A method for allocating and redistributing online competition papers

Abstract

The present invention belongs to the technical field of paper marking allocation, and discloses a method for paper marking allocation and redistribution for online competitions. Based on the online competition paper marking allocation, the judges are preliminarily allocated test papers. After a period of review, the test papers that have not been reviewed are redistributed based on the online competition paper marking redistribution. This method can achieve the randomness of paper marking allocation, while ensuring that each student is reviewed an equal number of times and the number of papers reviewed by each judge is close to the maximum limit. At the same time, in response to the situation where the marking speed of large-scale competitions is inconsistent, the present invention proposes a "redistribution" method, which can allocate the test papers of judges with slow marking speeds to other judges, and selects the optimal redistribution plan by calculating the variance of the number of allocations to each judge. The present invention is simple to implement, efficient in calculation, and meets the application requirements.

Description

Technical Field

[0001] The present invention relates to the technical field of paper marking and allocation, and in particular to a method for allocating and redistributing paper marking for an online competition. Background Art

[0002] With the development of society, many examinations and competitions are conducted online. Online examinations avoid large offline gatherings of people and have the advantages of convenient competition format, convenient invigilation and marking, and fast score summary. There is no paper waste during the examination process, which is more environmentally friendly.

[0003] During the entire online competition process, multiple systems are involved, including the server, registration end, examination end, invigilation end, marking end, and management end. Relevant functions include candidate registration, examination room creation, candidate allocation, teacher allocation, test paper answering, teacher invigilation, teacher marking, and score summary. Among them, in order to ensure the fairness of the examination marking, the same test paper is often required to be marked by multiple teachers. How to achieve random allocation and avoid unfair results caused by too many repeated test papers between teachers. At the same time, for some large-scale competitions (students participate online and judges mark the papers offline), there are too many applicants, resulting in a large number of test papers allocated to each teacher, and the marking speed varies greatly among teachers. In view of this situation, how to allocate a certain judge's test paper to other judges while ensuring the basic principles of marking allocation is also a very important issue.

[0004] The existing online marking model often has the following basic principles: one candidate's paper is marked by multiple judges (the judges marking the papers of different candidates should be as different as possible to prevent the judges from communicating and comparing with each other during the marking process); each judge cannot mark the same paper multiple times; each candidate's paper should be marked the same number of times; the number of papers marked by each judge should be as consistent as possible (when there is a remainder, one or some of the judges will mark more papers).

[0005] The technical solution described in the Chinese patent "CN114037345A An algorithm for assigning examiners to subjective questions in an exam" is mainly aimed at assigning examinees to subjective questions. It adds a self-assessment link to the assignment process, and adopts a "student carousel" method to assign each student to several other students in sequence. This method will result in too many repeated students when the judges are marking the papers after the competition. When the judges discuss with each other, the randomness of the assignment cannot be guaranteed, and thus the fairness of the test results cannot be guaranteed. It does not meet the first of the above basic principles mentioned in the present invention.

[0006] The technical solution described in the Chinese patent "CN111126845A A dynamic data processing method for assigning marking tasks" is mainly aimed at assigning markers. Its allocation scheme is to assign one to each teacher each time. When the marking teacher finishes, the next one will be assigned until the marking teacher's allocation reaches the upper limit. In this allocation method, the allocation of all marking teachers is not considered comprehensively. Since the marking speed of each judge is different, when the number of times each candidate is marked is greater than 1, it is easy for some candidates' papers to fail to meet the specified requirements and there is no way to make the next allocation. Summary of the invention

[0007] The purpose of the present invention is to provide a method for allocating and redistributing the marking of online competitions, in which the examination papers are randomly allocated and redistributed during the marking process of online competitions. It can solve the problems of insufficient randomness in the initial allocation of student examination papers, uneven number of teacher marking, and inconsistent number of times of student examination papers being marked, and can also allocate the examination papers of the slow-marking judges to other judges when the marking speed of the marking teachers is inconsistent, thereby improving the overall marking speed and completing the entire marking process as quickly as possible.

[0008] The technical solution of the present invention is as follows: a method for online competition paper marking allocation and redistribution, including online competition paper marking allocation and online competition paper marking redistribution; based on the online competition paper marking allocation, the judges are initially allocated test papers, and after a period of review, the test papers that have not been reviewed are redistributed based on the online competition paper marking redistribution.

[0009] The online competition marking assignment includes the following steps:

[0010] Step 1: Input the information of candidates to be assigned, the information of the examiners to be graded, and the number of times each candidate has been graded; based on the input information, calculate the number of candidates taking the exam and the number of examiners to be graded;

[0011] Step 2: Calculate the number of papers each judge should review;

[0012] Step 3: Traverse each judge; determine whether the judge is a node; if not, proceed to step 4; if it is a node, calculate all previously assigned candidates, screen out candidates who have not yet participated in the assignment, and directly assign the screened candidates who have not yet participated in the assignment to the judge, and then proceed to step 4;

[0013] During the screening process, the node judges are recorded. The first node screens candidates who have not appeared once, the second node screens candidates who have not appeared twice, and so on;

[0014] Step 4: Randomly select a candidate and determine whether the candidate needs to be reviewed; determine the number of times the candidate has appeared. If the number of times the candidate has appeared is equal to the number of times the candidate needs to be reviewed, it means that the candidate does not need to be reviewed, and return to step 4; if the number of times the candidate has appeared is less than the number of times the candidate needs to be reviewed, it means that the candidate needs to be reviewed, and continue to step 5;

[0015] Step 5: Determine whether the judge has reviewed the candidate; by screening whether the candidate has been assigned by the judge to review the candidates, to avoid the same judge reviewing the same candidate repeatedly;

[0016] Step 6: Store the examinee's marking information and return to step 4 until the number of students assigned to the examiner reaches the upper limit;

[0017] Step 7: Store the judge's marking information, and then return to step 3 until all judges have been assigned to mark the papers.

[0018] The calculation formula for calculating the number of papers that each judge should review is:

[0019] The number of papers that each judge should mark = the number of candidates taking the exam * the number of times each candidate is marked / the number of judges to mark;

[0020] When the number of papers that each judge is required to review is not an integer, the number will be rounded down.

[0021] The upper limit for the first judge is the "number of papers that each judge should mark" calculated in step 2. The upper limit for each subsequent judge is calculated as follows:

[0022] The number of papers each judge should mark = INT (number of candidates taking the exam * number of times each candidate has been marked - number of candidates assigned to mark) / number of judges who have not marked the papers.

[0023] The node rules of the node judges are as follows: when the number of candidates is N and each candidate is graded M times, the N-(M-1), N-(M-2), ..., N judges are node judges.

[0024] The online competition marking and redistribution comprises the following steps:

[0025] Step 1): Get the test papers to be assigned; get the test papers to be assigned of the teachers who have not completed the marking, and calculate the total number of test papers to be assigned;

[0026] Step 2): Get a copy of the test paper information;

[0027] Step 3): Traverse the marking information of each judge; determine whether the judge has read the test paper. If yes, traverse the next judge; if no, record it and make the judge the assignable judge for the test paper; and so on, until all judges are traversed;

[0028] Step 4): Random assignment: randomly select one judge from all assignable judges as the examiner for the test paper and record it; then return to step 2) until all test papers are assigned;

[0029] Step 5): Calculate the number of papers to be assigned to each judge;

[0030] Step 6): Select the best solution; calculate the variance based on the number of papers to be graded by each judge, and then return to step 1) to redistribute;

[0031] Step 7): Save the result; repeat the above process N times, and then select the one with the smallest variance as the final redistribution result.

[0032] The specific calculation formula for the variance is: the sum of squares of the number of papers redistributed for marking by each judge - the number of judges assigned to mark the papers * the average number of papers redistributed for marking by each judge * the average number of papers redistributed for marking by each judge.

[0033] Compared with the prior art, the beneficial effects of the present invention are as follows: the method proposed by the present invention can achieve the randomness of the distribution of the examination papers, while ensuring that each student is marked an equal number of times, and the number of examination papers marked by each judge is close to the maximum limit. At the same time, in view of the inconsistent examination speed of large-scale competitions, the present invention proposes a "redistribution" method, which can distribute the examination papers of the judges with slow examination speed to other judges, and select the optimal redistribution scheme by calculating the variance of the number of examination papers allocated to each judge. The present invention is simple to implement, efficient in calculation, and meets the application requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 Assign a flow chart for marking online competitions;

[0035] Figure 2 Flowchart of marking and redistribution for online competitions. DETAILED DESCRIPTION

[0036] Figure 1 This is the main flow chart of the marking distribution in the technical solution of the present invention. Figure 1 As shown, the method for online competition marking and distribution proposed by the present invention includes the following steps.

[0037] Step 1: Input information. This section requires inputting the information of candidates to be assigned, the information of the examiners to be graded, and the number of times each candidate has been graded. Then, based on the input information, the number of candidates taking the exam and the number of examiners to be graded are calculated.

[0038] Step 2: Calculate the number of papers each judge should review. The calculation formula is:

[0039] The number of papers each judge should mark = the number of candidates taking the exam * the number of times each candidate has his or her papers marked / the number of judges to mark.

[0040] If the number of papers that each judge is supposed to review is not an integer, round it down. For example, if the result is 3.3, round it down to 3.

[0041] Step 3: Go through each judge.

[0042] Step 4: Determine whether the judge is a node. If not, proceed to step 5; if yes, calculate all previously assigned candidates, screen candidates who have not yet participated in the assignment, and assign these candidates directly to the judge, and then proceed to step 5.

[0043] At the same time, during the screening process, we need to record the node judges. The first node screens candidates who have not appeared once, the second node screens candidates who have not appeared twice, and so on...

[0044] The node rules are as follows: when the number of candidates is N and each candidate is graded M times, the N-(M-1), N-(M-2), ..., N judges are node judges.

[0045] Step 5: Randomly select a candidate.

[0046] Step 6: Determine whether the candidate needs to be reviewed. This part determines the number of times the candidate has appeared. If the number of times the candidate has appeared is equal to the number of times the candidate needs to be reviewed, it means that the candidate does not need to be reviewed, and you can return to step 5. If the number of times the candidate has appeared is less than the number of times the candidate needs to be reviewed, it means that the candidate needs to be reviewed, and you can continue to step 7.

[0047] Step 7: Determine whether the judge has reviewed the candidate. This part is determined by screening whether the candidate has been assigned to review by the judge to avoid the same judge reviewing the same candidate repeatedly.

[0048] Step 8: Store the examinee's marking information, and then return to step 5 until the number of students assigned to the judge reaches the upper limit. The upper limit for the first judge is the "number of papers each judge should mark" calculated in step 2. The upper limit calculation formula for each subsequent judge is as follows:

[0049] Number of papers to be marked by each judge = INT (number of candidates taking the exam * number of times each candidate has been marked - number of candidates assigned to mark) / number of judges who have not marked the papers

[0050] Step 9: Store the information of the judge's marking, and then return to step 3 until all judges have been assigned to mark the papers.

[0051] Figure 2This is the main flow chart of the redistribution of examination papers in the technical solution of the present invention. Figure 2 As shown, the method for redistributing online competition marking proposed by the present invention comprises the following steps:

[0052] Step 1): Get the test papers to be assigned. Get the test papers to be assigned of the teacher who is slow in marking, and calculate the total number of test papers to be assigned.

[0053] Step 2): Get a copy of the test paper information.

[0054] Step 3): Traverse the marking information of each judge. Determine whether the judge has read the test paper. If yes, traverse the next judge. If no, record and make the judge the assignable judge of the test paper. Repeat this process until all judges have been traversed.

[0055] Step 4): Random assignment. Randomly select one of all assignable judges as the examiner for the test paper and record it. Then, return to step 2) until all test papers are assigned.

[0056] Step 5): Calculate the number of papers to be graded by each judge.

[0057] Step 6): Select the best solution. Calculate the variance based on the number of papers to be graded by each scorer, and then return to step 1) to redistribute.

[0058] The specific calculation formula is: the sum of the squares of the number of papers assigned to be marked by each judge - the number of judges assigned to mark the papers * the average number of papers assigned to be marked by each judge * the average number of papers assigned to be marked by each judge.

[0059] Step 7: Save the result. Repeat the above process N times (N=10), and then select the one with the smallest variance as the final redistribution result.

[0060] For example, in a national competition, there were 1,427 candidates and 26 invigilators. Each candidate's test paper was reviewed by two teachers. The final result of the first distribution was: 20 teachers had 110 test papers, and 6 teachers had 109 test papers. After three hours of marking, all but one teacher had finished marking, and the teacher had 30 papers left, so a redistribution was made. Except for this teacher, the remaining teachers were assigned 0-3 test papers. In the end, everyone finished marking all the test papers within 10 minutes.

[0061] In the allocation process, the present invention analyzes the allocation method, proposes the concept of "node", and performs data verification through "node rules", thereby avoiding the situation where some examinees cannot be allocated and the number of times of examination does not meet the requirements. At the same time, the method ensures that the number of examination papers initially allocated by each examiner is almost the same, achieving random allocation and ensuring fairness and justice in examination.

Claims

1. A method for allocating and redistributing online competition marking, characterized in that: Including online competition paper marking allocation and online competition paper marking reallocation; based on the online competition paper marking allocation, the judges are initially allocated papers, and after a period of marking, the papers that have not been marked are reallocated again based on the online competition paper marking reallocation; The online competition marking assignment includes the following steps: Step 1: Input the information of candidates to be assigned, the information of the examiners to be graded, and the number of times each candidate has been graded; based on the input information, calculate the number of candidates taking the exam and the number of examiners to be graded; Step 2: Calculate the number of papers each judge should review; Step 3: Traverse each judge; determine whether the judge is a node; if not, proceed to step 4; if it is a node, calculate all previously assigned candidates, screen out candidates who have not yet participated in the assignment, and directly assign the screened candidates who have not yet participated in the assignment to the judge, and then proceed to step 4; During the screening process, the node judges are recorded. The first node screens candidates who have not appeared once, the second node screens candidates who have not appeared twice, and so on; The node rules of node judges are as follows: when the number of judges is P and the number of times each candidate is graded is M, the P-(M-1), P-(M-2), ..., P judges are node judges; Step 4: Randomly select a candidate and determine whether the candidate needs to be reviewed; determine the number of times the candidate has appeared. If the number of times the candidate has appeared is equal to the number of times the candidate needs to be reviewed, it means that the candidate does not need to be reviewed, and return to step 4; if the number of times the candidate has appeared is less than the number of times the candidate needs to be reviewed, it means that the candidate needs to be reviewed, and continue to step 5; Step 5: Determine whether the assessor has reviewed the candidate; By screening whether the candidate has been judged among the candidates assigned to be graded by the assessor, it is possible to avoid the same assessor marking the same candidate repeatedly; Step 6: Store the examinee's marking information and return to step 4 until the number of students assigned to the examiner reaches the upper limit; Step 7: Store the judge's marking information, and then return to step 3 until all judges have been assigned to mark the papers.

2. The method for allocating and redistributing online competition marking according to claim 1, characterized in that: The calculation formula for calculating the number of papers that each judge should review is: The number of papers that each judge should mark = the number of candidates taking the exam * the number of times each candidate is marked / the number of judges to mark; When the number of papers that each judge is required to review is not an integer, the number will be rounded down.

3. The method for allocating and redistributing online competition marking according to claim 2, characterized in that: The first judge's upper limit is the "number of papers each judge should mark" calculated in step 2. The upper limit of each subsequent judge's marking is calculated as follows: The number of papers that each examiner should mark = INT((number of candidates taking the exam * number of times each candidate has been marked – number of candidates assigned to mark) / number of examiners who have not marked the papers), INT is a round-down function.

4. The method for allocating and redistributing online competition marking according to any one of claims 1 to 3, characterized in that: The online competition marking and redistribution comprises the following steps: Step 1): Get the test papers to be assigned; get the test papers to be assigned of the teachers who have not completed the marking, and calculate the total number of test papers to be assigned; Step 2): Get a copy of the test paper information; Step 3): Traverse the marking information of each judge; determine whether the judge has read the test paper. If so, traverse the next judge; if not, record it and make the judge the assignable judge for the test paper; And so on, until all the judges have been traversed; Step 4): Random assignment: randomly select one judge from all assignable judges as the examiner for the test paper and record it; then return to step 2) until all test papers are assigned; Step 5): Calculate the number of papers to be assigned to each judge; Step 6): Select the best solution; calculate the variance based on the number of papers to be graded by each judge, and then return to step 1) to redistribute; Step 7): Save the result; repeat the above process N times, and then select the one with the smallest variance as the final redistribution result.

5. The method for allocating and redistributing online competition marking according to claim 4, characterized in that: The specific calculation formula for the variance is: the sum of squares of the number of papers assigned to be graded by each judge - the number of judges assigned to grade the papers * the average number of papers assigned to be graded by each judge * the average number of papers assigned to be graded by each judge.

Images