CN111625631A - Method for generating option of choice question - Google Patents
Method for generating option of choice question Download PDFInfo
- Publication number
- CN111625631A CN111625631A CN202010290665.6A CN202010290665A CN111625631A CN 111625631 A CN111625631 A CN 111625631A CN 202010290665 A CN202010290665 A CN 202010290665A CN 111625631 A CN111625631 A CN 111625631A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- options
- question
- candidate
- candidate options
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3322—Query formulation using system suggestions
- G06F16/3323—Query formulation using system suggestions using document space presentation or visualization, e.g. category, hierarchy or range presentation and selection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/02—Knowledge representation; Symbolic representation
- G06N5/022—Knowledge engineering; Knowledge acquisition
- G06N5/025—Extracting rules from data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Computing Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Algebra (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
The invention provides a method for generating option options of choice questions, which comprises the following steps: obtaining a question making record of a target object, and obtaining a knowledge set from the question making record; acquiring a question to be made of the target object from a question bank according to the knowledge set; acquiring candidate options from the question bank according to knowledge points contained in the to-be-asked questions, and generating alternate options of the to-be-asked questions according to the candidate options; the learning method and the learning system can be used for strengthening the mastering condition of weak knowledge points and improving the learning efficiency by combining learning situation analysis and option generation.
Description
Technical Field
The invention relates to the field of intelligent education, in particular to a method for generating option choices of choice questions.
Background
Most of the existing problem schemes for generating choices of choice questions generate choices aiming at the question and real answers existing in the question, and in detail, some of the existing related technologies directly and randomly extract choices in a choice library, and some new technologies select other generated choices by analyzing semantic and form similarity among the choices in the choice library and combining the content of the question stem.
Most of the prior art focuses on a single direction, and in selecting a question and generating a question, the prior art determines other generated choices by training a data model under the condition that a real answer is known. In addition, in the case of obtaining known correct answers, the selection of other answers is realized by randomly filling wrong choices into a choice list. Although the methods realize the generation of the options, the method does not consider the information of the subject making person, and the problems and the options generated by students of different levels cannot be distinguished. On the aspect of student learning situation analysis, various learning situation analysis methods are provided in the prior art, such as a traditional personalized Bayes knowledge tracking model, an improved Bayes neural network deep knowledge tracking method, a complex neural network algorithm and the like, but such methods only perform modeling analysis on the learning process of students (such as the process of doing exercises by the students) and further obtain the mastering conditions of the students on the existing knowledge points through the models. The method is mainly applied to the personalized question recommendation of students on an online learning platform at present.
Disclosure of Invention
In view of the problems in the prior art, the invention provides a choice question option generation method, which mainly solves the problem that the existing choice question option generation method does not consider the knowledge point mastering condition of students.
In order to achieve the above and other objects, the present invention adopts the following technical solutions.
A choice question option generating method comprises the following steps:
obtaining a question making record of a target object, and obtaining a knowledge set from the question making record;
acquiring a question to be made of the target object from a question bank according to the knowledge set;
and acquiring candidate options from the question bank according to the knowledge points contained in the to-be-asked questions, and generating the alternative items of the to-be-asked questions according to the candidate options.
Optionally, before obtaining the to-be-themed title of the target object, the method further includes: acquiring knowledge points in the question bank, sequencing the knowledge points in the question bank, and creating a knowledge bank;
and comparing the knowledge set with the knowledge base to obtain the learning progress of the target object, and obtaining the to-be-made question according to the learning progress.
Optionally, before obtaining the candidate option, the method further includes: processing the knowledge points contained in the to-be-themed items through a knowledge tracking algorithm to obtain the mastering condition of the target object on the corresponding knowledge points;
and screening corresponding topics according to the mastery condition of the knowledge points, and acquiring options of the corresponding topics as the candidate options.
Optionally, a screening threshold is set, and the candidate options are obtained through the screening threshold.
Optionally, the knowledge tracking algorithm comprises at least one of: bayesian knowledge tracking algorithm, recurrent neural network algorithm, and depth knowledge tracking algorithm.
Optionally, the method further includes screening the candidate options according to a preset screening policy.
Optionally, the preset screening policy includes at least one of: acquiring the answer right rate of the target object according to the right options of the questions, and screening out the candidate options with the right rate reaching a set threshold;
screening the candidate options according to the selection rate of the incorrect options of the subject;
and screening the candidate options according to the similarity between the candidate options and the correct options.
Optionally, the election rate is obtained according to the frequency of the incorrect options being selected and the frequency of the corresponding questions being answered.
Optionally, the obtaining manner of the similarity between the candidate option and the correct option includes at least one of: euclidean distance, cosine distance, paradigm distance.
Optionally, when the number of the candidate options obtained according to the screening threshold is smaller than the preset incorrect number of the candidate options, obtaining a corresponding number of candidate options from other candidate options of the corresponding title, and complementing the number of the candidate options.
As described above, the method for generating choice questions of the present invention has the following advantageous effects.
By analyzing the knowledge points in the question making record and generating the option of the selected question according to the learning progress and the knowledge point mastering condition, the method is beneficial to assisting the target object to deepen the memory of the knowledge points with poor mastering.
Drawings
Fig. 1 is a flowchart of a method for generating selection questions in an embodiment of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
Referring to FIG. 1, the present invention provides a method for generating choice questions, comprising steps S01-S03.
In step S01, a question record of the target object is obtained, and a knowledge set is obtained from the question record:
in one embodiment, a question making record of a target object (such as a student, a web lesson student, etc.) can be obtained through an online learning platform. Wherein, the question making record comprises the question making quantity, the correctness of the question making, the number of times of answering the same question and the like.
According to the question making record of the target object, all the questions made by the target object can be extracted. Further, knowledge points in all manufactured topics can be extracted, and a knowledge point set is created.
In step S02, a question to be made of the target object is obtained from the question bank according to the knowledge set;
in an embodiment, before obtaining the to-be-themed title of the target object, the method further includes: acquiring knowledge points in an item bank, sequencing the knowledge points in the item bank, and creating a knowledge bank;
and comparing the knowledge set with a knowledge base to obtain the learning progress of the target object, and obtaining the to-be-made question according to the learning progress.
Specifically, knowledge points contained in all questions in the question bank are obtained, all the knowledge points can be sorted according to the sequence of the knowledge points in the teaching text, and the specific sorting rule can be adjusted according to the actual situation. And storing the sequenced knowledge points in a database of the online learning platform to obtain a knowledge base.
The knowledge set comprises all knowledge points in the questions which have been made by the target object, and the knowledge points in the knowledge set are compared with the knowledge points in the knowledge base, so that the learning progress of the target object is determined. The knowledge points in the knowledge set can also be pre-sorted according to the sorting rules in the knowledge base so as to simplify the comparison process. For example, the knowledge set may be expressed as { A }1，A2，…，AkGet A by alignmentkAnd if the sequencing position of the knowledge point in the corresponding knowledge base is n, the n represents the learning progress of the target object. And extracting the topics corresponding to all previous knowledge points in the knowledge base n as the topics to be made of the target object. The to-be-themed items can be used for the target object to perform knowledge point reinforced memory.
In step S03, candidate options are obtained from the question bank according to the knowledge points included in the to-be-asked questions, and an alternative item for the to-be-asked questions is generated according to the candidate options.
In one embodiment, before the candidate options are obtained, the knowledge points contained in the to-be-made questions can be processed through a knowledge tracking algorithm, and the mastering condition of the target object on the corresponding knowledge points is obtained;
and screening the corresponding questions according to the mastery condition of the knowledge points, and acquiring options of the corresponding questions as candidate options.
In one embodiment, the knowledge tracking algorithm may include at least one of: bayesian knowledge tracking algorithm, recurrent neural network algorithm, and depth knowledge tracking algorithm.
Taking a bayesian knowledge tracking algorithm (BKT) as an example, the BKT models different knowledge points, and generally how many knowledge points are in training data and how many sets of corresponding (L0, T, G, S) parameters exist. Wherein:
l0: indicating the mastery degree of a knowledge point (i.e. the probability of mastering the knowledge point) when the student does not start to do the question or starts to do the knowledge point continuously; it can be obtained by averaging the training data, or by using experience, such as half-probability, which is generally known, and L0 is 0.5;
t: the probability that the knowledge points never reach the scholars after the students practice questions is represented;
g: the probability that the student does not master the knowledge point but still has a Monte Pair is represented;
s: the probability that the student actually mastered the knowledge point but made a mistake is shown;
a Hidden Markov Model (HMM) can be constructed through the 4 parameters, and the mastering condition of the knowledge points of the students can be obtained through training of the HMM.
A set of items to be topical may be denoted as Q ═ Q1，q2，q3，…，qiThe set of all the knowledge point correspondences contained in the item to be addressed can be denoted as K ═ K1，k2，k3，…，km}; then the knowledge point contained in one of the to-be-topical items can be represented as r (q)1)＝{k2，k5By this methodAnd obtaining the knowledge point representation contained in each to-be-made subject.
And converting the knowledge point representation contained in the to-be-asked question into a binary variable, and inputting the binary variable into a Bayesian knowledge tracking algorithm for training an HMM model. Specifically, assume that the current topic qtThe corresponding knowledge point is { kt，kt+1A, B, C, D, four options are contained in the title, wherein the AB option is the correct option; if the student selects AC, set kt to 1 and kt+1Is 0; select BC, then set kt to 0 and kt+1To 1, select CD, set kt and kt+1Are all 0. The Bayesian knowledge tracking algorithm outputs a vector with the length of m as the mastery condition of the student on the m knowledge points.
In an embodiment, according to the grasping condition of the target object on a certain to-be-made question including a knowledge point obtained in the previous step, a screening threshold value can be set to further screen other to-be-made questions in the question bank corresponding to the response knowledge point, and other items of the question corresponding to the knowledge point are extracted as candidate items. If the question q is to be made3Including knowledge point k2And k is2The other questions to be made which meet the screening threshold in the corresponding question bank are q4，q6，q9. Then q will be4，q6，q9Corresponding alternative as q3Is selected.
The screening threshold is used for judging whether the mastery condition of a target object to a knowledge point meets the requirement, the theoretical value range of the threshold is (0, 1), the value range can be usually set to be [0.7-0.9], for the knowledge points of which the mastery conditions of m knowledge points output by a knowledge tracking algorithm are lower than the screening threshold and are not equal to 0, the knowledge points can be used as the knowledge points with poor mastery condition corresponding to the target object, then based on the corresponding relation between the topics and the knowledge points, all the items corresponding to the knowledge points are extracted to be used as candidate items, concretely, when the English class selection questions are generated in an individualized way, according to the topic characteristics of the language class, the mastery requirement on one knowledge point is lower, but more differences used among words need to be distinguished, therefore, the screening threshold can be set to be 0.7, when the mathematics class selection questions are generated in an individualized way, due to the characteristics of the subject of mathematics, the requirement for mastering a knowledge point is generally higher, and most problems of students have certain commonality, so that the screening threshold value can be set to be 0.9, and the high requirement for mastering the knowledge point of the students is reflected. The screening threshold setting can be adjusted according to actual application requirements.
In one embodiment, the candidate options may be further filtered according to a preset filtering policy. Wherein, the preset screening strategy comprises at least one of the following:
acquiring the answer right rate of the target object according to the right option of the question, and screening out candidate options with the right rate reaching a set threshold;
screening candidate options according to the selection rate of the incorrect options of the subject;
and screening the candidate options according to the similarity between the candidate options and the correct options. In another embodiment, as a simplified method, the candidate options may also be filtered by calculating the distance between the candidate option and the correct option, and the calculation method may be represented as: in the case where only characters are allowed to be inserted, deleted, and replaced, the number of steps required to switch from the candidate to the correct one is calculated. If english words have homology, the distance requirement is set to be greater than 2, and the english words are arranged in descending order to distinguish different forms and different derivatives of english words, and if the correct answer is beauty, then words such as beautiful, beautify and the like are ranked in the top and then are preferentially used as candidates of the title to be provided to students who do the title.
Specifically, for a first strategy, all options in the option set can be extracted in advance as options of correct answers, then the answer correct rate in the question bank corresponding to the candidate options is used as a basis for sorting, and sorting is performed in an ascending order or a descending order according to requirements;
aiming at the second strategy, options in all option sets serving as incorrect answers can be extracted in advance, the selection rate is set according to the frequency of selecting the incorrect options and the frequency of answering the corresponding questions, and if the selection rate is divided by the frequency of selecting the candidates and the frequency of answering the questions, the selection rate is used as the basis of sorting, and sorting is carried out in an ascending order or a descending order according to the requirements;
for the third strategy, similarity calculation methods such as Euclidean distance, cosine distance or canonical distance can be adopted to obtain the similarity between the candidate option and the correct option, and ascending or descending sorting is carried out according to the similarity.
In an embodiment, according to the sorting result of the candidate options, a corresponding number of candidate options are selected from the candidate options as the candidate options corresponding to the to-be-themed items. If A is the correct option in a radio topic containing four options, 3 options are selected from the candidate options according to the steps and form the alternative of the radio topic together with the A option.
In an embodiment, when the number of candidate options obtained according to the screening threshold is smaller than a preset incorrect number of candidate options, obtaining a corresponding number of candidate options from other candidate options of a corresponding topic, and complementing the number of candidate options. If the candidate options in the sorting result are less than three, other candidate options beyond the screening threshold are extracted as the alternative options until three alternative options are reached.
And feeding back the questions to be made and the generated corresponding alternative items as practice questions to the target object.
In summary, the method for generating the option of the choice question combines the mastering condition of the knowledge point of the target object with the option generation, is favorable for strengthening the memory of the weak knowledge point and improves the learning efficiency; through learning situation analysis, the option of the selected question is screened based on the knowledge point mastering condition, and the auxiliary target object is facilitated to distinguish the easily mixed knowledge points; the exercise questions are integrated and generated through the exercise making records of the target object, so that the influence on learning efficiency and effect caused by the fact that the target object makes the same exercise questions repeatedly and vividly is avoided; the relevance between the knowledge points is fully utilized to generate the option of the choice question, which is beneficial to enhancing the flexibility of the application of the knowledge points. Therefore, the invention effectively overcomes various defects in the prior art and has high industrial utilization value.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.
Claims (10)
1. A method for generating choice questions, comprising:
obtaining a question making record of a target object, and obtaining a knowledge set from the question making record;
acquiring a question to be made of the target object from a question bank according to the knowledge set;
and acquiring candidate options from the question bank according to the knowledge points contained in the to-be-asked questions, and generating the alternative items of the to-be-asked questions according to the candidate options.
2. The choice option generating method according to claim 1, further comprising, before obtaining the to-be-themed theme of the target object: acquiring knowledge points in the question bank, sequencing the knowledge points in the question bank, and creating a knowledge bank;
and comparing the knowledge set with the knowledge base to obtain the learning progress of the target object, and obtaining the to-be-made question according to the learning progress.
3. The method of claim 1, further comprising, prior to obtaining the candidate options: processing the knowledge points contained in the to-be-themed items through a knowledge tracking algorithm to obtain the mastering condition of the target object on the corresponding knowledge points;
and screening corresponding topics according to the mastery condition of the knowledge points, and acquiring options of the corresponding topics as the candidate options.
4. The method of claim 3, wherein a filtering threshold is set, and the candidate item is obtained by the filtering threshold.
5. The choice generating method of claim 3, wherein the knowledge tracking algorithm comprises at least one of: bayesian knowledge tracking algorithm, recurrent neural network algorithm, and depth knowledge tracking algorithm.
6. The method of claim 3, further comprising screening the candidate options according to a preset screening policy.
7. The choice generating method of claim 6, wherein the preset screening strategy comprises at least one of: acquiring the answer right rate of the target object according to the right options of the questions, and screening out the candidate options with the right rate reaching a set threshold;
screening the candidate options according to the selection rate of the incorrect options of the subject;
and screening the candidate options according to the similarity between the candidate options and the correct options.
8. The method of claim 7, wherein the selection rate is obtained according to a frequency of selecting the incorrect item and a frequency of answering a corresponding item.
9. The method of claim 7, wherein the similarity between the candidate option and the correct option is obtained in at least one of the following manners: euclidean distance, cosine distance, paradigm distance.
10. The method of claim 7, wherein when the number of candidate options obtained according to the filtering threshold is smaller than the preset incorrect number of candidate options, a corresponding number of candidate options are obtained from other candidate options of a corresponding topic to complement the number of candidate options.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010290665.6A CN111625631B (en) | 2020-04-14 | 2020-04-14 | Method for generating option of choice question |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010290665.6A CN111625631B (en) | 2020-04-14 | 2020-04-14 | Method for generating option of choice question |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111625631A true CN111625631A (en) | 2020-09-04 |
| CN111625631B CN111625631B (en) | 2022-06-17 |
Family
ID=72258860
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010290665.6A Active CN111625631B (en) | 2020-04-14 | 2020-04-14 | Method for generating option of choice question |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111625631B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112216168A (en) * | 2020-10-21 | 2021-01-12 | 李帮军 | Intelligent question type conversion system and method based on choice question editor |
| CN113239689A (en) * | 2021-07-07 | 2021-08-10 | 北京语言大学 | Selection question interference item automatic generation method and device for confusing word investigation |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101000643A (en) * | 2006-01-10 | 2007-07-18 | 英业达股份有限公司 | Study system and method with review function |
| CN101477755A (en) * | 2009-01-06 | 2009-07-08 | 陈诗周 | Error problem database system for student |
| CN104299473A (en) * | 2013-07-20 | 2015-01-21 | 郑州学生宝电子科技有限公司 | Teaching and student learning error question knowledge base |
| US20150179082A1 (en) * | 2013-12-23 | 2015-06-25 | International Business Machines Corporation | Dynamic Identification and Validation of Test Questions from a Corpus |
| CN104834729A (en) * | 2015-05-14 | 2015-08-12 | 百度在线网络技术（北京）有限公司 | Title recommendation method and title recommendation device |
| WO2017124343A1 (en) * | 2016-01-20 | 2017-07-27 | 吴晓敏 | Method for pushing information during sorting-out of knowledge points, and learning system |
| CN109977230A (en) * | 2019-04-02 | 2019-07-05 | 上海乂学教育科技有限公司 | It is suitble to the multiple-choice question mistake of adaptive teaching because of analysis method |
| CN110399541A (en) * | 2019-05-31 | 2019-11-01 | 平安国际智慧城市科技股份有限公司 | Topic recommended method, device and storage medium based on deep learning |
-
2020
- 2020-04-14 CN CN202010290665.6A patent/CN111625631B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101000643A (en) * | 2006-01-10 | 2007-07-18 | 英业达股份有限公司 | Study system and method with review function |
| CN101477755A (en) * | 2009-01-06 | 2009-07-08 | 陈诗周 | Error problem database system for student |
| CN104299473A (en) * | 2013-07-20 | 2015-01-21 | 郑州学生宝电子科技有限公司 | Teaching and student learning error question knowledge base |
| US20150179082A1 (en) * | 2013-12-23 | 2015-06-25 | International Business Machines Corporation | Dynamic Identification and Validation of Test Questions from a Corpus |
| CN104834729A (en) * | 2015-05-14 | 2015-08-12 | 百度在线网络技术（北京）有限公司 | Title recommendation method and title recommendation device |
| WO2017124343A1 (en) * | 2016-01-20 | 2017-07-27 | 吴晓敏 | Method for pushing information during sorting-out of knowledge points, and learning system |
| CN109977230A (en) * | 2019-04-02 | 2019-07-05 | 上海乂学教育科技有限公司 | It is suitble to the multiple-choice question mistake of adaptive teaching because of analysis method |
| CN110399541A (en) * | 2019-05-31 | 2019-11-01 | 平安国际智慧城市科技股份有限公司 | Topic recommended method, device and storage medium based on deep learning |
Non-Patent Citations (4)
| Title |
|---|
| BEI XU ET AL.: "Automatic wrong option generation based on confusion degree", 《2019 15TH INTERNATIONAL CONFERENCE ON SEMANTICS, KNOWLEDGE AND GRIDS》 * |
| LEYLA ZHUHADAR ET AL.: "Is there a gender difference in interacting with intelligent tutoring system? Can Bayesian Knowledge Tracing and Learning Curve Analysis Models answer this question?", 《COMPUTERS IN HUMAN BEHAVIOR》 * |
| 杨洲权: "基于B/S模式的高校无纸化考试系统设计与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 苏喻 等: "学生得分预测： 一种基于知识图谱的卷积自编码器", 《中国科学技术大学学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112216168A (en) * | 2020-10-21 | 2021-01-12 | 李帮军 | Intelligent question type conversion system and method based on choice question editor |
| CN113239689A (en) * | 2021-07-07 | 2021-08-10 | 北京语言大学 | Selection question interference item automatic generation method and device for confusing word investigation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111625631B (en) | 2022-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109903617B (en) | Personalized exercise method and system | |
| KR102194837B1 (en) | Method and apparatus for answering knowledge-based question | |
| CN110991645A (en) | Self-adaptive learning method, system and storage medium based on knowledge model | |
| CN112784608B (en) | Test question recommending method and device, electronic equipment and storage medium | |
| CN112508334A (en) | Personalized paper combining method and system integrating cognitive characteristics and test question text information | |
| CN108509439B (en) | Algebra teaching system | |
| CN110309300B (en) | Method for identifying knowledge points of physical examination questions | |
| CN109949637B (en) | Automatic answering method and device for objective questions | |
| US11989524B2 (en) | Knowledge-grounded dialogue system and method for language learning | |
| Heilman et al. | Personalization of reading passages improves vocabulary acquisition | |
| CN113344053B (en) | Knowledge tracking method based on examination question different composition representation and learner embedding | |
| CN111625631B (en) | Method for generating option of choice question | |
| CN112685550B (en) | Intelligent question-answering method, intelligent question-answering device, intelligent question-answering server and computer readable storage medium | |
| Bunch et al. | Automated scoring in assessment systems | |
| CN114201684A (en) | Knowledge graph-based adaptive learning resource recommendation method and system | |
| CN113239209A (en) | Knowledge graph personalized learning path recommendation method based on RankNet-transformer | |
| Singla et al. | From {Solution Synthesis} to {Student Attempt Synthesis} for block-based visual programming tasks | |
| CN116258613B (en) | Course planning method, course planning device, and readable storage medium | |
| CN116932778A (en) | Teaching knowledge mining method and system based on VR technology | |
| CN109800880B (en) | Self-adaptive learning feature extraction system based on dynamic learning style information and application | |
| CN111930908A (en) | Answer recognition method and device based on artificial intelligence, medium and electronic equipment | |
| CN116561274A (en) | Knowledge question-answering method based on digital human technology and natural language big model | |
| CN115617974A (en) | Dialogue processing method, device, equipment and storage medium | |
| CN112256869B (en) | Same-knowledge-point test question grouping system and method based on question meaning text | |
| CN116401341A (en) | Interactive answering system oriented to understanding |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |