Teaching data sharing method based on ant lion algorithm

Abstract

The invention relates to the technical field of data sharing, and discloses a teaching data sharing method based on an ant lion algorithm, which comprises the following steps of: marking a keyword I on a recorded teaching video, marking a keyword II on a test question, and marking a keyword II on the test question; then the test questions are captured by using the ant lion algorithm on the basis of the first keyword, and the teaching video and the captured test questions are linked to form a data chain, so that the process of manually selecting and editing the questions by teaching personnel is avoided, the online teaching efficiency is greatly improved while rapid association and sharing of the teaching data are realized, and the teaching experience is improved. And the workload of online teaching is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of data sharing, and particularly to a teaching data sharing method based on the antlion algorithm. Background Art

[0002] Online teaching is a remote teaching activity based on network operation, which is a method or technology for discussing teaching content, arranging teaching processes, and conducting teaching training with students in a dialogue manner. Cloud service refers to obtaining the required services in a demand-based and easily expandable manner through the network, and its main based on the increase, use, and interaction modes of Internet-related services, usually involving providing dynamically expandable and often virtualized resources through the Internet.

[0003] To achieve the above objectives, those skilled in the art have disclosed a teaching data sharing system, with the publication number: CN115454951 B; by setting a data receiving module, a data storage module, a data processing module, and a data pushing module, determining the association relationship between each operation step through the received error operation data of the user terminal, representing the similarity between operation steps through the association relationship, identifying error operation steps with higher similarity during the experimental operation process of the user terminal, and correspondingly dividing them into the same data group, determining other operation steps associated with the corresponding operation steps in the data group by calling the data group, and pushing the reference materials corresponding to other operation steps, establishing the association between each operation step, being able to accurately discover the similar errors that occur during the experimental operation process of the user terminal, and based on this, accurately pushing the reference materials corresponding to the subsequent operation steps similar to the errors that have occurred, making the material support more accurate and fast.

[0004] However, the technical solution of the above patent mainly uses the relevance of customer operations to push corresponding data, this method cannot meet the personalized teaching needs of teachers, and at the same time cannot accurately match corresponding teaching data for teachers, especially lacking relevance in the use of teaching data (mainly teaching videos and test questions), unable to quickly form a teaching data chain according to needs, and thus unable to effectively reduce the difficulty and workload of online teaching and improve the efficiency of online teaching. Therefore, those skilled in the art need a teaching data sharing method that can automatically match teaching videos and test question materials. Summary of the Invention

[0005] The purpose of the present invention is to solve the above problems and design a teaching data sharing method based on the antlion algorithm.

[0006] To achieve the above object, the technical solution of the present invention is a teaching data sharing method based on the antlion algorithm, and the system includes the following steps:

[0007] Step 1: Establish a teaching video database and a question database. Store teaching videos in the teaching video database, and mark a number of keyword ones on the video files of the teaching videos. Store a number of questions in the question database, and mark a number of keyword twos on the questions. It should be noted that the teaching videos can be uniformly recorded by teaching institutions or recorded by teachers themselves, while the questions are batch-crawled from the Internet using a search engine and then stored in the question database.

[0008] Step 2: Based on the keyword ones on the teaching videos, the teaching video database can automatically capture the questions related to the keyword ones in the question database through the ant lion algorithm and form a one-to-many data link.

[0009] Step 3: After a teaching video in the teaching video database is selected, the question database can automatically output a number of questions that form a data link with the teaching video. Teachers select the questions that form a data link with the teaching video according to teaching needs and form a corresponding test paper.

[0010] Step 4: Output the selected teaching video and the formed test paper into a compressed file, and the teacher sends the compressed file to the students for online teaching.

[0011] The number of keyword ones on the teaching video is at least one; the number of keyword twos on the question is more than one.

[0012] The keyword ones include but are not limited to: the knowledge points explained in the teaching video and the difficulty of the content explained in the teaching video; the keyword twos include but are not limited to: the knowledge points included in the question and the difficulty of the question setting.

[0013] The process by which the teaching video database automatically captures relevant questions through the ant lion algorithm is as follows:

[0014] Set the keyword two on the question as the ant, and establish the ant random walk expression as:

[0015] X(t)=[0,cussum(2r(t 1 )-1),...,cussum(2r(t n )-1)] (1)

[0016] In the formula: X(t) is the step set of the ant random walk; cussum is to calculate the cumulative sum; t is the number of steps of the random walk; r(t) is a random function, defined as:

[0017]

[0018] In the formula: rand is a random number in [0,1];

[0019] Perform normalization processing on formula (2) to obtain:

[0020]

[0021] Where: a i is the minimum value of the random walk of the i-th dimensional variable, b i is the maximum value of the random walk of the i-th dimensional variable, and are the maximum and minimum values of the i-th dimensional variable in the t-th iteration respectively;

[0022] Set the first keyword on the teaching video as antlion. According to the influence of the trap made by the antlion on the random walking route of the ant, the following formula is established:

[0023]

[0024] Where: c is the minimum value of all variables in the t-th iteration, d is the maximum value of all variables in the t-th iteration, is the position of the selected i-th antlion in the t-th iteration;

[0025] According to the adaptive mechanism, the simulation equation for the antlion to randomly capture the ant is:

[0026]

[0027] Where: I is the proportionality coefficient, T is the maximum number of iterations, and v is a number that changes as the number of iterations increases;

[0028] When the fitness value of the ant is smaller than that of the antlion, it is considered that the antlion has captured it. At this time, the antlion will update its position according to the position of the ant:

[0029]

[0030] Where: is the position of the i-th ant in the t-th iteration, is the position of the t-th ant in the (t + 1)-th iteration, and f is the fitness function.

[0031] The position of the t-th ant in the (t + 1)-th iteration is determined by Equation (8);

[0032]

[0033] Where: is the value generated by the ant randomly walking the l-th step around an antlion selected by roulette in the t-th iteration, is the value generated by the ant randomly walking the l-th step around the elite antlion in the t-th generation.

[0034] The fitness function J is as follows:

[0035]

[0036] Among them, H δ (a) represents the Huber loss function, which is defined as follows:

[0037]

[0038] Among them, y(t) represents the actual Jaccard similarity coefficient between keyword one and keyword two at time t, represents the standard Jaccard similarity coefficient between keyword one and keyword two at time t, and δ is the threshold parameter in the Huber loss function, which is used to distinguish small errors and large errors, and the value of δ is 0.001.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1. The present invention uses the ant lion algorithm to quickly generate the data chain of teaching videos and test questions, and uses the teaching data chain to assist teachers in completing the preparation process of online teaching materials, and then realizes interactive online teaching through the demonstration of teaching content and the answering of test questions, thereby more effectively improving the online teaching effect;

[0041] 2. After the teacher selects the teaching video, the present invention can automatically generate the corresponding test question data chain, so as to achieve the purpose of efficiently integrating teaching resources and realizing teaching data sharing, effectively reducing the difficulty and workload of online teaching, and at the same time improving teaching efficiency, and can better meet the needs of online teaching. Description of the Drawings

[0042] Figure 1 is the flowchart of a teaching data sharing method based on the ant lion algorithm described in Embodiment 1 of the present invention;

[0043] Figure 2 is the standard Jaccard similarity coefficient table of keyword one and keyword two at different times described in Embodiment 1 of the present invention. Detailed Embodiments

[0044] The embodiments of the present invention will be specifically described below with reference to the drawings:

[0045] Embodiment 1

[0046] A teaching data sharing method based on the ant lion algorithm, as Figure 1 and Figure 2 shown, the system includes the following steps:

[0047] Step 1: Establish a teaching video database and a question database. Store teaching videos in the teaching video database, and mark a number of keyword 1 on the video files of the teaching videos. Store a number of questions in the question database, and mark a number of keyword 2 on the questions. It should be noted that the teaching videos can be uniformly recorded by teaching institutions or recorded by teachers themselves.

[0048] Step 2: Based on keyword 1 on the teaching video, the teaching video database can automatically capture the questions in the question database related to keyword 1 through the ant lion algorithm and form a one-to-many data link.

[0049] Step 3: After a teaching video in the teaching video database is selected, the question database can automatically output a number of questions that form a data link with the teaching video. The teacher selects the questions that form a data link with the teaching video according to teaching needs and forms a corresponding test paper.

[0050] Step 4: Output the selected teaching video and the formed test paper into a compressed file, and the teacher sends the compressed file to the students for online teaching.

[0051] The number of keyword 1 on the teaching video is at least one. When the number of keyword 1 on the teaching video exceeds one, the keyword 1 on the teaching video is sorted from largest to smallest according to the proportion explained in the teaching video. The number of keyword 2 on the question is greater than one, and the several keyword 2 on the question are randomly sorted.

[0052] The keyword 1 includes but is not limited to: the knowledge points explained in the teaching video and the difficulty of the content explained in the teaching video. The keyword 2 includes but is not limited to: the knowledge points included in the question and the difficulty of the question. The teaching video database captures the relevant questions in the question database one by one according to the sorting of keyword 1.

[0053] It should be noted that the difficulty of the content explained in the teaching video and the difficulty of the question can be divided according to the teaching level. Among them, the teaching level refers to the level at which the learner is in the learning process, and the level is divided according to the K12 education system.

[0054] The process by which the teaching video database automatically captures relevant questions through the ant lion algorithm is as follows:

[0055] Set keyword 2 on the question as an ant, and establish an ant random walk expression as:

[0056] X(t)=[0,cussum(2r(t 1 )-1),...,cussum(2r(t n )-1)] (1)

[0057] Where: \(X(t)\) is the set of steps of the ant's random walk; cussum is to calculate the cumulative sum; \(t\) is the number of steps of the random walk; \(r(t)\) is a random function, defined as:

[0058]

[0059] Where: rand is a random number in \([0, 1]\);

[0060] Normalizing Equation (2) gives:

[0061]

[0062] Where: \(a\) i is the minimum value of the random walk of the \(i\)-th dimensional variable, \(b\) i is the maximum value of the random walk of the \(i\)-th dimensional variable, and are the maximum and minimum values of the \(i\)-th dimensional variable in the \(t\)-th iteration respectively;

[0063] Assume that the first keyword in the teaching video is the antlion. According to the influence of the trap made by the antlion on the random walk route of the ant, the following formula is established:

[0064]

[0065] Where: \(c\) is the minimum value of all variables in the \(t\)-th iteration, \(d\) is the maximum value of all variables in the \(t\)-th iteration, is the position of the \(i\)-th selected antlion in the \(t\)-th iteration;

[0066] According to the self-adaptive mechanism, the simulation equation for the antlion to randomly capture the ant is:

[0067]

[0068] Where: \(I\) is the proportionality coefficient, \(T\) is the maximum number of iterations, \(v\) is a number that changes as the number of iterations increases;

[0069] When the fitness value of the ant is smaller than that of the antlion, it is considered that the antlion has captured it. At this time, the antlion will update its position according to the position of the ant:

[0070]

[0071] Where: is the position of the \(i\)-th ant in the \(t\)-th iteration, is the position of the \(t\)-th ant in the \((t + 1)\)-th iteration, \(f\) is the fitness function.

[0072] It should be noted that the antlion (i.e., the first keyword) is in

[0073] The position of the $t$-th ant in the $(t + 1)$-th iteration is determined by Equation (8);

[0074]

[0075] where: is the value generated by the $l$-th step of the random walk of the ant around an antlion selected by roulette in the $t$-th iteration, is the value generated by the $l$-th step of the random walk of the ant around the elite antlion in the $t$-th generation.

[0076] The fitness function $J$ is as follows:

[0077]

[0078] where $H$ δ (a) represents the Huber loss function, which is defined as follows:

[0079]

[0080] where $y(t)$ represents the actual Jaccard similarity coefficient between keyword one and keyword two at time $t$, represents the standard Jaccard similarity coefficient between keyword one and keyword two at time $t$, and $\delta$ is the threshold parameter in the Huber loss function, which is used to distinguish small errors and large errors, and the value of $\delta$ is 0.001.

[0081] In the process of implementing this method, the teacher can generate multiple compressed files according to teaching needs and send them to students one by one in groups or send them specifically according to the situations of different students. After receiving the compressed files, the students decompress the teaching videos and test papers in the compressed files, first start the teaching videos to learn the corresponding content, then answer the questions on the test papers, and send the test papers to the teacher after completion. The teacher adjusts the learning situation of the students by grading the test papers, so as to improve the learning efficiency of the learners.

[0082] The creative point of this application lies in using the antlion algorithm to process the linking problem between teaching video materials and multiple test questions, and using the capture characteristics of the antlion algorithm to improve the integration efficiency between teaching videos and test questions, so as to achieve the effect of efficient utilization and sharing of data.

[0083] It should be noted that the calculation process of the standard Jaccard similarity coefficient of keyword one and keyword two is as follows:

[0084] First, select a set of standard texts, such as a group of subject-related articles, academic papers, or standard teaching videos, etc. These texts will be used as the basis for calculating the Jaccard similarity coefficient. Among them, the standard teaching video refers to the template video recognized within this subject, which can be selected by multiple teaching experts in this subject.

[0085] Second, count the occurrences of keyword one and keyword two in each standard text or standard teaching video. The keywords one and two can be extracted from the text through text processing tools (such as the NLTK or spaCy libraries in Python), and then record the text numbers (or indices) where each keyword appears.

[0086] Third, calculate the Jaccard similarity coefficient:

[0087] (1) Construct sets: Let the set of texts where keyword one appears be set A, and the set of texts where keyword two appears be set B. For example, if keyword one appears in texts A1, 3, 5, then A = {1, 3, 5}; if keyword B appears in texts 3, 4, 5, then B = {3, 4, 5}.

[0088] (2) Calculate the sizes of the intersection and union: Calculate the number of elements in the intersection (A ∩ B) of sets A and B, that is, the number of texts where both keywords appear; and the number of elements in their union (A ∪ B), that is, the total number of texts where the two keywords appear (removing duplicates). In the above example, (A ∩ B) = {3, 5}, and the number of elements is 2; (A ∪ B) = {1, 3, 4, 5}, and the number of elements is 4.

[0089] (3) Calculate the Jaccard similarity coefficient: According to the formula Divide the number of intersection elements by the number of union elements. In the example,

[0090] It should be noted that the value range of the Jaccard similarity coefficient is from 0 to 1; where 0 indicates no correlation: when it is 0, it means that the text sets in which the two keywords appear have no intersection, that is, they never appear in the same text, indicating that the two keywords have little correlation. For example, "historical novels" and "quantum physics" may rarely appear in the same text under normal circumstances, and the Jaccard similarity coefficient may be close to 0; while 1 indicates complete correlation: when it is 1, it means that set A and set B are exactly the same, that is, the two keywords always appear in the same text, indicating that they have a very strong correlation. For example, in medical literature, "diabetes" and "hyperglycemia" often appear together, and their Jaccard similarity coefficient may be relatively high, close to 1; and the intermediate value indicates different degrees of correlation: a value between 0 and 1 of the coefficient indicates different degrees of correlation; generally speaking, the higher the value, the stronger the correlation; the lower the value, the weaker the correlation; for example, a Jaccard similarity coefficient of 0.3 may indicate that the two keywords have a certain correlation, but not as close as when the coefficient is 0.7.

[0091] Similarly, the calculation process of the actual Jaccard similarity coefficient between keyword one and keyword two is as follows:

[0092] First, select a teaching video and use a keyword one marked on the teaching video as the basis for calculating the Jaccard similarity coefficient, and at the same time determine keyword two related to keyword one in the question database.

[0093] Second, count the occurrences of keyword one and keyword two in each question in the question database and extract keyword one and keyword two, and then record the question numbers (or indexes) where each keyword appears.

[0094] Third, calculate the Jaccard similarity coefficient:

[0095] (1) Construct sets: Let the set of questions in which keyword one appears be set A, and the set of questions in which keyword two appears be set B. For example, if keyword one appears in questions 1, 2, 4, 5, 7 of set A, then A = {1, 2, 4, 5, 7}; if keyword two appears in questions 2, 4, 5, 6, 7, 8 of set B, then B = {2, 4, 5, 6, 7, 8}.

[0096] (2) Calculate the sizes of the intersection and union: Calculate the number of elements in the intersection (A ∩ B) of set A and set B, that is, the number of questions in which the two keywords appear together; and the number of elements in their union (A ∪ B), that is, the total number of questions in which the two keywords appear (removing duplicates). In the above example, (A ∩ B) = {2, 4, 5, 7}, and the number of elements is 4; (A ∪ B) = {1, 2, 4, 5, 6, 7, 8}, and the number of elements is 7.

[0097] (3) Calculate the Jaccard similarity coefficient: According to the formula Divide the number of intersection elements by the number of union elements, and the Jaccard similarity coefficient is 0.57.

[0098] The above technical solution only reflects the preferred technical solution of the technical solution of the present invention. Some changes that those skilled in the art may make to some parts thereof all reflect the principle of the present invention and fall within the protection scope of the present invention.

Claims

1. A teaching data sharing method based on the ant lion algorithm, characterized in that: The system includes the following steps: Step 1, establishing a teaching video database and a test question database, storing teaching videos in the teaching video database, marking a number of key words 1 on the video files of the teaching videos, and storing a number of test questions in the test question database, marking a number of key words 2 on the test questions; Step 2, based on the keyword 1 in the teaching video, the test questions in the test question database are automatically captured by the ant lion algorithm and a data link is formed; Step 3: After a teaching video in the teaching video database is selected, the test question database can automatically output test questions that form a data link with the teaching video and form a corresponding test paper; Step 4: Output the selected teaching video and the generated test volume in a compressed format.

2. The teaching data sharing method based on the ant lion algorithm according to claim 1, characterized in that: The number of keyword one in the teaching video is at least one, and the number of keyword two in the test question is greater than one.

3. A teaching data sharing method based on the ant lion algorithm according to claim 2, characterized in that: The keyword one includes but is not limited to: the knowledge points explained in the teaching video and the difficulty of the content explained in the teaching video; the keyword two includes but is not limited to: the knowledge points included in the test questions and the difficulty of the test questions.

4. A teaching data sharing method based on the ant lion algorithm according to claim 3, characterized in that: The process of automatically capturing relevant test questions in the teaching video database through the ant lion algorithm is as follows: Assume that the second keyword in the test question is ant, and establish the expression of ant random walk as: X(t)=[0,cussum(2r(t1)-1),...,cussum(2r(t n )-1)] (1) Where: X(t) is the set of steps of the ant's random walk; cussum is the calculated cumulative sum; t is the number of steps of the random walk; r(t) is a random function, defined as: Where: rand is a random number in [0,1]; Normalizing formula (2) yields: Where: a i is the minimum value of the random walk of the i-th dimension variable, b i is the maximum value of the random walk of the i-th dimension variable, and are the maximum and minimum values ​​of the i-th dimension variable in the t-th iteration respectively; Assuming that the first keyword in the teaching video is antlion, the following formula is established based on the impact of the traps made by antlions on the random walk routes of ants: Where: c is the minimum value of all variables at the tth iteration, d is the maximum value of all variables at the tth iteration, is the position of the selected ith antlion at the tth iteration; According to the adaptive mechanism, the simulation equation for ant lions to randomly capture ants is established as: Where: I is the proportional coefficient, T is the maximum number of iterations, and v is a number that changes as the number of iterations increases; When the fitness value of an ant is smaller than that of an antlion, it is considered that the antlion has captured it. At this time, the antlion will update its position according to the position of the ant: Where: is the position of the i-th ant at the t-th iteration, is the position of the t-th ant at the t+1th iteration, and f is the fitness function.

5. The teaching data sharing method based on the ant lion algorithm according to claim 4 is characterized in that: The position of the t-th ant at the t+1-th iteration is determined by formula (8); Where: is the value generated by the ant's random walk in step l around an antlion selected by roulette at iteration t, is the value generated by the ant's random walk in the lth step around the elite antlion of the tth generation.

6. The teaching data sharing method based on the ant lion algorithm according to claim 4 is characterized in that: The fitness function J is as follows: Among them, H δ (a) represents the Huber loss function, which is defined as follows: in, y(t) represents the actual Jaccard similarity coefficient between keyword 1 and keyword 2 at time t. It represents the standard Jaccard similarity coefficient between keyword 1 and keyword 2 at time t. δ is the threshold parameter in the Huber loss function, which is used to distinguish small errors from large errors. The value of δ is 0.001.

Images