Knowledge point determination method and system based on wrong test questions

By collecting and analyzing wrong questions in offline paper exams by mobile terminals, weighted sorting is performed based on semantic features and answering features, personalized recommendation problems for determining knowledge points in offline exams are solved, and the efficiency of checking for deficiencies and filling in gaps is improved.

CN119067209AActive Publication Date: 2024-12-03BEIJING FENGHUANG XUE YI SCI & TECH CO LTD
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202411195600.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-12-03
Estimated Expiration
2044-08-29

AI Technical Summary

Technical Problem

It is difficult for the existing technology to accurately determine the knowledge points to be improved for individual testers in offline paper examination scenarios, and most wrong question management plans cannot be applied to common paper examinations, and personalized recommendations are missing.

Method used

Images are collected on the target test area of ​​the test paper through M mobile terminals, semantic features and answering features of each group of wrong test questions are extracted, knowledge points are given weight points based on similarity, and weight points are divided and pre-set based on the allocation weights of each wrong test question. The benchmark scores determine multiple target knowledge points, and weighted scores are sorted in consideration of individual and overall situations.

Benefits of technology

It realizes accurate knowledge points recommendations for each individual, improves the individual and overall efficiency of checking and filling gaps, and is suitable for more common offline examination scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067209A_ABST
    Figure CN119067209A_ABST
Patent Text Reader

Abstract

The invention provides a knowledge point determination method and system based on wrong test questions, and belongs to the technical field of data processing and educational informatization. The method comprises the steps of obtaining M groups of wrong test question pictures; extracting a plurality of semantic features and / or answer features of each group of wrong test question pictures; assigning a knowledge point weight score to each group of wrong test question pictures; the knowledge point weight is distributed to each wrong test question corresponding to the group; and determining a plurality of target knowledge points. The system comprises an error test question collection module, a feature extraction module, a similarity determination module, a score determination module and a target knowledge point determination module. According to the technical scheme, when the target knowledge point is determined based on the wrong test question, the condition of each respondent and the overall condition of all respondents are considered at the same time, and weighted assignment sorting is executed based on the individual condition and the overall condition, so that the target knowledge point for each individual is obtained and pushed, and the pushing efficiency is improved. And the individual and overall defect checking and repairing efficiency can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of data processing and educational information technology, and in particular relates to a method and system for determining knowledge points based on incorrect test questions, a computer-readable storage medium for implementing the method, a computer program product, and an electronic device. Background Art

[0002] Wrong questions are a kind of learning resource, and they are tailor-made for learners. Wrong question management is an important way to learn knowledge. It can not only help teachers improve the quality of teaching, but also help students to conduct personalized learning, play a role in learning from one example and get rid of the tactics of the sea of ​​questions. It is a measure that can help students improve their learning efficiency and develop scientific learning methods. Consolidation exercises for wrong questions are the most important part of the wrong question management system. Its main purpose is to recommend a certain number of exercises of relevant knowledge to learners according to their needs, so as to practice and consolidate their knowledge.

[0003] Chinese invention patent CN117648934B has proposed a method, device, equipment and medium for determining knowledge points based on incorrect test questions, which accurately determines incorrect test questions caused by errors from the test question keyword dimension and the test question option dimension, thereby achieving the purpose of determining the knowledge points corresponding to incorrect test questions that are not caused by errors as knowledge points to be improved, and solving the technical problem of poor accuracy in determining knowledge points to be improved in related technologies.

[0004] In actual applications, it is found that most of the error management solutions in the existing technology are aimed at online (electronic) examination systems or online (electronic) exercise systems, but the above solutions cannot be directly applied to the more common offline (paper) examination or exercise scenarios; in addition, the existing error management solutions focus on collecting existing errors as a whole for analysis, and the recommended knowledge points are also for the overall test takers, and do not consider the different situations of individual test takers to give targeted recommendations.

[0005] Therefore, according to the above practical problems and based on the existing technologies, the present application proposes a method and system for determining knowledge points based on incorrect questions that can be applicable to more general offline (paper-based) examinations or test scenarios, while taking into account the situation of each respondent and the overall situation of all respondent, and performing weighted scoring and sorting based on individual and overall situations, as well as a computer-readable storage medium, a computer program product and an electronic device for implementing the method. Summary of the invention

[0006] In order to solve the above technical problems, the present invention proposes a method and system for determining knowledge points based on wrong questions that can be applied to more general offline (paper) examinations or practice questions, taking into account the individual situation of each answerer and the overall situation of all answerers, and performing weighted scoring and sorting based on individual and overall situations, as well as a computer-readable storage medium, a computer program product, and an electronic device for implementing the method as follows: In a first aspect of the present invention, a method for determining knowledge points based on incorrect test questions is provided, wherein the method is executed based on a mobile terminal; The method comprises the following steps: S100: Through M mobile terminals to the target test area of ​​the test paper to collect images, to obtain M groups of wrong test pictures, the target test paper for the wrong answer test; M> 1; S200: extracting multiple semantic features and / or answer features of each group of wrong test picture; S300: assigning a knowledge point weight score to each group of wrong test question pictures according to the similarity of the wrong test question pictures in different groups; S400: Allocating the knowledge point weight corresponding to each group of wrong test question images to each wrong test question corresponding to the group; S500: Determine a plurality of target knowledge points based on the assigned weight score of each incorrect test question, the number of each group of incorrect test question pictures, and a pre-set benchmark score.

[0007] The step S100 specifically includes: Through M mobile terminals Capturing images of the target test question area of ​​the test paper; Among them, mobile terminals The test papers include Wrong answer on the test question, mobile terminal get Wrong test picture ; ; Count the number of wrong test pictures obtained by each mobile terminal among M mobile terminals ; Calculate the maximum number of wrong test pictures obtained by all mobile terminals ; For mobile terminals Obtained Wrong test picture , like , then the mobile terminal Obtained Wrong test picture Expand to The wrong test pictures are taken as the i-th group of wrong test pictures .

[0008] The step S200 extracts multiple semantic features and / or answer features of each group of wrong test question images, specifically including: The semantic feature is a semantic segmentation feature of the incorrectly answered test question stem; The answer feature is an answer identification feature corresponding to the test question with an incorrect answer.

[0009] The step S300 assigns knowledge point weights to each group of wrong test question pictures according to the similarities of the wrong test question pictures of different groups, specifically including: Calculate the i-th group of wrong test questions and the jth group of wrong test pictures The semantic feature similarity of ; ; Based on semantic feature similarity Determine the i-th group of wrong test questions and the jth group of wrong test pictures Knowledge point weight score; similarity between the knowledge point weight score and the semantic feature Positive correlation.

[0010] The step S400 allocates the knowledge point weights corresponding to each group of wrong test question images to each wrong test question corresponding to the group, specifically including: Get the i-th group of wrong test questions The answer identification features for each incorrect test question included; determining an actual score for each incorrect test question based on the answer identification feature; Determine the knowledge point weight score actually assigned to each wrong question based on the actual score of the wrong question; Among them, the knowledge point weight score actually assigned to each wrong test question is inversely correlated with the actual score of each wrong test question.

[0011] The step S500 of determining a plurality of target knowledge points specifically includes: Based on the assigned weight score of each wrong test question, the number of each group of wrong test question pictures and a preset benchmark score, M groups of target knowledge points are determined, and the M groups of target knowledge points correspond to the M mobile terminals; After step S500, the method further includes: S600: Pushing the M groups of knowledge points to the corresponding M mobile terminals.

[0012] In order to execute the method described in the first aspect, in the second aspect of the present invention, a knowledge point determination system based on incorrect test questions is proposed, and the system includes an incorrect test question collection module, a feature extraction module, a similarity determination module, a score determination module and a target knowledge point determination module.

[0013] The wrong test question collection module includes M mobile terminals, through the M mobile terminals to the target test question area of ​​the test paper to collect images, to obtain M groups of wrong test question pictures, the target test question is the test question in the test paper with the wrong answer; M> 1; The feature extraction module extracts multiple semantic features and / or answer features of each group of wrong test question images; The score determination module includes a group weight determination unit and a test question allocation determination unit; The group weight determination unit assigns a knowledge point weight to each group of wrong test question pictures according to the similarities of the wrong test question pictures of different groups determined by the similarity determination module; The test question allocation score determination unit allocates the knowledge point weight score corresponding to each group of wrong test question pictures to each wrong test question corresponding to the group; The target knowledge point determination module determines a plurality of target knowledge points based on the assigned weight score of each wrong test question, the number of wrong test question pictures in each group, and a preset benchmark score; Wherein, the test question allocation score determination unit determines the knowledge point weight score actually allocated to each wrong test question based on the actual score of the wrong test question; Among them, the knowledge point weight score actually assigned to each wrong test question is inversely correlated with the actual score of each wrong test question.

[0014] The semantic feature is a semantic segmentation feature of the incorrectly answered test question stem; The answer feature is an answer recognition feature corresponding to a test question with an incorrect answer; The grouping weight determination unit assigns a knowledge point weight to each group of wrong test question pictures according to the similarities of different groups of wrong test question pictures determined by the similarity determination module, specifically including: The similarity determination module calculates the semantic feature similarity of the semantic segmentation of the wrong test pictures of different groups; The grouping weight determination unit determines the knowledge point weights of the wrong test pictures of different groups based on the similarity of semantic features; The knowledge point weight score is positively correlated with the semantic feature similarity.

[0015] The target knowledge point determination module determines M groups of target knowledge points based on the assigned weight score of each wrong test question, the number of wrong test question pictures in each group, and a preset benchmark score. The M groups of target knowledge points correspond to the M mobile terminals; The target knowledge point determination module also includes a knowledge point pushing unit; The knowledge point pushing unit pushes the M groups of knowledge points to the corresponding M mobile terminals.

[0016] Part or all of the steps of the method for determining knowledge points based on incorrect test questions described in the first aspect can be automatically implemented through various forms of electronic devices and computer program instructions; the computer program instructions can be stored in storage media of different forms and loaded into computer electronic devices for execution.

[0017] Therefore, in the third aspect of the present invention, a computer-readable storage medium is also provided for storing computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes a method for determining knowledge points based on incorrect test questions as described in the first aspect.

[0018] In the fourth aspect of the present invention, an electronic device is also proposed, which includes a processor and a memory, the memory is used to store instructions, and the processor is used to call the instructions in the memory, so that the electronic device executes the knowledge point determination method based on incorrect test questions described in the first aspect above.

[0019] In a fifth aspect of the present invention, a computer program product is also proposed, the product comprising a computer program, and when the computer program is executed, the method for determining knowledge points based on incorrect test questions described in the first aspect above is performed.

[0020] When the technical solution of the present application determines the target knowledge points based on incorrect test questions, it takes into account the situation of each answerer and the overall situation of all answerers, and performs weighted scoring and sorting based on the individual situation and the overall situation, so as to complete the push of target knowledge points for each individual, which can improve the efficiency of checking for deficiencies and filling gaps for both individuals and the whole.

[0021] Further advantages of the present invention will be further reflected in detail in the specific embodiments section in conjunction with the drawings of the specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0023] Figure 1This is a schematic diagram of the main flow of a method for determining knowledge points based on incorrect test questions according to an embodiment of the present invention. Figure 2 yes Figure 1 Schematic diagram of the principle of expanding the wrong test picture obtained by each mobile terminal in the method Figure 3 Schematic diagram of the main flow of the method for determining knowledge points based on wrong test questions according to a more preferred embodiment of the present invention Figure 4 This is a schematic diagram of the functional modules of a knowledge point determination system based on incorrect test questions according to an embodiment of the present invention. Figure 5 This is a schematic diagram of a preferred embodiment of a knowledge point determination system based on wrong test questions of the present invention. DETAILED DESCRIPTION

[0024] Here, exemplary embodiments will be described in detail, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments that are the same as the present application. Instead, they are only examples of devices and methods that are the same as some aspects of the present application as detailed in the attached claims.

[0025] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0026] At the same time, in the specific implementation of this application, if user-related data is involved, when the embodiment of this application is applied to a specific product or technology, the user's permission or consent must be obtained, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0027] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0028] See also Figure 1 , Figure 1A schematic diagram showing the main flow of a method for determining knowledge points based on incorrect test questions according to an embodiment of the present invention is shown. Figure 1 The method is executed based on multiple mobile terminals and specifically includes: Figure 1 The steps S100-S500 are specifically implemented as follows: S100: Through M mobile terminals to the target test area of ​​the test paper to collect images, to obtain M groups of wrong test pictures, the target test paper for the wrong answer test; M> 1; S200: extracting multiple semantic features and / or answer features of each group of wrong test picture; S300: assigning a knowledge point weight score to each group of wrong test question pictures according to the similarity of the wrong test question pictures in different groups; S400: Allocating the knowledge point weight corresponding to each group of wrong test question images to each wrong test question corresponding to the group; S500: Determine a plurality of target knowledge points based on the assigned weight score of each incorrect test question, the number of each group of incorrect test question pictures, and a pre-set benchmark score.

[0029] Next, the above steps are further explained in combination with a specific scenario.

[0030] After an exam or practice assignment, the instructor (also known as a teacher, administrator, etc.) marks all the test questions (also known as test papers, practice questions, etc.) and returns the marked test questions to the test takers (also known as students, practitioners, examinees, etc.). After each test taker receives the returned test paper, the test paper will have the score of each question, whether the answer to each question is wrong (marked by the instructor), and the original answer (submitted by the test taker).

[0031] Therefore, the scenario targeted by this embodiment can be applicable to more general offline (paper-based) examinations or test question scenarios.

[0032] On this basis, step S100 is executed to collect images of the target test question area of ​​the test paper through M mobile terminals to obtain M groups of wrong test question pictures, where the target test question is a test question with an incorrect answer in the test paper; M>1; Specifically, after the M test subjects receive the returned test papers, they use their own or their parents' mobile terminals to collect images of the target test question area of ​​their test papers to obtain M groups of wrong test question images; Although each examinee has the same test paper, the test questions that each examinee gets wrong are generally not exactly the same. Therefore, step S100 allows each examinee to take a photo of each wrong question by himself, which is more in line with the actual situation.

[0033] For the convenience of description, assume that M mobile terminals are identified as , corresponding to M test subjects, and each test subject answered at least one question incorrectly (not considering the full score).

[0034] At this time, the mobile terminal The test papers targeted include If the test paper contains Wrong answer on the test question, mobile terminal Take a separate photo of each test area to obtain Pictures of incorrect test questions; Preferably, the mobile terminal Will The wrong test pictures are sorted by priority to form a picture queue set ; ; Preferably, the priority level is determined when the subject is taking the photo, that is, the subject first takes a photo of the error area that he thinks is most important to obtain the image. , and then take a picture of the less important error area , and so on, we finally get Pictures of incorrect test questions; It should be noted that the priority determined by this shooting method is determined by the photographer based on his or her own actual situation, which is more in line with actual teaching scenarios, because each test subject has a different understanding of the importance of different knowledge points and a different degree of mastery. That is, this application fully considers the actual situation of individual test subjects, which is one of the initial improvement bases for proposing the technical solution of this application.

[0035] More importantly, the image queue set formed by the above sorting method , which determines the subsequent processing of the image queue set The method of element expansion is more targeted, which will be further emphasized in the following steps, and is one of the improvements of this application.

[0036] Figure 2 yes Figure 1 A schematic diagram of the principle of expanding the wrong test question images obtained by each mobile terminal in the method.

[0037] First, count the number of wrong test pictures obtained by each mobile terminal among the M mobile terminals. ; Continuing from the above description, it can be seen that the mobile terminal Will The wrong test pictures are sorted by priority to form a picture queue set , that is, mobile terminals The number of wrong test pictures obtained is ; Next, calculate the maximum number of wrong test pictures obtained by all mobile terminals ; In order to highlight the importance of the highest priority wrong questions among the wrong question images obtained by each mobile terminal, this embodiment is further improved as follows: like , then the mobile terminal Obtained Wrong test picture Expand to The wrong test pictures are taken as the i-th group of wrong test pictures .

[0038] Specifically, the expansion method is: The elements in are copied one by one from the beginning to and keep looping until Included elements.

[0039] Right now , ; Because each expansion starts from the beginning (from the element with the highest priority).

[0040] According to statistical principles, the scores of most test subjects are distributed in the same range, that is, the number of incorrect test pictures obtained by most mobile terminals will not differ too much.

[0041] Therefore, the above cyclic copy process will generally not exceed 2 or 3 times. No matter how many times the copy process is repeated, as long as there is repetition, it will be copied to the element at the beginning (the highest priority).

[0042] After the above expansion operation, each mobile terminal Each corresponds to an expanded set of wrong test picture queues , and each incorrect trial picture queue set contains the same elements (all containing elements).

[0043] Queue all wrong test pictures By arranging the elements of The matrix KM is as follows: To better describe the above process, Figure 2 A simple specific parameterization process is shown.

[0044] exist Figure 2In the example, M=3 is used (i.e., assuming there are three test subjects A, B, and C). Test subject A submitted three wrong test pictures A1, A2, and A3 (arranged according to the importance of test subject A); test subject B submitted two wrong test pictures B1 and B2 (arranged according to the importance of test subject B); test subject C submitted two wrong test pictures C1 and C2 (arranged according to the importance of test subject C); at this time, ; Therefore, it is necessary to expand {B1, B2} and {C1, C2} to obtain: (No expansion required) As a further explanation, Assumptions ,but at this time This process is the aforementioned The elements in are copied one by one from the beginning to and keep looping until Included elements.

[0045] In the first case (M=3, ) as an example, the matrix KM is as follows: .

[0046] On the basis of obtaining the matrix KM, the subsequent process can be continued as follows: S200: extracting multiple semantic features and / or answer features of each group of wrong test picture; Specifically, the semantic feature is a semantic segmentation feature of the incorrectly answered test question stem; The answer feature is an answer identification feature corresponding to the test question with an incorrect answer.

[0047] It can be understood that the semantic word segmentation feature is necessary for the wrong test question images, but the answer recognition feature corresponding to the wrongly answered test questions is optional, because there may be a situation where a question is not answered (that is, the answer recognition feature is blank).

[0048] When the answer identification feature does not exist, the score of the question can be determined as 0 based on the answer identification feature.

[0049] This application considers answer recognition features mainly to correctly distinguish the relative scores of each question. In actual scenarios, teachers may not give scores for each question. Teachers may only count the correct answers, so there is no need to score wrong questions. However, the subsequent steps of this application involve the need to use the actual scoring of each (wrong) question. Therefore, the answer feature is introduced here, which also reflects the completeness of the technical solution of this application.

[0050] It can be understood that the semantic word segmentation features for identifying test questions and the answer identification features corresponding to incorrectly answered questions belong to the existing technology. This is not the focus of the technical solution of this application. This application will not elaborate on this. For details, please refer to the relevant existing technology (such as the relevant literature mentioned in the background technology).

[0051] On this basis, step S300 assigns knowledge point weights to each group of wrong test pictures according to the similarities of wrong test pictures of different groups, specifically including: Calculate the i-th group of wrong test questions and the jth group of wrong test pictures The semantic feature similarity of ; ; Based on semantic feature similarity Determine the i-th group of wrong test questions and the jth group of wrong test pictures Knowledge point weight score; similarity between the knowledge point weight score and the semantic feature Positive correlation.

[0052] It is understandable that the i-th group of wrong test pictures and the jth group of wrong test pictures The semantic feature similarity of , and the jth group of wrong test pictures and the i-th group of wrong test pictures The semantic feature similarity of are the same, that is ; Then, the i-th group of wrong test pictures and the jth group of wrong test pictures The weights of the knowledge points are also the same, which is the overall situation of all the respondents considered in this application.

[0053] Semantic feature similarity Generally, it is between (0,1]. For the sake of simplicity, the semantic feature similarity can be directly The value of is used as the knowledge point weight score. Of course, other scoring methods (such as ten points, one hundred points, etc.) can also be used, as long as the knowledge point weight score is ensured to be similar to the semantic feature. A positive correlation will do.

[0054] Next, the step S400 allocates the knowledge point weights corresponding to each group of wrong test question images to each wrong test question corresponding to the group, specifically including: Get the i-th group of wrong test questions The answer identification features for each incorrect test question included; determining an actual score for each incorrect test question based on the answer identification feature; Determine the knowledge point weight score actually assigned to each wrong question based on the actual score of the wrong question; Among them, the knowledge point weight score actually assigned to each wrong test question is inversely correlated with the actual score of each wrong test question.

[0055] The negative correlation here means that the higher the actual score of each wrong question (an wrong question refers to a question that has errors and does not get full marks, but it is possible to get partial marks), the lower the knowledge point weight score actually assigned to each wrong question will be. This is because the lower the score, the lower the degree of mastery of the knowledge point, and the higher the importance of the knowledge point, which is also in line with the actual situation of the individual.

[0056] For example, when a test question scores 0 points, the knowledge point weight score actually assigned to the wrong test question may be directly equal to the knowledge point weight score of the group to which it belongs (the highest value).

[0057] After the knowledge point weights assigned to each wrong test picture are obtained, the aforementioned matrix KM can be transformed into the knowledge point weight matrix SM as follows: , in, For wrong test picture The weight of the knowledge point assigned.

[0058] It can be understood that identifying the semantic word segmentation features of the test question stem, the answer recognition features corresponding to the wrongly answered questions, the similarity comparison calculation between the features, etc. all belong to the existing technology, which is not the focus of the technical solution of this application. This application will not expand on this. For details, please refer to the relevant existing technology (such as the relevant literature mentioned in the background technology).

[0059] Next, step S400 is executed to determine a plurality of target knowledge points based on the assigned weight score of each incorrect test question, the number of each group of incorrect test question pictures, and a pre-set benchmark score.

[0060] The step S500 of determining a plurality of target knowledge points specifically includes: Based on the assigned weight score of each incorrect test question, the number of each group of incorrect test question pictures and a pre-set benchmark score, M groups of target knowledge points are determined, and the M groups of target knowledge points correspond to the M mobile terminals.

[0061] Preferably, all knowledge points corresponding to the wrong questions whose assigned weights are greater than a preset benchmark numerator and whose number of wrong question images exceeds the overall average are taken as recommended knowledge points, and the target knowledge points are determined based on the recommended knowledge points.

[0062] Specifically, determining the target knowledge point based on the recommended knowledge point can be to determine the same or similar test questions or exercises based on the recommended knowledge point as the target knowledge point. This can be achieved through the related technologies based on text recommendation and data mining in the prior art, which will not be elaborated in this application. For example, the following prior art can be referred to:

[0063] [1] Wang Wenquan. Design and implementation of personalized recommendation exercise algorithm in wrong question management system [J]. China Education Informatization, 2016(11):67-70+92. [2] Guo Hongxia. Research and development of data mining system for middle school students’ wrong questions[D]. North China University of Technology, 2012. Optionally, to facilitate computer program implementation, the matrix SM may be partially deleted of duplicate elements, and then subjected to principal component analysis and dimension reduction to an X-dimensional matrix, where X <min{M, }.

[0064] Based on the X-dimensional matrix, the weight points assigned to each wrong test question and the number of wrong test question pictures in each group are obtained.

[0065] On this basis, Figure 3 It is a schematic diagram of the main flow of a method for determining knowledge points based on incorrect test questions in a more preferred embodiment of the present invention.

[0066] Combination Figure 3 As well as the aforementioned embodiments, a method for determining knowledge points based on incorrect test questions in a more preferred embodiment of the present invention is implemented as follows: A method for determining knowledge points based on incorrect test questions, the method being executed based on a mobile terminal, characterized in that the method comprises the following steps: S100: Through M mobile terminals to the target test area of ​​the test paper to collect images, to obtain M groups of wrong test pictures, the target test paper for the wrong answer test; M> 1; S200: extracting multiple semantic features and / or answer features of each group of wrong test picture; S300: assigning a knowledge point weight score to each group of wrong test question pictures according to the similarity of the wrong test question pictures in different groups; S400: Allocating the knowledge point weight corresponding to each group of wrong test question images to each wrong test question corresponding to the group; S500: Based on the assigned weight score of each incorrect test question, the number of each group of incorrect test question pictures and a pre-set benchmark score, M groups of target knowledge points are determined, and the M groups of target knowledge points correspond to the M mobile terminals.

[0067] S600: Pushing the M groups of knowledge points to the corresponding M mobile terminals.

[0068] Based on the method embodiment, Figure 4 The present invention is a schematic diagram of the functional modules of a knowledge point determination system based on incorrect test questions according to an embodiment of the present invention.

[0069] The system includes an incorrect test question collection module, a feature extraction module, a similarity determination module, a score determination module and a target knowledge point determination module.

[0070] Figure 5 This is a schematic diagram of a preferred embodiment of a knowledge point determination system based on wrong test questions of the present invention. Among them, it is shown that the score determination module includes a grouping weight score determination unit and a test question allocation score determination unit.

[0071] The wrong test question collection module includes M mobile terminals, through the M mobile terminals to the target test question area of ​​the test paper to collect images, to obtain M groups of wrong test question pictures, the target test question is the test question in the test paper with the wrong answer; M> 1; The feature extraction module extracts multiple semantic features and / or answer features of each group of wrong test question images; The semantic features are obtained by the semantic feature extraction unit, and the answer features are obtained by the answer feature extraction unit; The score determination module includes a group weight determination unit and a test question allocation determination unit; The group weight determination unit assigns a knowledge point weight to each group of wrong test question pictures according to the similarities of the wrong test question pictures of different groups determined by the similarity determination module; The test question allocation score determination unit allocates the knowledge point weight score corresponding to each group of wrong test question pictures to each wrong test question corresponding to the group; The target knowledge point determination module determines a plurality of target knowledge points based on the assigned weight score of each wrong test question, the number of wrong test question pictures in each group, and a preset benchmark score; Wherein, the test question allocation score determination unit determines the knowledge point weight score actually allocated to each wrong test question based on the actual score of the wrong test question; Among them, the knowledge point weight score actually assigned to each wrong test question is inversely correlated with the actual score of each wrong test question.

[0072] The semantic feature is a semantic segmentation feature of the incorrectly answered test question stem; The answer feature is an answer recognition feature corresponding to a test question with an incorrect answer; The grouping weight determination unit assigns a knowledge point weight to each group of wrong test question pictures according to the similarities of different groups of wrong test question pictures determined by the similarity determination module, specifically including: The similarity determination module calculates the semantic feature similarity of the semantic segmentation of the wrong test pictures of different groups; The grouping weight determination unit determines the knowledge point weights of the wrong test pictures of different groups based on the similarity of semantic features; The knowledge point weight score is positively correlated with the semantic feature similarity.

[0073] The target knowledge point determination module determines M groups of target knowledge points based on the assigned weight score of each wrong test question, the number of wrong test question pictures in each group, and a preset benchmark score. The M groups of target knowledge points correspond to the M mobile terminals; The target knowledge point determination module also includes a knowledge point pushing unit; The knowledge point pushing unit pushes the M groups of knowledge points to the corresponding M mobile terminals.

[0074] It is understandable that after the embodiment in the form of method flow is elaborated in detail, the corresponding embodiment in the form of system function architecture does not need to be repeated again, but those skilled in the art can directly and unambiguously determine that the embodiment in the form of system function architecture also corresponds to all the contents of the embodiment in the form of method flow. The same is true for the embodiment in the form of computer program code pseudo language.

[0075] The present invention provides multiple embodiments, each of which can constitute an independent technical solution and may contribute to the prior art and solve corresponding technical problems. However, it should be pointed out that different embodiments can be combined with each other without violating logic; at the same time, each embodiment can solve at least one technical problem, but it is not required that each individual embodiment solves multiple or all technical problems.

[0076] For other technologies, principles, algorithms or models not elaborated in detail in this application, please refer to the prior art.

[0077] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0078] It should be noted that, for the knowledge point determination method based on wrong questions in the embodiment of the present application, a person of ordinary skill in the art can understand that all or part of the process of implementing the knowledge point determination method based on wrong questions in the embodiment of the present application can be completed by controlling the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, such as a memory of an electronic device, and executed by at least one processor in the electronic device, and the execution process may include the process of the embodiment of the knowledge point determination method based on wrong questions. Among them, the storage medium can be a disk, an optical disk, a read-only memory, a random access memory, etc.

[0079] In summary, according to the actual problems encountered in educational scenarios, on the basis of existing technologies, this application proposes a method and system for determining knowledge points based on wrong questions, which can be applied to more general offline (paper) examinations or test scenarios, taking into account the situation of each answerer and the overall situation of all answerers, and performing weighted scoring and sorting based on individual and overall situations, and a computer-readable storage medium, computer program product, and electronic device for implementing the method. The technical solution of the present application determines the target knowledge points based on wrong questions, taking into account the situation of each answerer and the overall situation of all answerers, and performing weighted scoring and sorting based on individual and overall situations, so as to complete the push of target knowledge points for each individual, which can improve the efficiency of checking for deficiencies and filling gaps for individuals and the whole.

[0080] The foregoing has shown and described the method embodiments and system of the present invention, but it is understandable to those skilled in the art that various changes, modifications, substitutions and variations may be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for determining knowledge points based on incorrect test questions, the method being executed on a mobile terminal, characterized in that: The method comprises the following steps: S100: Through M mobile terminals to the target test area of ​​the test paper to collect images, to obtain M groups of wrong test pictures, the target test paper for the wrong answer test; M> 1; S200: extracting multiple semantic features and / or answer features of each group of wrong test picture; S300: assigning a knowledge point weight score to each group of wrong test question pictures according to the similarity of the wrong test question pictures in different groups; S400: Allocating the knowledge point weight corresponding to each group of wrong test question images to each wrong test question corresponding to the group; S500: Determine a plurality of target knowledge points based on the assigned weight score of each incorrect test question, the number of each group of incorrect test question pictures, and a pre-set benchmark score.

2. A method for determining knowledge points based on incorrect test questions as claimed in claim 1, characterized in that: The step S500 of determining a plurality of target knowledge points specifically includes: Based on the assigned weight score of each wrong test question, the number of each group of wrong test question pictures and a preset benchmark score, M groups of target knowledge points are determined, and the M groups of target knowledge points correspond to the M mobile terminals; After step S500, the method further includes: S600: Pushing the M groups of target knowledge points to the corresponding M mobile terminals.

3. A method for determining knowledge points based on incorrect test questions as claimed in claim 1, characterized in that: The step S100 specifically includes: Through M mobile terminals Capturing images of the target test question area of ​​the test paper; Among them, mobile terminals The test papers include Wrong answer on the test question, mobile terminal get Wrong test picture ; ; Count the number of wrong test pictures obtained by each mobile terminal among M mobile terminals ; Calculate the maximum number of wrong test pictures obtained by all mobile terminals ; For mobile terminals Obtained Wrong test picture , like , then the mobile terminal Obtained Wrong test picture Expand to The wrong test pictures are taken as the i-th group of wrong test pictures .

4. A method for determining knowledge points based on incorrect test questions as claimed in claim 1, characterized in that: The step S200 extracts multiple semantic features and / or answer features of each group of wrong test question images, specifically including: The semantic feature is a semantic segmentation feature of the incorrectly answered test question stem; The answer feature is an answer identification feature corresponding to the test question with an incorrect answer.

5. A method for determining knowledge points based on incorrect test questions as claimed in claim 3, characterized in that: The step S300 assigns knowledge point weights to each group of wrong test question pictures according to the similarities of the wrong test question pictures of different groups, specifically including: Calculate the i-th group of wrong test questions and the jth group of wrong test pictures The semantic feature similarity of ; ; Based on semantic feature similarity Determine the i-th group of wrong test questions and the jth group of wrong test pictures Knowledge point weight score; similarity between the knowledge point weight score and the semantic feature Positive correlation.

6. A method for determining knowledge points based on incorrect test questions as claimed in claim 3, characterized in that: The step S400 allocates the knowledge point weights corresponding to each group of wrong test question images to each wrong test question corresponding to the group, specifically including: Get the i-th group of wrong test questions The answer identification features for each incorrect test question included; determining an actual score for each incorrect test question based on the answer identification feature; Determine the knowledge point weight score actually assigned to each wrong question based on the actual score of the wrong question; Among them, the knowledge point weight score actually assigned to each wrong test question is inversely correlated with the actual score of each wrong test question.

7. A knowledge point determination system based on incorrect test questions, the system comprising an incorrect test question collection module, a feature extraction module, a similarity determination module, a score determination module and a target knowledge point determination module; characterized in that: The wrong test question collection module includes M mobile terminals, through the M mobile terminals to the target test question area of ​​the test paper to collect images, to obtain M groups of wrong test question pictures, the target test question is the test question in the test paper with the wrong answer; M> 1; The feature extraction module extracts multiple semantic features and / or answer features of each group of wrong test question images; The score determination module includes a group weight determination unit and a test question allocation determination unit; The group weight determination unit assigns a knowledge point weight to each group of wrong test question pictures according to the similarities of the wrong test question pictures of different groups determined by the similarity determination module; The test question allocation score determination unit allocates the knowledge point weight score corresponding to each group of wrong test question pictures to each wrong test question corresponding to the group; The target knowledge point determination module determines a plurality of target knowledge points based on the assigned weight score of each wrong test question, the number of wrong test question pictures in each group, and a preset benchmark score; Wherein, the test question allocation score determination unit determines the knowledge point weight score actually allocated to each wrong test question based on the actual score of the wrong test question; Among them, the knowledge point weight score actually assigned to each wrong test question is inversely correlated with the actual score of each wrong test question.

8. A knowledge point determination system based on incorrect test questions as claimed in claim 7, characterized in that: The semantic feature is a semantic segmentation feature of the incorrectly answered test question stem; The answer feature is an answer recognition feature corresponding to a test question with an incorrect answer; The grouping weight determination unit assigns a knowledge point weight to each group of wrong test question pictures according to the similarities of different groups of wrong test question pictures determined by the similarity determination module, specifically including: The similarity determination module calculates the semantic feature similarity of the semantic segmentation of the wrong test pictures of different groups; The grouping weight determination unit determines the knowledge point weights of different groups of wrong test question pictures based on the semantic feature similarity; the knowledge point weights are positively correlated with the semantic feature similarity.

9. A knowledge point determination system based on incorrect test questions as claimed in claim 7, characterized in that: The target knowledge point determination module determines M groups of target knowledge points based on the assigned weight score of each wrong test question, the number of each group of wrong test question pictures and the preset benchmark score. The M groups of target knowledge points correspond to the M mobile terminals; The target knowledge point determination module also includes a knowledge point pushing unit; The knowledge point pushing unit pushes the M groups of target knowledge points to the corresponding M mobile terminals.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, it is used to implement a method for determining knowledge points based on incorrect test questions as described in any one of claims 1 to 6 above.

Citation Information

Patent Citations

  • Wrong sentence statistical method and system for teaching

    CN102436547A

  • Combined wrong question recommendation method based on knowledge graph

    CN107273490A

  • Intelligent recommendation method, device and equipment and computer storage medium

    CN113934922A

  • Weak knowledge point identification method and learning path planning method and device

    CN114254208A

  • Student ability diagnosis method, related equipment and storage medium

    CN114358570A