtain the comprehensive score of the learner.

[0015] As a preferred solution of a method for evaluating the learning quality of adult online education according to the present invention, wherein the obtaining of the teaching video text content includes:

[0016] Obtain the teaching video;

[0017] Perform speech recognition on the audio track of the teaching video to obtain the text content with time stamps.

[0018] As a preferred solution of a method for evaluating the learning quality of adult online education according to the present invention, wherein the dividing of the text content into multiple teaching units carrying time range markers, function type identifiers, and knowledge point identifiers according to teaching functions includes:

[0019] Perform semantic segmentation on the text content, and extract teaching function keywords through natural language processing;

[0020] Based on a preset teaching function classification model, cluster adjacent text paragraphs with similar semantics into independent teaching units;

[0021] Mark time range markers and function type identifiers for each teaching unit;

[0022] Mark knowledge point identifiers for each teaching unit according to the core knowledge points in the teaching unit.

[0023] As a preferred solution of a method for evaluating the learning quality of adult online education according to the present invention, wherein after outputting the set of teaching units carrying time range markers, function type identifiers, and knowledge point identifiers, the correctness of the teaching unit division is verified by at least one of the following methods:

[0024] Verify whether the time ranges of adjacent teaching units are continuous and non-overlapping;

[0025] Verify the logical relevance between the function type identifier and the knowledge point identifier;

[0026] Check whether the duration of the teaching unit meets the preset threshold range.

[0027] As a preferred solution of a method for evaluating the learning quality of adult online education according to the present invention, wherein, after verifying the correctness of the division of the teaching unit, it further includes:

[0028] If the verification is passed, proceed to the subsequent steps;

[0029] If the verification fails, mark the teaching unit that fails the verification in the video text content and send it for manual review. After manual review, verify again. After the verification is passed, proceed to the subsequent steps.

[0030] As a preferred solution of a method for evaluating the learning quality of adult online education according to the present invention, wherein, obtaining the concentration index of the learner when playing each teaching unit includes:

[0031] Obtain the facial expression data of the learner during the playback of the entire teaching video, and continuously extract the expression feature data through a computer vision model;

[0032] Associate the expression feature data to the corresponding teaching unit according to the time axis;

[0033] Based on the expression feature data within each teaching unit, calculate the concentration index through an emotion recognition model.

[0034] As a preferred solution of a method for evaluating the learning quality of adult online education according to the present invention, wherein the basic score calculation formula is:

[0035]

[0036] Wherein, the M, C, and F respectively represent the highest score of the teaching video, the playback integrity, and the concentration index, and n represents the number of teaching units.

[0037] As a preferred solution of a method for evaluating the learning quality of adult online education according to the present invention, wherein generating a test question set according to the knowledge point identifier and the basic score includes:

[0038] Calculate the ratio of the basic score to the highest score of the current teaching unit;

[0039] Select the associated question bank according to the knowledge point identifier;

[0040] Use the ratio of the basic score to the highest score of the current teaching unit as the difficulty coefficient to match and generate a test question set from the question bank.

[0041] As a preferred embodiment of the method for evaluating the learning quality of adult online education of the present invention, the test question set at least includes:

[0042] Test questions with a difficulty coefficient higher than the above;

[0043] Test questions with a difficulty coefficient lower than the above;

[0044] Test questions with a difficulty coefficient equal to the above.

[0045] As a preferred embodiment of the method for evaluating the learning quality of adult online education of the present invention, the final score of each teaching unit is obtained by correcting the basic score as follows:

[0046] Based on the basic score:

[0047] If a test question with a difficulty coefficient higher than the above is answered correctly, points are added and no points are deducted for incorrect answers;

[0048] If a test question with a difficulty coefficient lower than the above and a test question with a difficulty coefficient equal to the above are answered incorrectly, points are deducted and no points are added for correct answers;

[0049] The maximum threshold of the final score is equal to the highest score of the teaching unit.

[0050] Advantages of the present invention:

[0051] 1. By dividing the teaching video into different teaching units and identifying them based on the time range, function type, and knowledge points, the evaluation of learning quality is made more accurate. Compared with the traditional method of evaluating the whole lesson, this solution can identify the learning quality of different teaching contents and provide a more targeted improvement direction for curriculum optimization.

[0052] 2. By using computer vision technology to obtain the facial expression data of the students and combining with the emotion recognition model to calculate the concentration index of the students in different teaching units, the scientific nature of the evaluation is improved. Compared with the traditional method, it can effectively identify the attention state of the students during the learning process and avoid misjudgment caused by relying only on the viewing duration or test scores.

[0053] 3. It can dynamically generate test questions according to the knowledge points of the teaching unit and the basic score of the students, and correct the basic score through the answering results. This method ensures that the test link can accurately reflect the students' mastery of knowledge. Compared with the traditional single scoring mechanism, it can measure the real learning quality of the students more comprehensively. Description of the Drawings

[0054] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings. Among them:

[0055] Figure 1 It is a flowchart of a method for evaluating the learning quality of adult online education of the present invention. Detailed implementation manners

[0056] To make the above objects, features and advantages of the present invention more obvious and understandable, the following will give a detailed description of the specific implementation manners of the present invention with reference to the accompanying drawings of the specification.

[0057] Many specific details are set forth in the following description to facilitate a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0058] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure or characteristic that may be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not all refer to the same embodiment, nor is it an independent or alternative embodiment that mutually excludes other embodiments.

[0059] Furthermore, the present invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of the present invention, for the convenience of explanation, the cross-sectional views showing the device structure will be enlarged locally not in accordance with the general ratio, and the schematic diagrams are only examples and should not limit the scope of protection of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions including length, width and depth should be included.

[0060] Embodiment

[0061] A method for evaluating the learning quality of adult online education, which includes the following steps:

[0062] S1, obtaining the text content of the teaching video;

[0063] Specifically, obtaining the text content of the teaching video includes:

[0064] Obtaining the teaching video; performing speech recognition on the audio track of the teaching video to obtain the text content with time stamps;

[0065] For the video content, it is difficult to directly analyze. By extracting the audio track information, this pain point is solved, and the unstructured video content is converted into structured text data, providing a basis for subsequent analysis;

[0066] For example, in a teaching video "Introduction to the Pythagorean Theorem", the identified text content with timestamps is as follows:

[0067] [00:00:00] Pythagorean theorem formula: a 2 + b 2 = c 2

[0068] [00:02:16] Example: Given a = 3 and b = 4, find the value of the hypotenuse c.

[0069] [00:05:40] This is the end of the course.

[0070] S2. Divide the text content into multiple teaching units carrying time range markers, function type identifiers, and knowledge point identifiers according to the teaching function;

[0071] Among them, the data carried by the teaching unit is represented here as [time range, function type, knowledge point]. The function type is a predefined type, which can include knowledge point explanation, example exercise, experimental verification, knowledge expansion, etc. according to the characteristics of teaching. The knowledge point is the teaching knowledge point within the teaching unit;

[0072] Specifically, dividing the text content into multiple teaching units carrying time range markers, function type identifiers, and knowledge point identifiers according to the teaching function includes: performing semantic segmentation on the text content, extracting teaching function keywords through natural language processing; clustering adjacent text paragraphs with similar semantics into independent teaching units based on a preset teaching function classification model; annotating time range markers and function type identifiers for each teaching unit; annotating knowledge point identifiers for each teaching unit according to the core knowledge points within the teaching unit;

[0073] For example, for the text content of "Introduction to the Pythagorean Theorem" identified in S1, the divided teaching units are: [00:00:00 - 00:02:15, knowledge point explanation, Pythagorean theorem formula], [00:02:16 - 00:05:40, example exercise, application of the Pythagorean theorem];

[0074] And, after outputting the set of teaching units carrying time range markers, function type identifiers, and knowledge point identifiers, the correctness of the teaching unit division is verified through at least one of the following methods: verifying whether the time ranges of adjacent teaching units are continuous and non - overlapping; verifying the logical relevance between the function type identifier and the knowledge point identifier; detecting whether the duration of the teaching unit meets the preset threshold range;

[0075] Among the above three verification methods:

[0076] Verifying the continuity of the time ranges of adjacent teaching units can avoid data fragmentation. The seamless connection of timestamps is the basis for calculating subsequent concentration metrics. If there are time overlaps or gaps, it will lead to inaccurate matching between the learner's concentration metrics and teaching units. Moreover, continuous time series division can reflect the coherence of teaching logic. For example, in "Introduction to the Pythagorean Theorem", if there is a 1-second blank between the teaching unit of knowledge point explanation (00:00:00 - 00:02:15) and the teaching unit of example practice (00:02:16 - 00:05:40), key transitional content may be missed, disrupting the knowledge transfer chain;

[0077] Verifying the logical relevance between the function type and knowledge points is for semantic consistency verification. The function type needs to form a teaching logic closed-loop with the knowledge point type. For example, in "Introduction to the Pythagorean Theorem": If the function type is "experimental verification", but the knowledge point identifier is "Pythagorean Theorem", it is a logical contradiction. If the knowledge point corresponding to the "example practice" unit is "trigonometric functions", it exposes a classification error in the semantic segmentation model;

[0078] Detecting the duration threshold of teaching units is to avoid the "phantom unit" problem caused by speech recognition errors or model mis-segmentation. Suppose a unit is wrongly divided as [00:05:41 - 00:06:00, knowledge point explanation, course summary, 0.5], but the actual video ends at 00:05:40. At this time, the time range verification will immediately find that 05:41 - 06:00 exceeds the total video duration, fundamentally ensuring the reliability of the input data;

[0079] Specifically, after verifying the correctness of teaching unit division, it also includes: If the verification passes, proceed to the subsequent steps; if the verification fails, mark the teaching units that fail the verification in the video text content and send them for manual review. After manual review, verify again, and if the verification passes, proceed to the subsequent steps.

[0080] S3. Obtain the concentration metrics of the learner when playing each teaching unit;

[0081] As mentioned above, the concentration metric reflects the degree of attentiveness of the learner during the viewing of the teaching video. Obtaining the concentration metrics of the learner for each teaching unit requires analyzing the facial expressions of the learner for each teaching unit;

[0082] Therefore, obtaining the concentration metrics of the learner when playing each teaching unit includes: Obtaining the facial expression data of the learner during the entire playback of the teaching video, continuously extracting expression feature data through a computer vision model; Associating the expression feature data to the corresponding teaching unit along the time axis; Based on the expression feature data within each teaching unit, calculating the concentration metrics through an emotion recognition model.

[0083] S4. Obtain the playback completeness of each teaching unit of the learner;

[0084] Specifically, the calculation method of the playback completeness here is the ratio of the actual playback duration of the video time period corresponding to the teaching unit to the total duration of this time period. For example, when playing "Introduction to the Pythagorean Theorem", if the trainee skipped the time period from 00:01:48 to 00:03:58, then the playback completeness of the knowledge explanation teaching unit is 80%, and the playback completeness of the example exercise teaching unit is 50%.

[0085] S5. Calculate the basic score for each teaching unit according to the playback completeness and concentration index;

[0086] Specifically, the basic score calculation formula is:

[0087]

[0088] Among them, M, C, and F respectively represent the highest score of the teaching video, playback completeness, and concentration index, and n represents the number of teaching units;

[0089] For example, according to the example situation of "Introduction to the Pythagorean Theorem" in S1, S2, and S4, assuming that in S3, the trainee obtains a concentration index of 0.8 for the knowledge explanation teaching unit and a concentration index of 0.84 for the example exercise teaching unit, and the highest score of this video is 100 points. After substituting into the formula for calculation, the basic score obtained by the trainee in the knowledge explanation teaching unit is 32, and the basic score obtained in the example exercise teaching unit is 21.

[0090] S6. Judge whether to generate test questions according to the function type identifier. If so, generate a test question set according to the knowledge point identifier and the basic score, obtain the answer result to correct the basic score, and obtain the final score of each teaching unit. Otherwise, use the basic score as the final score;

[0091] As described above, judging whether to generate test questions according to the function type is to distinguish the differences in teaching actions. For the function types of knowledge transmission, such as knowledge explanation and example exercises, tests are required. For the function types of expansion, such as knowledge expansion, since the relevance to the core of the course is relatively low, tests are not required;

[0092] Specifically, generating a test question set according to the knowledge point identifier and the basic score includes: calculating the ratio of the basic score to the highest score of the current teaching unit; selecting the associated question bank according to the knowledge point identifier; using the ratio of the basic score to the highest score of the current teaching unit as the difficulty coefficient to match and generate a test question set from the question bank;

[0093] Among them, the associated question bank is a question bank established according to the course knowledge points. The test questions in the question bank are associated with knowledge points and difficulty coefficients. For example, according to the example in "Introduction to the Pythagorean Theorem" in S5, the basic score of this student in the knowledge point explanation teaching unit is 32, and the basic score in the example exercise teaching unit is 21. Then, for the knowledge point explanation teaching unit, a set of test questions with the knowledge point of the Pythagorean theorem formula should be matched from the question bank according to the difficulty coefficient of 0.64. For the example exercise teaching unit, a set of test questions with the knowledge point of the application of the Pythagorean theorem should be matched from the question bank according to the difficulty coefficient of 0.42;

[0094] Moreover, the set of test questions should at least include test questions with a difficulty coefficient higher than, lower than, and equal to the specified difficulty coefficient;

[0095] Specifically, the basic score is corrected to obtain the final score for each teaching unit as follows: Based on the basic score: If a test question with a difficulty coefficient higher than the specified one is answered correctly, points are added and no points are deducted for wrong answers; If a test question with a difficulty coefficient lower than or equal to the specified one is answered wrong, points are deducted and no points are added for correct answers; The maximum threshold of the final score is equal to the highest score of the teaching unit;

[0096] For example, according to the example in "Introduction to the Pythagorean Theorem" in S5, assume that for the knowledge point explanation teaching unit and the example exercise teaching unit, three test questions with the knowledge points of the Pythagorean theorem formula and the application of the Pythagorean theorem are generated respectively. The student's answering situation is that the test questions with a difficulty coefficient lower than the specified one are all answered correctly, and the rest are answered wrong. Assume that the point addition and deduction strategy is to use the lost points of the student in the teaching unit as the total score of the test questions with a difficulty coefficient higher than the specified one, use 1 / 4 of the student's basic score in the teaching unit as the total score of the test questions with a difficulty coefficient equal to the specified one, and use 3 / 4 of the student's basic score in the teaching unit as the total score of the test questions with a difficulty coefficient lower than the specified one. Then the final score of this student in the knowledge point explanation teaching unit is 24, and the final score in the example exercise teaching unit is 15.75.

[0097] This design combines the basic score and test correction, which not only avoids the fallacy of "viewing the playback volume as the learning effect" in traditional online education but also prevents the destruction of learning motivation caused by excessive testing.

[0098] S7. The final scores of all teaching units are weighted and calculated to obtain the comprehensive score of the student.

[0099] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not restrictive. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A method for evaluating the learning quality of adult online education, characterized in that, It includes the following steps: Obtain the text content of the teaching video; Divide the text content into multiple teaching units carrying time range marks, function type identifiers, and knowledge point identifiers according to teaching functions; Obtain the concentration index of the learner when playing each teaching unit; Obtain the playback completeness of each teaching unit of the learner; Calculate the basic score for each teaching unit according to the playback completeness and concentration index; Judge whether to generate test questions according to the function type identifier. If so, generate a test question set according to the knowledge point identifier and the basic score, obtain the answer result to correct the basic score, and obtain the final score of each teaching unit. Otherwise, use the basic score as the final score; Perform weighted calculation on the final scores of all teaching units to obtain the comprehensive score of the learner.

2. The method for evaluating the learning quality of adult online education according to claim 1, characterized in that, The obtaining of the text content of the teaching video includes: Obtain the teaching video; Perform speech recognition on the audio track of the teaching video to obtain the text content with time stamps.

3. The method for evaluating the learning quality of adult online education according to claim 2, characterized in that, The dividing of the text content into multiple teaching units carrying time range marks, function type identifiers, and knowledge point identifiers according to teaching functions includes: Perform semantic segmentation on the text content, and extract teaching function keywords through natural language processing; Based on a preset teaching function classification model, cluster adjacent text paragraphs with similar semantics into independent teaching units; Mark time range marks and function type identifiers for each teaching unit; Mark knowledge point identifiers for each teaching unit according to the core knowledge points in the teaching unit.

4. A method for evaluating the learning quality of adult online education according to claim 3, characterized in that, After outputting the teaching unit set carrying time range marks, function type identifiers, and knowledge point identifiers, verify the correctness of the teaching unit division through at least one of the following methods: Verify whether the time ranges of adjacent teaching units are continuous and non-overlapping; Verify the logical relevance between the function type identifier and the knowledge point identifier; Detect whether the duration of the teaching unit meets the preset threshold range.

5. A method for evaluating the learning quality of adult online education according to any one of claims 4, characterized in that After verifying the correctness of the teaching unit division, it also includes: If the verification is passed, enter the subsequent steps; If the verification fails, mark the teaching units that fail the verification in the video text content and send them for manual review. After manual review, verify again. After the verification is passed, enter the subsequent steps.

6. A method for evaluating the learning quality of adult online education according to any one of claims 1 to 5, characterized in that, The obtaining of the concentration index of the learner when playing each teaching unit includes: Obtain the facial expression data of the learner during the entire playback process of the teaching video, and continuously extract the expression feature data through a computer vision model; Associate the expression feature data with the corresponding teaching unit according to the time axis; Based on the expression feature data in each teaching unit, calculate the concentration index through an emotion recognition model.

7. The method for evaluating the learning quality of adult online education according to claim 6, characterized in that, The formula for the basic score is: Wherein, M, C, and F respectively represent the highest score of the teaching video, playback completeness, and concentration index, and n represents the number of teaching units.

8. A method for evaluating the learning quality of adult online education according to claim 7, characterized in that The generating of the test question set according to the knowledge point identifier and the basic score includes: Calculate the ratio of the basic score to the highest score of the current teaching unit; Select the associated question bank according to the knowledge point identifier; Use the ratio of the basic score to the highest score of the current teaching unit as the difficulty coefficient to match and generate a test question set from the question bank.

9. A method for evaluating the learning quality of adult online education according to claim 8, characterized in that: The test question set at least includes: Test questions higher than the difficulty coefficient; Test questions with a difficulty coefficient lower than the specified one; Test questions with a difficulty coefficient equal to the specified one.

10. A method for evaluating the learning quality of adult online education according to claim 9, characterized in that, The final score for each teaching unit is obtained by correcting the basic score as follows: Based on the basic score: If a test question with a difficulty coefficient higher than the specified one is answered correctly, points are added and no points are deducted for incorrect answers; If a test question with a difficulty coefficient lower than the specified one or equal to the specified one is answered incorrectly, points are deducted and no points are added for correct answers; The maximum threshold of the final score is equal to the highest score of the teaching unit.

Images