CN118365492A - Knowledge point missing and gap-filling analysis method and system based on big data - Google Patents

Abstract

The invention discloses a knowledge point missing detection and leak repair analysis method and system based on big data, wherein the method comprises the following steps of S1: acquiring the total score of a certain subject examination paper of a certain examination of students in the whole target area; obtaining the score of each small question in the subject test paper of the whole students; calculating the expected score of each small question in the subject paper of the current examination of the student; s2: calculating the expected deviation value of the score of each small question in the examination paper of the subject of the examination of the student; s3: calculating a weighted average of the scores of the Z exams of the subject in Z exams before the exam; s4: calculating the priority short-supplementing coefficient of each small question in the subject test paper of the student; s5: and feeding back the questions with the priority short-compensating coefficients to the students and the corresponding teachers in the step S1. The method and the system can calculate the priority short-compensating coefficient, find out the knowledge defects of students according to the priority short-compensating coefficient and make up for the knowledge defects in a targeted way, and are helpful for rapidly solving or improving the knowledge defects.

Description

Knowledge point missing and gap-filling analysis method and system based on big data

Technical Field

The present invention belongs to the field of big data and computer technology, and specifically relates to a knowledge point deficiency checking and gap-filling analysis method and system based on big data.

Background technique

The current big data online education software platform operates as follows: First, the student's test paper is scanned into the online platform; then, the platform grades the objective questions in the test paper, and the subjective questions are graded by the teacher. Finally, the platform calculates the scores of each question in the student's test paper and the total score of the test paper.

The platform can make the following preliminary analyses on the data: 1. Based on the total score of the student's test paper, the ranking of a certain student can be known according to the total score of all students' test papers; 2. Based on the score of a certain question in the student's test paper, the ranking of a certain student in that question can be known according to the score of all students' questions; 3. The highest total score of the test paper and the highest score of each question in the test paper; 4. The average score of each question in the test paper, etc.

Students can use this platform to compare their own scores on each question, the average score of all students on each question, and the highest score of all students on each question, to gain a preliminary understanding of which subjects and questions they have knowledge gaps in.

For individual students, some shortcomings are easy to improve, while others are more difficult to improve.

The knowledge gaps that students understand through the above method are relatively primitive, and cannot accurately guide students to find knowledge gaps that are easy to improve, which is not conducive to students quickly making up for their knowledge gaps.

Summary of the invention

The knowledge point deficiency checking and gap-filling analysis method and system based on big data of the present invention can assist teachers or students to quickly find knowledge points that are easy to improve, thereby assisting teachers and students to quickly fill in the gaps.

In order to achieve the above-mentioned purpose, the knowledge point deficiency checking and gap filling analysis method based on big data of the present invention comprises the following steps:

S1: Obtain the total score of a certain subject test paper of all students in the target area in a certain exam; obtain the score of each question in the test paper of the subject of all students; calculate the expected score of each question in the test paper of the subject of a certain student according to the total score of the test paper of the subject of this exam; the said all students in the target area refer to all students who take the exam with the same test paper;

S2: Calculate the expected deviation value of the score of each question in the test paper of the subject of the student in step S1;

S3: Calculate the weighted average of the student's Z test scores for the subject in the Z tests before the current test;

S4: Calculate the priority compensation coefficient of each question in the subject test paper of the student in step S1;

S5: Feedback the small questions with large short-completion coefficients to the student and the corresponding teacher in step S1.

Furthermore, step S1 specifically includes the following steps:

S1.1: Get the total score of a certain subject in a certain exam for all students in the target area; the matrix composed of the scores of each question in a certain subject in a certain exam for all students is:

;

Each row in the matrix A represents the score of each question in the test paper of this subject for the corresponding student in this exam; there are m students in total, and there are n questions in the test paper of this subject in this exam; A is the score matrix of each question in the test paper of this subject for all students in this exam;

S1.2: Sort the scores of all students in this subject in this exam from large to small; the matrix after sorting is:

;

S1.3: Calculate the expected score of each question in the subject paper of the student in this exam based on the student's total score of the subject paper in this exam;

set up:

;

The value of each row in the matrix S _{is Sk} , k = 1, 2, ... m; let the total score of the student's test paper for this subject _be Etotal , let the expected total score of the student's test paper for this _{subject be Ek} , let Ek equal Etotal ; if there is an expected score _Sk with the same _rank as Etotal in the matrix S _, then the formula for the expected score of the i- th question in the test paper _of this subject for this student is:

(1);

Let E _k be the expected total score of the student in this subject paper of this exam; is the expected score of the student for question i in the subject test paper of this exam; The expected score for the student in the same position;

If the total score of the student's test paper for this subject _{is Etotal} , and there is no same expected score of the same rank in the matrix S , then the formula for the expected score of the i- th question in the test paper of this subject for this student is:

(2);

Among them, S _t+1 is the expected total score of the t+1th row that is less than E _total and closest to E _total ; S _t is the expected total score of the tth row that is greater than E _total and closest to E _total ; u _ti is the score of the ith question in the tth row of the matrix U ; u _(t+1)i is the score of the ith question in the t+1th row of the matrix U ; E _k is the expected total score of the student in this exam for this subject. is the expected score of the student for question i in the subject test paper of this exam; The expected score for the student in the same ranking.

Further, in step S2,

;

Among them, d _i is the deviation between the actual score of question i and the expected score of question i in the test paper of this subject of the student in this exam; is the expected score of the i- th question in the test paper of this subject for this student in this examination; a _i is the actual score of the i- th question in the test paper of this subject for this student in this examination.

Furthermore, in step S3, the ranking of the student's score in the jth test of the subject among all students before the current test is set to p _{j ,} ; The weighted coefficient of the student's score ranking in the subject in the jth examination is f _j ;

Then, the weighted average w of the student's scores in this subject in the z exams before this exam is:

.

Furthermore, let the actual score of the student on the i - th question in the test paper of this subject be ranked among all students as h _i ;

The priority short-filling coefficient formula is:

(3);

In the formula , It is the priority short-compensation coefficient for the i- th question in the test paper of this subject for this student in this exam.

The knowledge point checking and gap-filling analysis system based on big data includes:

The question bank is used to store questions of various subjects and knowledge points. The question information includes text and pictures, and the keywords of the questions are extracted through natural language processing technology;

The priority-compensation coefficient calculation module can calculate the priority-compensation coefficient of each question in the test paper of this subject according to the student's score of each question in the test paper of this subject, the ranking of each question in the test paper of this subject, and the weighted average of the ranking in the Z test papers of this subject before the test paper of this subject;

The question recognition module can identify the subjects and knowledge points of each question in the test paper of this subject through text and pictures; it can extract keywords through natural language processing technology and find similar pictures in the question bank through reverse image search engine, so as to associate the subjects and knowledge points corresponding to each question;

The analysis module is connected with the priority compensation coefficient calculation module and the question identification module respectively; the subject and knowledge point of each question in the test paper of the subject of this exam are obtained respectively, and the priority compensation coefficient of each question in the test paper of the subject of this exam is obtained, and the subject and knowledge point that need to be compensated are judged and analyzed;

The similar exercise recommendation module is connected to the analysis module, can receive instructions from the analysis module, and can find similar exercises corresponding to the knowledge points from the question bank;

The display module can display the exercises of corresponding knowledge points found by the similar exercise recommendation module.

Beneficial effects: This method uses big data to analyze the differences in students' answers to each question in the same test paper, and calculates the priority compensation coefficient based on the student's total score and the ranking of each question. Based on the priority compensation coefficient, the student's knowledge deficiencies are found and targeted compensation is made, which helps to quickly solve or improve these knowledge deficiencies. By finding the knowledge deficiencies that are easiest to improve and improving them, comprehensive and comprehensive learning is avoided, thereby improving the pertinence and effectiveness of learning. It can effectively reduce the excessive academic burden on students.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a flow chart of a method for analyzing knowledge points based on big data;

FIG. 2 is a module diagram of the knowledge point deficiency and gap-filling analysis based on big data of the present invention.

Detailed ways

In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

Embodiment 1: As shown in FIG1 , a knowledge point deficiency checking and gap filling analysis method based on big data includes the following steps:

S1: Get the total score of a certain subject in a certain exam for all students in the target area; get the score of each question in the subject in the exam for all students; calculate the expected score of each question in the subject in the exam for a student based on the total score of the subject in the exam; all students in the target area refer to all students who take the same exam paper; the target area is, for example, the same exam paper in a unified exam in a district or county.

The specific steps include:

S1.1: Get the total score of a certain subject in a certain exam for all students in the target area; get the score of each question in the subject in this exam for all students;

Suppose the matrix composed of the scores of each question in the test paper of this subject for all students is:

;

Each row in the matrix A represents the score of each question in the test paper of this subject for the corresponding student in this exam; there are m students in total and n questions in the test paper of this subject for this exam; A is the score matrix of each question in the test paper of this subject for all students in this exam;

S1.2: Sort the scores of all students in this exam from the highest to the lowest.

Sort the scores of each question in matrix A from large to small; the sorted matrix is as follows:

;

S1.3: Based on the student’s total score for this subject in this exam, calculate the expected score for each question in this subject in this exam.

Specifically, suppose:

;

The value of each row in the matrix S _{is Sk} , k = 1, 2, ... m; let the total score of the student's test paper for this subject _be Etotal , let the expected total score of the student's test paper for this _{subject be Ek} , let Ek equal Etotal ; if there is an expected score _Sk with the same _rank as Etotal in the matrix S _, then the formula for the expected score of the i- th question in the test paper _of this subject for this student is:

(1);

Let E _k be the expected total score of the student in this subject paper of this exam; is the expected score of the student for question i in the subject test paper of this exam; The expected score for the student in the same ranking.

If _{the total} score E of the student's test paper for this subject in this exam does not exist in the matrix S with the same expected score of the same rank, then the formula for the expected score of the i-th question in the test paper of this subject in this exam is:

(2);

Among them, S _t+1 is the expected total score of the t+1th row that is less than E _total and closest to E _total ; S _t is the expected total score of the tth row that is greater than E _total and closest to E _total ; u _ti is the score of the ith question in the tth row of the matrix U ; u _(t+1)i is the score of the ith question in the t+1th row of the matrix U. E _k is the expected total score of the student in this exam. is the expected score of the student for question i in the subject test paper of this exam; The expected score for students in the same ranking.

S2: Calculate the expected deviation value of the score of each question in the test paper of the subject of the student in step S1.

Specifically, the formula is:

;

Among them, d _i is the deviation between the actual score of question i and the expected score of question i in the test paper of this subject of the student in this exam; is the expected score of the i- th question in the test paper of this subject for this student in this examination; a _i is the actual score of the i- th question in the test paper of this subject for this student in this examination.

S3: Calculate the weighted average of the student's scores and rankings in the Z examinations before this examination.

Assume that before this exam, the ranking of the student's score in the jth exam of this subject among all students is p _{j ,} ; Let the weighted coefficient of the student's score ranking in the subject in the jth examination be f _j ;

For example: the student is in the third year of junior high school, and the value of Z is the number of all the examinations for this subject before this examination in the third year of junior high school.

Then, the weighted average w of the student's score ranking in this subject in the Z exams before this one is:

.

S4: Calculate the priority short-answer coefficient for each question in the test paper of this subject for the student in S1.

Suppose the actual score of the student on the i- th question in the test paper of this subject ranks among all students as h _i ;

Specifically, the priority short-filling coefficient formula is:

(3);

It is the priority short-compensation coefficient for the i- th question in the test paper of this subject for this student in this exam.

S5: Feedback the small questions with large short-term coefficients to the student and his/her teacher in S1.

Prioritizing questions with a large coefficient of shortcoming indicates that the teacher needs to quickly make up for the student's shortcomings.

Embodiment 2: In order to more clearly reflect the scheme of embodiment 1, an example is used for illustration in this embodiment.

In step S1.1, it is assumed that there are six students in total, and the matrix A composed of the scores of each question in a certain subject test paper of the six students in a certain exam is:

;

That is, the scores of the first student's test paper for each question are [8, 17, 26, 40] in order; the actual score of the student's test paper is 91 points;

The scores of the questions in the second student's test paper are [10, 15, 29, 39] respectively; the actual score of this student is 93 points.

... and so on.

The scores of the sixth student's test paper for each question are [1, 19, 30, 38], and the actual score of the student is 88 points.

In step S1.2, the scores of all students in the test paper of this subject are sorted from large to small, and the sorted matrix U is:

;

In step S1.3,

;

For the first student whose actual score is 91, since there is no expected score of the same rank as 91 in the matrix S , the expected score of each question is calculated using formula (2) in Example 1: Then, the expected scores of the first student for each question are 7.5, 17.5, 28 and 38 respectively.

specifically:

Take the expected score of the first question of the first student as an example:

From the matrix U , we know that: u _(t+1)i =7, u _ti =8,

From the matrix S we know: Etotal = ₉₁ , S _{t +1} = 89, S _t = 93.

Therefore, combined with formula (2), the expected score of the student for the first question is calculated as:

7+(91-89)(8-7)/(93-89)=7+2/4=7.5. The expected scores of the other questions of this student are similar and will not be repeated here. After calculation, the expected scores of each question of this student are 7.5, 17.5, 28 and 38.

For the second student whose actual score is 93, because there is an expected score of the same rank as 93 in the third row of matrix S , the expected score of each question is calculated using formula (1) in Example 1, that is, the expected score of each question of the second student is directly obtained from the third row of matrix U , which are 8, 18, 29 and 38 respectively. If there are multiple identical expected scores of the same rank in matrix S , you can arbitrarily select one expected score of the same rank, and let the expected score of each question in the expected score of the same rank represent the expected score of each question in the student's test paper.

... and so on.

For the sixth student whose actual score is 88, since there is no expected score of 88 in the matrix S , the expected score of each question is calculated using formula (2) in Example 1:

Then, the expected scores of the sixth student for each question are: 6.4, 16.9, 26.9 and 37.8 respectively.

In step S2, the expected deviation values of the first student's scores for each question are: -0.5, 0.5, 2 and -2.

The expected deviations of the second student's scores on each question are: -2, 3, 0, and -1.

... and so on.

The expected deviation values of the sixth student's scores for each question are: 5.4, -2.1, -3.1 and -0.2.

Step S3, w is the weighted average of the ranking of the student in the subject before the current exam. w has been calculated from the Z times of the subject test papers before the current exam and is regarded as a fixed value.

In step S4, since w in equation (3) is a constant, it can be seen from equation (3) that: The size of is mainly determined by h _i and d _i . Taking the sixth student as an example, his d ₁ =5.4. As the ranking of the student's score on this question increases _, The larger the value is, the more the student needs to improve on the first question.

Embodiment 3: As shown in FIG2 , a knowledge point deficiency checking and gap filling analysis system based on big data includes:

The question bank is used to store questions of various subjects and knowledge points. The question information includes text and pictures, and the keywords of the questions are extracted through natural language processing technology;

The priority-compensation coefficient calculation module can calculate the priority-compensation coefficient of each question in the test paper of this subject according to the student's score of each question in the test paper of this subject, the ranking of each question in the test paper of this subject, and the weighted average of the ranking in the Z test papers of this subject before the test paper of this subject;

The question recognition module can identify the subject and knowledge points of each question in the test paper of this subject through text and pictures; it can extract keywords through natural language processing technology and find similar pictures in the question bank through reverse image search engine, so as to associate the subject and knowledge points corresponding to each question;

The analysis module is connected with the priority-making-up coefficient calculation module and the exercise identification module respectively; it obtains the subject, knowledge point, etc. of each question in the test paper of this subject in this examination, and can obtain the priority-making-up coefficient of each question in the test paper of this subject in this examination, and can judge and analyze the subjects and knowledge points that need to be made up.

The similar exercise recommendation module is connected to the analysis module, can receive instructions from the analysis module, and can find similar exercises corresponding to the knowledge points from the question bank;

The display module can display the exercises of corresponding knowledge points found by the similar exercise recommendation module.

Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the content in the specification, and its technical scope must be determined according to the scope of the claims.

Claims (6)

1. A knowledge point deficiency checking and gap filling analysis method based on big data, characterized in that it includes the following steps: S1: Obtain the total score of a certain subject test paper of all students in the target area in a certain exam; obtain the score of each question in the test paper of the subject of all students; calculate the expected score of each question in the test paper of the subject of a certain student according to the total score of the test paper of the subject of this exam; the said all students in the target area refer to all students who take the exam with the same test paper; S2: Calculate the expected deviation value of the score of each question in the test paper of the subject of the student in step S1; S3: Calculate the weighted average of the student's Z test scores for the subject in the Z tests before the current test; S4: Calculate the priority compensation coefficient of each question in the subject test paper of the student in step S1; S5: Feedback the small questions with large short-completion coefficients to the student and the corresponding teacher in step S1. 2. According to the big data-based knowledge point deficiency checking and gap-filling analysis method of claim 1, step S1 specifically comprises the following steps: S1.1: Get the total score of a certain subject in a certain exam for all students in the target area; the matrix composed of the scores of each question in a certain subject in a certain exam for all students is: ; Each row in the matrix A represents the score of each question in the test paper of this subject for the corresponding student in this exam; there are m students in total, and there are n questions in the test paper of this subject in this exam; A is the score matrix of each question in the test paper of this subject for all students in this exam; S1.2: Sort the scores of all students in this subject in this exam from large to small; the matrix after sorting is: ; S1.3: Calculate the expected score of each question in the subject paper of the student in this exam based on the student's total score of the subject paper in this exam; set up: ; The value of each row in the matrix S , k = 1, 2, ... m; let the total score _{of the student in this exam for this subject be Etotal, let the expected total score of the student in this exam for this subject be Ek, let Ek equal Etotal ;} if _there is _an expected score with the same _{rank as Etotal} in the matrix S , then the formula for the student's expected score for question i in this subject's test paper is: (1); Let E _k be the expected total score of the student in this subject paper of this exam; is the expected score of the student for question i in the subject test paper of this exam; The expected score for the student in the same position; If the total score of the student's test paper for this subject _{is Etotal} , and there is no same expected score of the same rank in the matrix S , then the formula for the expected score of the i- th question in the test paper of this subject for this student is: (2); Among them, S _t+1 is the expected total score of the t+1th row that is less than E _total and closest to E _total ; S _t is the expected total score of the tth row that is greater than E _total and closest to E _total ; is the score of the ith question in the tth row of matrix U ; is the score of the ith question in the t+1th row of the matrix U ; Ek _is the expected total score of the student in this exam. is the expected score of the student for question i in the subject test paper of this exam; The expected score for the student in the same ranking. 3. The method for analyzing knowledge points based on big data according to claim 2 is characterized in that, in step S2, ; in, is the deviation between the actual score of question ith and the expected score of question ith in the test paper of the subject of this exam for this student; is the expected score of the student for question i in the subject test paper of this exam; It is the actual score of the student for question i in this subject test paper. 4. The method for analyzing knowledge points based on big data according to claim 3 is characterized in that, in step S3, the ranking of the student's score in the subject in the jth test before the current test among all students is set as p _{j ,} ; Let the weighted coefficient of the student's score ranking in the subject in the jth examination be f _j ; Then, the weighted average w of the student's scores in the subject in the z exams before this exam is: . 5. The method for analyzing knowledge points based on big data according to claim 4 is characterized in that in step S4, Suppose the actual score of the student on the i- th question in the test paper of this subject ranks among all students as h _i ; The priority short-filling coefficient formula is: (3); In the formula , It is the priority short-compensation coefficient for the i- th question in the test paper of this subject for this student in this exam. 6. A knowledge point deficiency checking and supplementing analysis system based on big data, characterized in that the knowledge point deficiency checking and supplementing analysis method based on big data according to any one of claims 1 to 5 comprises: The question bank is used to store questions of various subjects and knowledge points. The question information includes text and pictures, and the keywords of the questions are extracted through natural language processing technology; The priority-compensation coefficient calculation module can calculate the priority-compensation coefficient of each question in the test paper of this subject according to the student's score of each question in the test paper of this subject, the ranking of each question in the test paper of this subject, and the weighted average of the ranking in the Z test papers of this subject before the test paper of this subject; The question recognition module can identify the subjects and knowledge points of each question in the test paper of this subject through text and pictures; it can extract keywords through natural language processing technology and find similar pictures in the question bank through reverse image search engine, so as to associate the subjects and knowledge points corresponding to each question; The analysis module is connected with the priority compensation coefficient calculation module and the question identification module respectively; the subject and knowledge point of each question in the test paper of the subject of this exam are obtained respectively, and the priority compensation coefficient of each question in the test paper of the subject of this exam is obtained, and the subject and knowledge point that need to be compensated are judged and analyzed; The similar exercise recommendation module is connected to the analysis module, can receive instructions from the analysis module, and can find similar exercises corresponding to the knowledge points from the question bank; The display module can display the exercises of corresponding knowledge points found by the similar exercise recommendation module.