CN111539588A - Method for predicting learning result by correcting short answer question score - Google Patents
Method for predicting learning result by correcting short answer question score Download PDFInfo
- Publication number
- CN111539588A CN111539588A CN202010653451.0A CN202010653451A CN111539588A CN 111539588 A CN111539588 A CN 111539588A CN 202010653451 A CN202010653451 A CN 202010653451A CN 111539588 A CN111539588 A CN 111539588A
- Authority
- CN
- China
- Prior art keywords
- question
- answer
- score
- test
- knowledge point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06395—Quality analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/20—Education
- G06Q50/205—Education administration or guidance
Abstract
The invention discloses a method for predicting learning achievement by correcting simple answer scores. The method determines an answer set according to a score list, calls a first profile in a first question bank, calls a second profile in a second question bank, and determines a test question adjusting coefficient of each test question and a correction score of each question-answer identification. And finally, determining the learning result of each knowledge point code according to the question-answer score and the correction score. The invention determines the mastery degree of each question and answer knowledge point by the answerers through the corrected test paper scores, and can more accurately predict the learning achievement. The learning results of a plurality of students in a unified class or the same specialty are collected, and the teaching quality can be evaluated.
Description
Technical Field
The invention relates to a method for predicting learning achievement by correcting simple answer question scores, belonging to the technology of processing and predicting subject achievement data.
Background
CN110929020A discloses a knowledge point mastery analysis method based on test results. The method systematically disperses different knowledge points into different test questions, and combines the error checking of different test questions of students to form a knowledge point mastering database for each student. In the specific test, the test content is generated into a question number, a knowledge point number is generated for each knowledge point, and the knowledge point mastery degree of the test result of one student in each list is analyzed. The knowledge point mastery degree calculation formula in the case considers the difficulty coefficient and student capability of the knowledge point, but does not consider the time length factor influencing the test score.
Some applicants have identified the impact of the time duration factor on performance, such as the performance prediction method of CN 110826796A. However, in this case, the duration is used as a preset parameter and not as an adjustment item. The related art related to the present application also discloses a method for processing result data disclosed in CN108710602A, which is a method for converting and processing result data by using various data tables such as a result data table and a result analysis table, and can be referred to in the present application. The method of knowledge point set recommendation of CN108573628A discloses a solution for creating question bank according to knowledge point category. CN104809677A discloses an automatic scoring method based on statistics and analysis of knowledge point grasping conditions, which mainly relates to scoring steps, and does not disclose how to analyze knowledge point grasping conditions.
In addition, the question type structure of the short answer questions is more complicated than the blank filling questions and the selection questions. Documents CN101587513A and CN110175585A only disclose examination paper marking methods for simple answers, and few inventors propose analytical prediction models for simple answers.
Therefore, the prior art lacks a method for determining learning outcome by correcting the question-answer score of a short-response question.
Disclosure of Invention
The invention provides a method for predicting learning achievement by correcting the score of a short answer question, which determines the mastery degree of each question and answer knowledge point of an answerer by correcting the score of a test paper and can more accurately predict the learning achievement.
A method for predicting learning outcome by correcting short-answer question score is characterized in that,
The first question bank stores a plurality of first profiles, and the first profiles comprise question identifications and question stem durationAnd multiple question-answer identifiers corresponding to the question stem identifier;
The second question bank stores a plurality of second profiles, and the second profiles comprise question and answer identifiers, knowledge point codes and question and answer scoresLength of question and answer;
The score database stores score lists of multiple answerers, each score list has multiple question-answer identifiers and question-answer scores uniquely corresponding to the question-answer identifiers;
Determining a set of answers based on the score listThe answer set comprises any question-answer mark with a non-zero question-answer score,for anyAll satisfy,;
According to the answer set, calling the first profile with the question-answer identification in the first question bank to determine the question amount regulating coefficient,。
Calling a second profile with the question-answer identifier in a second question bank according to the answer set to determine the test question adjustment coefficient of each test question,;
According to question-answer scoreAnd the correction scoreDetermining learning outcome for each knowledge point code,。
In the present invention, each knowledge point code contains one or more knowledge points.
In the invention, the learning result of the knowledge point code corresponding to the question-answer identification with any question-answer score not equal to zero is output.
In the present invention, in the case of the present invention,as the total number of questions that have been answered,the total number of questions and answers that have been answered in the test question.
In the invention, the learning achievement of a plurality of answerers for coding any knowledge point is predicted, the average learning achievement is obtained, and the teaching quality is determined.
in the invention, each test question has more than one question and answer, and each test question mark corresponds to a plurality of question and answer marks.
The method for predicting the learning result by correcting the simple answer score distributes the answer time of each question and determines the correction coefficient according to the time distribution. And adjusting the score of each question and answer by using the correction coefficient, and finally summarizing to determine the test paper score. And determining the mastery degree of the answerer on each question and answer knowledge point according to the corrected test paper scores, so that the learning result can be predicted more accurately. The learning results of a plurality of students in a unified class or the same specialty are collected, and the teaching quality can be evaluated.
Drawings
Fig. 1 is a flowchart of a method for predicting learning outcome by correcting the score of a simple response question according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention.
The method for predicting learning outcome by modifying the score of a simple response question according to the present invention as shown in fig. 1 is used for predicting learning outcome according to the score of a test paper containing a plurality of simple response questions. Such test papers are, for example, the subjective questions of judicial examinations. The method mainly comprises the following steps:
determining total length of test. The total test duration refers to the total test duration of the test paper and represents the longest allowable answer of all the simple answersThe subject time is, for example, 2 hours or 3 hours. Before the examination, the operator assigns the total length of the examination to each short-response question to generate the length of the examination question. Meanwhile, the test question time length is divided into the time length for reading the question stem and the time length for completing each question and answer. If the answerer completes all the simple answers, the influence of time factors on answer quality can be not considered. If the answerer only completes partial simple answers, the actual answering efficiency of the partial simple answers is lower than the preset answering efficiency, and the proficiency of the answerer on the question-answering knowledge point is evaluated. It is necessary to have a reasonable allocation duration impact.
The first question bank stores a plurality of first profiles. The first profile comprises test question identification, test question duration and question stem durationAnd multiple question-answer identifiers corresponding to the question stem identifier. The question stem duration refers to the basic duration of reading the question stem by the answering machine. Each test question has more than one question and answer, so each test question mark corresponds to a plurality of question and answer marks. Based on the question and answer identification, a unique stem identification and stem duration may be determined from the first profile of the first question bank. For example, a first profile of a test paper for a subjective question of a judicial examination is shown in the following table.
The second question bank stores a plurality of second profiles. The second profile comprises question-answer identification, knowledge point code and question-answer scoreLength of question and answer. The question-and-answer time length is a time length expected to be assigned to the corresponding question and answer. When a question-answer relates to multiple knowledge points, the corresponding knowledge point code may contain multiple knowledge pointsAnd (6) recognizing points. The second profile of the test paper is as follows.
Question-answering mark | Question and answer score | Duration of question and answer | Knowledge point coding |
11 | 5 | 5 | 113 |
12 | 5 | 10 | 117 |
21 | 5 | 10 | 256 |
22 | 10 | 10 | 259 |
31 | 5 | 5 | 365 |
32 | 10 | 5 | 371 |
33 | 10 | 10 | 377/241 |
41 | 5 | 5 | 119 |
42 | 10 | 5 | 633 |
43 | 10 | 10 | 671 |
44 | 15 | 15 | 678/114 |
The score database stores score lists of multiple answerers, each score list has multiple question-answer identifiers and question-answer scores uniquely corresponding to the question-answer identifiers. And after the examination paper is read, generating a score list according to the score condition of each question and answer of each simple answer of the answerer. The coaching performance of a student is shown in the table below.
Question-answering mark | Question and answer score |
11 | 5 |
12 | 3 |
21 | 5 |
22 | 8 |
31 | 0 |
32 | 0 |
33 | 0 |
41 | 5 |
42 | 6 |
43 | 3 |
44 | 0 |
Determining a set of answers based on the score listThe answer set includes any question-answer identifiers with a non-zero question-answer score. Or, extracting the scored question-answer identifiers from all question-answer identifiers. In general, the question-answer identification is not zero, so that the answering machine can determine that the answering machine tries to answer the test questions, and the actual score can be regarded as the learning result of the answering machine. Answering setFor anyAll satisfy. The preset test questions are always n, and each test question comprises the number of questions and answers of m.,。As the total number of questions that have been answered,the number of questions and answers that have been answered in the test questions.
According to the answer set, calling the first profile with the question-answer identification in the first question bank to determine the question amount regulating coefficient. Because partial answerers only answer partial test questions, the time actually consumed for the answered test questions exceeds the preset time length, and the answering person does not scoreCan be used as the actual learning result of the test question. The question volume adjusting coefficient is the degree that the answering time of the answerer exceeds the preset total answering time.。
Calling a second profile with the question-answer identifier in a second question bank according to the answer set to determine the test question adjustment coefficient of each test question. For partial respondents, the same test question only answers partial questions. Since the question stem duration is a lead time, the smaller the number of completed questions, the longer the time allocated for the questions is.。
Determining a revised score for each question and answer identifier,. According to question-answer scoreAnd the correction scoreDetermining learning outcome for each knowledge point code,. S =120min was determined from the previous table as in the following table. Length of examination paper basic answer5+10+15+(5+10)+(10+10)+(5+5+10)=85 min. For the test question 10, the basic answer duration is long=5+5+10=20, basic answer time length of all questions and answers=5+10= 15. For the test question 20, the basic answer duration is long=10+10+10=30, basic answer time length of all questions and answers=10+10= 20. For the test question 40, the basic answering duration is long=15+5+5+10=35, basic answer time length of all questions and answers=5+5+10= 20. The following table is specific.
Generally, the test questions which are not answered cannot predict whether the time is short or the test questions which are not answered, so that the system only outputs the learning result of the knowledge point codes corresponding to the question and answer identifications with any question and answer score which is not zero with the cautious purpose. For large-scale examinations of multiple students in the same specialty, the learning results of multiple respondents coding any knowledge points can be predicted, and the average learning results can be determined. The average learning outcome can be regarded as the teaching outcome of the teacher to the knowledge point. It should be noted that since the actual number of questions to be answered by the answerer cannot exceed the total number of questions, the total answering time cannot exceed the total examination time, and therefore。
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (7)
1. A method for predicting learning outcome by correcting short-answer question score is characterized in that,
The first question bank stores a plurality of first profiles, and the first profiles comprise question identifications and question stem durationAnd multiple question-answer identifiers corresponding to the question stem identifier;
The second question bank stores a plurality of second profiles, and the second profiles comprise question and answer identifiers, knowledge point codes and question and answer scoresLength of question and answer;
The score database stores score lists of multiple answerers, each score list has multiple question-answer identifiers and question-answer scores uniquely corresponding to the question-answer identifiers;
Determining a set of answers based on the score listThe answer set includes any questions and answersThe question-answer labels with the scores not equal to zero,for anyAll satisfy,;
According to the answer set, calling the first profile with the question-answer identification in the first question bank to determine the question amount regulating coefficient,;
Calling a second profile with the question-answer identifier in a second question bank according to the answer set to determine the test question adjustment coefficient of each test question,;
2. The method of claim 1, wherein each knowledge point code comprises one or more knowledge points.
3. The method for predicting learning outcome by modifying the simple answer score as claimed in claim 1, wherein the learning outcome of the knowledge point code corresponding to the question mark with any score not equal to zero is outputted.
5. The method of claim 1, wherein the learning outcome of any knowledge point coding of a plurality of respondents is predicted, an average learning outcome is obtained, and the teaching quality is determined.
7. the method of claim 1, wherein each test question has more than one question and answer, and each test question mark corresponds to multiple question and answer marks.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010653451.0A CN111539588A (en) | 2020-07-08 | 2020-07-08 | Method for predicting learning result by correcting short answer question score |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010653451.0A CN111539588A (en) | 2020-07-08 | 2020-07-08 | Method for predicting learning result by correcting short answer question score |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111539588A true CN111539588A (en) | 2020-08-14 |
Family
ID=71979787
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010653451.0A Pending CN111539588A (en) | 2020-07-08 | 2020-07-08 | Method for predicting learning result by correcting short answer question score |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111539588A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112785039A (en) * | 2020-12-31 | 2021-05-11 | 科大讯飞股份有限公司 | Test question answering score prediction method and related device |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005257483A (en) * | 2004-03-11 | 2005-09-22 | Sumitomo Pharmaceut Co Ltd | Differential method of analyzing proteome |
CN108763588A (en) * | 2018-06-19 | 2018-11-06 | 深圳习习网络科技有限公司 | A kind of knowledge point quantitative analysis method and device |
CN109242736A (en) * | 2018-09-27 | 2019-01-18 | 广东小天才科技有限公司 | A kind of method and system for the study situation for assisting teacher to understand student |
CN109859553A (en) * | 2018-12-04 | 2019-06-07 | 广东宜教通教育有限公司 | A kind of ability evaluating method doing topic situation based on student |
-
2020
- 2020-07-08 CN CN202010653451.0A patent/CN111539588A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2005257483A (en) * | 2004-03-11 | 2005-09-22 | Sumitomo Pharmaceut Co Ltd | Differential method of analyzing proteome |
CN108763588A (en) * | 2018-06-19 | 2018-11-06 | 深圳习习网络科技有限公司 | A kind of knowledge point quantitative analysis method and device |
CN109242736A (en) * | 2018-09-27 | 2019-01-18 | 广东小天才科技有限公司 | A kind of method and system for the study situation for assisting teacher to understand student |
CN109859553A (en) * | 2018-12-04 | 2019-06-07 | 广东宜教通教育有限公司 | A kind of ability evaluating method doing topic situation based on student |
Non-Patent Citations (1)
Title |
---|
胡春安等: "智能仿真上机指导系统ICAI的研究", 《江西理工大学学报》 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112785039A (en) * | 2020-12-31 | 2021-05-11 | 科大讯飞股份有限公司 | Test question answering score prediction method and related device |
CN112785039B (en) * | 2020-12-31 | 2023-12-01 | 科大讯飞股份有限公司 | Prediction method and related device for answer score rate of test questions |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111368182A (en) | Individualized self-adaptive learning recommendation method based on big data analysis of education platform | |
JP2008547038A5 (en) | ||
US20110045452A1 (en) | Computer-Implemented Systems and Methods for Generating an Adaptive Test | |
CN108734379B (en) | Online training method for realizing differentiation of customer service personnel | |
JP4993773B2 (en) | Development of teaching materials based on multiple choice questions | |
CN110413728B (en) | Method, device, equipment and storage medium for recommending exercise problems | |
CN110929020A (en) | Knowledge point mastery degree analysis method based on test achievement | |
CN111507584A (en) | Data analysis system and method | |
US20090291425A1 (en) | Educational tool | |
CN111539588A (en) | Method for predicting learning result by correcting short answer question score | |
Blazar et al. | Professional Staff Diversity and Student Outcomes: Extending Our Understanding of Race/Ethnicity-Matching Effects in Education. EdWorkingPaper No. 21-500. | |
CN114202978A (en) | Intelligent training method and system based on big data | |
CN111104455B (en) | Multi-source multi-dimensional school teaching transverse information difference comparison and analysis method | |
CN110910043A (en) | Analysis method of test paper information | |
CN116542829A (en) | Personalized intelligent group scroll and on-line examination method | |
CN111369140A (en) | Teaching evaluation system and method | |
CN110378814A (en) | The intelligent tutoring ecosystem | |
CN112015783B (en) | Interactive learning process generation method and system | |
CN112015830B (en) | Question storage method suitable for adaptive learning | |
CN114298872A (en) | Learning situation information processing method | |
CN112163975A (en) | Intelligent learning guiding and prompting method and system | |
CN109858769B (en) | Gramineae graduation design guidance quality evaluation method | |
CN111881284A (en) | Auxiliary verification method for subjective questions in K12 teaching | |
CN112164262A (en) | Intelligent paper reading tutoring system | |
CN111243364A (en) | HanLP-based method for automatically scoring examination subjective questions of registered electrical engineers |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200814 |
|
RJ01 | Rejection of invention patent application after publication |