Test question generation method, device and system, computer storage medium and program product
The method addresses the limitations of conventional question banks by dynamically generating test questions from multiple databases, ensuring accuracy and adaptability to user needs and educational reforms, providing timely and flexible question formats.
Patent Information
- Application Number
- JP2023550270
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2021-02-20
- Filing Date
- 2022-02-18
- Publication Date
- 2024-11-27
- Estimated Expiration
- 2042-02-18
AI Technical Summary
Conventional question creation services are limited by existing question banks, unable to generate test questions for specific answer steps, fail to meet normativity requirements of different exam formats, and lack effective user response analysis, leading to insufficient and outdated question generation.
A method and system that generates test questions by combining initial and intermediate elements from databases, including knowledge bases, user history, and pattern databases, adjusting parameters to meet preset conditions, and dynamically updating to ensure accuracy and relevance.
Enables the creation of test questions that accurately reflect user needs, adapt to educational reforms, and provide timely, flexible question formats, enhancing the directionality and accuracy of question generation.
Smart Images

Figure 0007590743000001 
Figure 0007590743000002 
Figure 0007590743000003
Abstract
Description
[Technical field]
[0001] The present disclosure relates to the field of computer-aided education, and in particular to a test question generation method, a test question generation device, a test question generation system, a method of using the test question generation system, an electronic device, a computer-readable storage medium, and a computer program product. [Background technology]
[0002] In recent years, with the advancement of big data technology, various algorithms and related hardware computing power, and smart terminal devices, computer-aided teaching technology has developed at a dizzying pace. Various smart education websites, applications, and related devices provide large-scale and smart teaching support services for both educators and learners. Smart question creation services have always been the focus of attention in computer-aided teaching services. Users hope to use the relevant smart question creation services of service providers to obtain test questions for their weak knowledge items and receive training.
[0003] At present, the conventional question creation service mainly generates questions and question papers from existing question banks. Specifically, the conventional question creation service selects the test questions or question papers composed of multiple test questions required by the user from the existing question bank through a recommendation system based on the content to be tested, and provides them to the user. However, the question bank can only recognize the knowledge items related to each question as a whole, and the question itself may have multiple different answer steps, and each answer step may correspond to a different specific knowledge item, so the question creation service from the question bank cannot create test questions for the knowledge items related to a specific answer step, and is insufficiently directional. The question creation service from the question bank is limited by the question bank itself, and the test questions it can create are only the existing test questions in the question bank itself, so it cannot meet the normative requirements of test questions for different test formats. The past question creation service from the question bank cannot respond to the demand of education and evaluation reform to provide test questions with the latest timeliness. In addition, due to the lack of an effective method of communication between the question creation service and the users who use the service, providers of the question creation service generally have difficulty understanding the user's answering status, and are therefore unable to provide a more accurate question creation service by analyzing the user's current answers and historical records. Summary of the Invention [Problem to be solved by the invention]
[0004] The present disclosure provides a test question generation method, a test question generation device, a test question generation system, a method for using the test question generation system, an electronic device, a computer-readable storage medium, and a computer program product. [Means for solving the problem]
[0005] According to one aspect of the present disclosure, there is provided a test question generation method including: a preset condition receiving step of receiving a preset condition; an initial element generating step of generating a test question initial element from at least one preset database based on the preset condition; an intermediate element generating step of generating a test question intermediate element from the test question initial element and the at least one preset database based on the preset condition; and a test question determining step of determining whether a test question constituted by the test question initial element and the test question intermediate element satisfies the preset condition, and if the test question does not satisfy the preset condition, returning to the test question initial element generating step or the test question intermediate element generating step, and if the test question satisfies the preset condition, entering a test question generating step and generating the test question from the test question initial element and the test question intermediate element, wherein one of the test question initial element and the test question intermediate element generated by executing the test question intermediate element generating step most recently is for generating an answer to the test question, and the other of the two is for generating a question of the test question.
[0006] Furthermore, according to the test question generation method of one aspect of the present disclosure, the test question intermediate elements other than the test question intermediate element generated by executing the test question intermediate element generation step most recently correspond to the answer step of the test question.
[0007] Furthermore, according to a test question generation method of one aspect of the present disclosure, the at least one preset database includes a knowledge base database, a test information database, a user group pattern database, and a user history database.
[0008] Furthermore, according to one embodiment of the test question generation method of the present disclosure, the knowledge base database is for storing information related to the knowledge to be tested, including one or more of basic knowledge information, common sense information, answering approach information, answering method information, and information on areas where mistakes are likely to be made.
[0009] In addition, according to one embodiment of the test question generation method of the present disclosure, the test information base is for storing information related to the test, including one or more of question paper configuration information, question format information, difficulty distribution information, test time information, and test style information.
[0010] In addition, according to one embodiment of the test question generation method of the present disclosure, the user group pattern database is for storing information related to a user group, and the information related to the user group is obtained from the usage record of the user group and includes one or more of answer speed information, knowledge item information that is prone to mistakes, answering habit information, and psychological tolerance limit information of a particular user group.
[0011] In addition, according to one embodiment of the test question generation method of the present disclosure, the user history database is for storing information related to a particular user, including one or more of the particular user's answer history information, knowledge acquisition information, and learning progress information.
[0012] In addition, according to a test question generation method of one aspect of the present disclosure, the preset conditions include one or more of a question format constraint condition, a test content constraint condition, and a test style constraint condition.
[0013] In addition, according to one aspect of the test question generation method of the present disclosure, if the test question satisfies the preset conditions, generating the test question from the test question initial element and the test question intermediate element further includes generating additional elements based on question format constraint conditions in the preset conditions, and combining the test question initial element, the test question intermediate element, and the additional element to generate the test question that satisfies the question format constraint conditions.
[0014] According to a test question generation method of one aspect of the present disclosure, the test question initial element generation step includes determining the test items from the knowledge base database, the user history database, and / or the user group pattern database.
[0015] According to a test question generating method of one aspect of the present disclosure, the test question intermediate element generating step includes determining an answer concept from the test question initial element and the at least one preset database.
[0016] In addition, according to one aspect of the test question generation method of the present disclosure, the method further includes a question paper generation step of repeatedly executing the test question initial element generation step, the test question intermediate element generation step, the test question determination step, and the test question generation step to generate a plurality of test questions, and a question paper determination step of determining whether the plurality of test questions satisfy the preset condition, and if the plurality of test questions do not satisfy the preset condition, returning to the question paper generation step and generating one or more of the plurality of test questions again, and if the plurality of test questions satisfy the preset condition, entering into an overall generation step and generating a question paper from the plurality of test questions.
[0017] In addition, according to one aspect of the test question generation method of the present disclosure, the question paper determination step includes generating global constraint parameters from the preset conditions, the global constraint parameters including one or more of an overall question format parameter, a parameter for each test question alone, a related parameter for each test question overall, a generation time parameter, a maximum allowable error parameter, and an answer time parameter, and determining whether the multiple test questions satisfy the global constraint parameters.
[0018] In addition, according to one aspect of the test question generation method of the present disclosure, if the plurality of test questions do not satisfy the preset conditions, returning to the question paper generation step and regenerating one or more of the plurality of test questions includes dynamically adjusting the overall constraint parameters, returning to the question paper generation step, and regenerating one or more of the plurality of test questions.
[0019] Furthermore, according to the test question generation method of one aspect of the present disclosure, the generation error of one or more of the plurality of test questions is smaller than a predetermined threshold value.
[0020] In addition, according to one aspect of the test question generation method of the present disclosure, the test question initial element and the test question intermediate element each have a corresponding first parameter range interval, and the test question initial element and the test question intermediate element generated after returning to the test question initial element generation step each have a corresponding second parameter range interval, and the second parameter range interval is smaller than the first parameter range interval.
[0021] According to another aspect of the present disclosure, there is provided a test question generating unit, comprising: a preset condition receiving unit configured to receive a preset condition; an initial element generating unit configured to generate a test question initial element from at least one preset database based on the preset condition; an intermediate element generating unit configured to generate a test question intermediate element from the test question initial element and the at least one preset database based on the preset condition; a test question determining unit configured to determine whether a test question constituted by the test question initial element and the test question intermediate element satisfies the preset condition; and a test question generating unit configured to generate the test questions from the test question initial element and the test question intermediate element, and a test question generation unit that, if the test question does not satisfy the preset condition, returns to generating the test question initial element again by the test question initial element generation unit, or generates the test question intermediate element again by the test question intermediate element generation unit, and, if the test question satisfies the preset condition, generates the test question from the test question initial element and the test question intermediate element by the test question intermediate element generation unit, wherein one of the test question initial element and the test question intermediate element most recently generated by the test question intermediate element generation unit is for generating an answer to the test question, and the other of the two is for generating a question of the test question.
[0022] According to another aspect of the present disclosure, there is provided a test question generation method including: a preset condition receiving step of receiving preset conditions; a test question generating step of generating test questions from at least one preset database based on the preset conditions; a question paper generating step of repeatedly executing the test question generating step a plurality of times to generate a plurality of test questions, and then organizing the plurality of test questions to generate a question paper; and a question paper determining step of determining whether the question paper satisfies the preset conditions, returning to the test question generating step or the question paper generating step if the question paper does not satisfy the preset conditions, and outputting the question paper if the question paper satisfies the preset conditions.
[0023] According to a test question generating method according to another aspect of the present disclosure, the at least one preset database includes a knowledge base database, a test information database, a user group pattern database, and a user history database.
[0024] In addition, according to another aspect of the test question generation method of the present disclosure, the knowledge base database is for storing information related to the knowledge to be tested, including one or more of basic knowledge information, common sense information, answering method information, answering method information, and information on easy-to-make mistakes.
[0025] In addition, according to another aspect of the test question generation method of the present disclosure, the test information base is for storing information related to the test, including one or more of question paper configuration information, question format information, difficulty distribution information, test time information, and test style information.
[0026] In addition, according to another aspect of the test question generation method of the present disclosure, the user group pattern database is for storing information related to a user group, and the information related to the user group is obtained from the usage record of the user group and includes one or more of answering speed information, knowledge item information that is prone to mistakes, answering habit information, and psychological tolerance limit information of a particular user group.
[0027] In addition, according to another aspect of the test question generation method of the present disclosure, the user history database is for storing information related to a particular user, including one or more of the particular user's answer history information, knowledge acquisition information, and learning progress information.
[0028] According to a test question generation method according to another aspect of the present disclosure, the preset conditions include one or more of a question type constraint condition, a test content constraint condition, and a test style constraint condition.
[0029] According to another aspect of the test question generation method of the present disclosure, the test question generation step includes an initial element generation substep of generating a test question initial element from at least one preset database based on the preset conditions, an intermediate element generation substep of generating a test question intermediate element from the test question initial element and the at least one preset database based on the preset conditions, and a test question determination substep of determining whether a test question composed of the test question initial element and the test question intermediate element satisfies the preset conditions, and if the test question does not satisfy the preset conditions, returning to the test question initial element generation substep or the test question intermediate element generation substep, and if the test question satisfies the preset conditions, entering the test question generation substep to generate the test question from the test question initial element and the test question intermediate element, and one of the test question initial element and the test question intermediate element generated by executing the test question intermediate element generation step most recently is set as the answer of the test question, and the other of the two is set as the question of the test question.
[0030] In addition, according to another aspect of the test question generation method of the present disclosure, the test question intermediate elements other than the test question intermediate element generated by executing the test question intermediate element generation sub-step most recently correspond to the answer step of the test question.
[0031] In addition, according to another aspect of the test question generation method of the present disclosure, if the test question satisfies the preset condition, generating the test question from the test question initial element and the test question intermediate element further includes generating an additional element based on a question format constraint condition in the preset condition, and combining the test question initial element, the test question intermediate element and the additional element to generate the test question that satisfies the question format constraint condition.
[0032] According to another aspect of the test question generation method of the present disclosure, the test question initial element generation sub-step includes determining the test items from the knowledge base database, the user history database and / or the user group pattern database.
[0033] According to a test question generating method according to another aspect of the present disclosure, the test question intermediate element generating sub-step includes determining an answer concept from the test question initial element and the at least one preset database.
[0034] In addition, according to another aspect of the test question generation method of the present disclosure, if the question paper does not satisfy the preset conditions, returning to the test question generation step or the question paper generation step includes, if a plurality of test questions do not satisfy the preset conditions, returning to the question paper generation step and generating one or more of the plurality of test questions again.
[0035] In addition, according to another aspect of the test question generation method of the present disclosure, the question paper determination step includes generating overall constraint parameters from the preset conditions, the overall constraint parameters including one or more of an overall question format parameter, a parameter for each test question alone, a related parameter for each test question overall, a generation time parameter, and a maximum allowable error parameter, and determining whether the multiple test questions satisfy the overall constraint parameters.
[0036] In addition, according to another aspect of the test question generation method of the present disclosure, if the plurality of test questions do not satisfy the preset conditions, returning to the question paper generation step and regenerating one or more of the plurality of test questions includes dynamically adjusting the overall constraint parameters, returning to the question paper generation step, and regenerating one or more of the plurality of test questions.
[0037] According to a test question generation method according to another aspect of the present disclosure, the generation error of one or more of the plurality of test questions is smaller than a predetermined threshold value.
[0038] In addition, according to another aspect of the test question generation method of the present disclosure, the test question initial element and the test question intermediate element each have a corresponding first parameter range interval, and the test question initial element and the test question intermediate element generated after returning to the test question initial element generation sub-step each have a corresponding second parameter range interval, and the second parameter range interval is smaller than the first parameter range interval.
[0039] According to yet another aspect of the present disclosure, there is provided a test question generation device, including: a preset condition receiving unit configured to receive preset conditions; a test question generation unit configured to generate test questions from at least one preset database based on the preset conditions; a question paper generation unit configured to generate a question paper by organizing a plurality of test questions after the test question generation unit generates the plurality of test questions; and a question paper determination unit configured to determine whether the question paper satisfies the preset conditions, and if the question paper does not satisfy the preset conditions, to generate new test questions by the test question generation unit or to generate a new question paper by the question paper generation unit, and if the question paper satisfies the preset conditions, to output the question paper.
[0040] According to yet another aspect of the present disclosure, there is provided a method for using a test question generation system, the method including: a preset condition receiving step for receiving preset conditions; a test question / question paper generating step for generating test questions and / or question papers based on the test question generation method; an answer result receiving step for receiving answer results in which a user answers the test questions and / or question papers; an answer result recording step for recording the answer results on an answer step basis; and a database updating step for updating the at least one preset database from the answer results.
[0041] According to yet another aspect of the present disclosure, there is provided a test question generation system including: a preset condition receiving unit configured to receive preset conditions; a test question / question paper generating unit configured to generate test questions and / or question papers using the test question generation method; an answer result receiving unit configured to receive answer results of a user answering the test questions and / or question papers; an answer result recording unit configured to record the answer results on an answer step basis; and a database updating unit configured to update the at least one preset database from the answer results.
[0042] According to yet another aspect of the present disclosure, there is provided an electronic device including a memory for storing computer readable instructions and a processor for executing the computer readable instructions and performing the test question generation method.
[0043] According to yet another aspect of the present disclosure, there is provided a computer-readable storage medium for storing computer-readable instructions, which, when executed by a computer, cause the computer to perform the test question generation method.
[0044] According to yet another aspect of the present disclosure, there is provided a computer program product including computer program instructions which, when executed by a processor, realise the test question generation method. Effect of the Invention
[0045] As described in detail below, the test question generation method, test question generation device, test question generation system, method of using the test question generation system, electronic device, computer-readable storage medium, and computer program product of the embodiments of the present disclosure realize that when there is no question bank, the user can be provided with a single test question or a set of question papers that meet different test formats in a directional manner, with the knowledge items related to the answer step as the smallest unit, and the difficulty, type, and number of test questions can be flexibly configured based on the user's needs. Furthermore, they can promptly respond to the demands of education and evaluation reform to provide test questions that are up-to-date and timely, and can update related information and data based on the user's answer status, thereby providing more accurate question creation services later.
[0046] It should be understood that both the foregoing general description and the following detailed description are exemplary and intended to provide further explanation of the claimed technology. [Brief description of the drawings]
[0047] The above and other objects, features and advantages of the present disclosure will become more apparent by describing the embodiments of the present disclosure in more detail with reference to the drawings. The drawings are provided for a better understanding of the embodiments of the present disclosure, and are provided for explaining the present disclosure together with the embodiments of the present disclosure as a part of the specification, and are not intended to be limiting of the present disclosure. In the drawings, the same reference numerals generally represent the same parts or steps. [Figure 1] FIG. 1 is a schematic diagram of a test question generation system according to an embodiment of the present disclosure. [Diagram 2] FIG. 2 is a flow chart illustrating a test question generation method according to an embodiment of the present disclosure. [Diagram 3] FIG. 3 is a schematic diagram illustrating a test question generation method according to an embodiment of the present disclosure. [Figure 4] FIG. 4 is a schematic diagram of a specific example of a test question generation method according to an embodiment of the present disclosure. [Diagram 5] FIG. 5 is a flow chart illustrating a test question generation method according to an embodiment of the present disclosure. [Figure 6] FIG. 6 is a block diagram illustrating a test question generation device according to an embodiment of the present disclosure. [Figure 7] FIG. 7 is a block diagram illustrating another test question generating device according to an embodiment of the present disclosure. [Figure 8] FIG. 8 is a flow chart illustrating a method of using a test question generation system according to an embodiment of the present disclosure. [Figure 9] FIG. 9 is a block diagram illustrating a test question generation system according to an embodiment of the present disclosure. [Figure 10] FIG. 10 is a block diagram illustrating hardware of an electronic device according to an embodiment of the present disclosure. [Figure 11] FIG. 11 is a schematic diagram illustrating a computer-readable storage medium according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0048] In order to make the objectives, technical solutions and advantages of the present disclosure more apparent, exemplary embodiments of the present disclosure will be described in detail below with reference to the drawings. Note that the described embodiments are only a part of the embodiments of the present disclosure, and are not all of the embodiments of the present disclosure. It should be understood that the present disclosure is not limited to the exemplary embodiments described herein.
[0049] First, with reference to FIG. 1, a test question generation system according to an embodiment of the present disclosure will be generally described.
[0050] 1, a test question generation device 100 is configured in a test question generation system 10. Users 101a to 101c use the test question creation service provided by the test question generation system 10 by accessing the test question generation device 100 in the test question generation system 10 via a communication network 103 using their respective user devices 102a to 102c.
[0051] 1, in one embodiment of the present disclosure, the user devices 102a-102c, the test question generator 100, can be operated to connect to the communication network 103 by various wired or wireless communication protocols, such as Transmission Control Protocol and Internet Protocol (TCP / IP), User Datagram Protocol (UDP), Hypertext Transfer Protocol (HTTP), and / or File Transfer Protocol (FTP). Examples of the communication network 103 may include, but are not limited to, the Internet, an IPTV network, plain old telephone service (POTS), a Wi-Fi (Wi-Fi) network, a wireless local area network (WLAN), a local area network (LAN), and / or a metropolitan area network (MAN).
[0052] It should be noted that the test question generation device 100 in the test question generation system 10 shown in FIG. 1 is merely exemplary, and in other embodiments of the present disclosure, the test question generation device 100 according to the embodiments of the present disclosure may be, for example, suitable logic, circuitry, interfaces and / or code configured in user devices 102a-102c, which are used to provide question creation services to users 101a-101c.
[0053] In the test question generation device 100 in the test question generation system 10 according to the embodiment of the present disclosure, a database 104 is exemplarily configured. Alternatively, the database 104 may be configured separately from the test question generation device 100, and the test question generation device 100 can access the database 104 via various wired or wireless communication protocols.
[0054] As described in detail below, the test question generation system 10 according to an embodiment of the present disclosure does not directly select test questions or question papers available to the user from, for example, a question bank database using the test question generation device 100, but instead provides the user with a single test question or a set of question papers that meet different test formats in a directional manner, with the knowledge items related to the answer step as the smallest unit, when a question bank is not available.The difficulty, type, and number of test questions can be flexibly configured based on the user's needs, and the demand for education and evaluation reform to provide test questions that are up-to-date and timely can be met in a timely manner.Based on the user's answering status, related information and data can be updated, allowing for the provision of more accurate question creation services later.
[0055] Next, the process of generating test questions using the test question generating method according to the embodiment of the present disclosure will be described with reference to Figures 2 to 4. Figure 2 is a flowchart showing the test question generating method according to the embodiment of the present disclosure, Figure 3 is a schematic diagram showing the test question generating method according to the embodiment of the present disclosure, and Figure 4 is a schematic diagram of a specific example of the test question generating method according to the embodiment of the present disclosure.
[0056] As shown in FIGS. 2 and 3, the test question generation method according to an embodiment of the present disclosure includes the following steps.
[0057] In step S201, a preset condition receiving step is executed to receive the preset conditions.
[0058] In one embodiment of the present disclosure, the test question generation device 100 receives preset conditions 300 input by a user. The preset conditions 300 include one or more of a question type constraint, a test content constraint, and a test style constraint. For example, the question type constraint in the preset conditions 300 is for setting to generate one or more of a whole, an answer question, a multiple choice question, a judgment question, a fill-in-the-blank question, a connecting question, etc. The test content constraint in the preset conditions 300 is for setting one or more of a knowledge range to be tested, a difficulty level, an expected completion time, etc. The test style constraint in the preset conditions 300 is for setting one or more of a question paper configuration, a test question creation style, etc.
[0059] Alternatively, the preset conditions 300 may be automatically generated by the test question generation device 100 based on the recognized user and its related information (e.g., the user's age, gender, grade, region, history of solving questions, etc.) to suit the user's current learning needs and learning progress.
[0060] In step S202, an initial element generating step is executed to generate an initial element of a test question from at least one preset database based on the preset conditions.
[0061] In one embodiment of the present disclosure, the at least one preset database includes, for example, a knowledge base database 1041, a test information database 1042, a user group pattern database 1043, and a user history database 1044.
[0062] Specifically, the knowledge base database 1041 is for storing information related to the knowledge to be tested, including one or more of basic knowledge information, common sense information, answering method information, answering method information, and information on easy-to-make mistakes.
[0063] The exam information base 1042 is for storing information related to exams, including one or more of question paper configuration information, question format information, difficulty distribution information, exam time information, and exam style information.
[0064] The user group pattern database 1043 is for storing information related to a user group, and the information related to the user group is obtained from the usage records of the user group, and includes one or more of answering speed information, knowledge item information that is likely to be mistaken, answering habit information, and psychological tolerance limit information of a specific user group.
[0065] The user history database 1044 is for storing information related to a particular user, including one or more of answer history information, knowledge acquisition information, and learning progress information of the particular user.
[0066] It is easy to understand that the at least one preset database according to the embodiment of the present disclosure is not limited to the exemplary database and may include any other database. Similarly, the information stored in the at least one preset database is not limited to the exemplary information and may include any other information.
[0067] Furthermore, the at least one preset database may be dynamically updated.
[0068] On the one hand, the knowledge base database 1041 can be dynamically updated according to the educational outline, teaching materials, and the progress of educational research, thereby ensuring that it can meet the latest knowledge acquisition requirements in real time, and the examination information base 1042 can be dynamically updated according to the latest published examination outline and examination information, thereby ensuring that it can meet the latest examination requirements in real time. In other words, the dynamic update of the database realizes the timeliness and standardity of the creation of examination questions.
[0069] On the other hand, the user group pattern database 1043 and the user history database 1044 dynamically record the relevant information of a specific user and a specific user group in real time, thereby ensuring that directional training can be provided to the specific user and the specific user group.
[0070] In one embodiment of the present disclosure, taking a response question as an example, the test question initial element test question generated from at least one preset database based on the preset conditions is the test question 301. In another embodiment of the present disclosure, taking a response question as an example, the test question initial element is the last answer step n of the reference answer 302 of the test question. In other words, the test question generation method according to the embodiment of the present disclosure may generate answers in a forward direction from the generation of the question, or may generate questions in a backward direction from the generation of the final answer.
[0071] Specifically, the test question generation method according to the embodiment of the present disclosure determines the test contents, question formats, styles, etc. to be tested based on the preset conditions 300, and determines the test items to be tested, etc. from the relevant information in the knowledge base database 1041, the user group pattern database 1043, and the user history database 1044, and ensures that the generated results meet the criteria of the actual test from the test information database 1042. In addition, for a user who uses for the first time or a user who has insufficient relevant records in the user history database 1044, the default group user parameters in the user group pattern database 1043 can be used.
[0072] In step S203, an intermediate element generating step is performed to generate test question intermediate elements from the test question initial elements and the at least one preset database based on the preset conditions.
[0073] In one embodiment of the present disclosure, taking the answer question as an example, generating the test question intermediate element in step S203 after generating the question 301 of the test question in step S202 is answer step 1. In another embodiment of the present disclosure, similarly taking the answer question as an example, generating the test question intermediate element in step S203 after generating the last answer step n of the reference answer 302 of the test question in step S202 is answer step n-1.
[0074] Specifically, the test question generation method according to an embodiment of the present disclosure evaluates what answer concept should be used to construct the test question from the initial test question elements (e.g., the generation result of answer step n) using related data from at least one preset database, so that the relationship between answer step n-1 and answer step n is optimized and satisfies the preset conditions.
[0075] In step S204, a test question determination step is executed to determine whether the test question constituted by the test question initial element and the test question intermediate element satisfies the preset condition.
[0076] If a negative result is obtained in step S204, that is, the test question constituted by the test question initial element and the test question intermediate element does not satisfy the preset condition, the process returns to the test question initial element generating step S202 or the test question intermediate element generating step S203.
[0077] Specifically, by returning to the test question initial element generation step S202 or the test question intermediate element generation step S203, more intermediate answer steps can be generated (for example, after answer step n-1, answer step n-2 to answer step 1 can be generated), and the generation parameters can be adjusted to generate answer step n-1 or answer step n that was already generated last time, which ultimately helps to construct a test question that satisfies the preset conditions from the test question initial element and the test question intermediate element.
[0078] If a positive result is obtained in step S204, that is, the test question constituted by the test question initial element and the test question intermediate element satisfies the preset condition, the process enters step S205.
[0079] In step S205, a test question generation step is executed to generate the test questions from the test question initial elements and the test question intermediate elements.
[0080] Specifically, one of the test question initial element and the test question intermediate element generated by executing the test question intermediate element generation step most recently is for generating an answer to the test question, and the other of the two is for generating a question of the test question. The other test question intermediate elements except for the test question intermediate element generated by executing the test question intermediate element generation step most recently correspond to the answer step of the test question.
[0081] That is, in one embodiment of the present disclosure, taking a response question as an example, after generating question 301 of a test question in step 202, the test question intermediate element generated by executing the test question intermediate element generating step S203 once most recently is the last answer step n of reference answer 302. Similarly, in another embodiment of the present disclosure, taking a response question as an example, after generating the last answer step n of reference answer 302 of a test question in step 202, the test question intermediate element generated by executing the test question intermediate element generating step S203 once most recently is question 301 of the test question.
[0082] In addition, if the test question satisfies the preset conditions, generating the test question from the test question initial element and the test question intermediate element further includes generating additional elements based on question format constraint conditions in the preset conditions, and combining the test question initial element, the test question intermediate element and the additional elements to generate the test question that satisfies the question format constraint conditions.
[0083] In one embodiment of the present disclosure, the additional element may be relevant explanatory text, for example, the directly generated content is: Completing the Square: x 2 +4x-5=(x+5)(x-1) is used. After generating the related additional elements, the final combination is as follows. Although it does not conform to the perfect square method, it conforms to the conditions for using the method of completing the square because -5=-1×5, 4=5-1. Here, the method of completing the square: x 2 Use +4x-5=(x+5)(x-1).
[0084] In addition, generating additional elements based on the question type constraints in the preset conditions further includes generating and combining additional elements based on different question types. For example, if the question type is a multiple choice question, use existing test question initial elements and test question intermediate elements to generate a question part of the multiple choice question and a corresponding correct option, and optionally supplement other incorrect options. If the question type is a judgment question, use existing test question initial elements and test question intermediate elements to generate a question part of the judgment question and a corresponding correct statement, and optionally supplement other incorrect statements. If the question type is a fill-in-the-blank question, use existing test question initial elements and test question intermediate elements to generate a question part of the fill-in-the-blank question and a corresponding missing part. If the question type is a linking question, use existing test question initial elements and test question intermediate elements to generate a corresponding relationship of the linking question, and optionally supplement related figures.
[0085] As described above, the test question generation method according to the embodiment of the present disclosure controls and adjusts the test question generation process with the answer step as the smallest unit, and can adjust the generation parameters of one or more answer steps to meet the final purpose, and each answer step can accurately correspond to the corresponding knowledge item and answer concept, thereby realizing a more accurate and directional question creation service.
[0086] For example, in one embodiment of the present disclosure, after the test question initial element generating step S202 or the test question intermediate element generating step S203 is performed for the first time, the test questions that are the test question initial element and the test question intermediate element or a combination thereof have a first parameter range interval, and after returning to the test question initial element generating step S202 or the test question intermediate element generating step S203, the test questions that are the test question initial element and the test question intermediate element or a combination thereof that are subsequently generated have a second parameter range interval. The second parameter range interval is smaller than the first parameter range interval. That is, the parameter range is gradually narrowed by iteration.
[0087] For example, take the difficulty parameter range as an example. The difficulty parameter range generated the first time is as follows: a. Step 1: 73-75 b. Step 2: 7-10.5 c. Step 3: 5 to 20.
[0088] The test question initial element generating step S202 or the test question intermediate element generating step S203 is carried out again, that is, the parameter interval is gradually narrowed to obtain the new difficulty parameter range as follows: a. Step 1: 73-74.2 b. Step 2: 8.2 to 9.1 c. Step 3: 6-11.3.
[0089] After multiple rounds of iteration, the final difficulty parameters are obtained. a. Step 1: 73.5 b. Step 2: 8.7 c. Step 3: 8.2.
[0090] It should be noted that in the test question generation process, any related parameters can be transformed into intuitive visualizations. For example, for a particular test question, the related parameters can be shown to the user as follows: Target test difficulty: 60 Knowledge Test Difficulty: 82 Calculation difficulty: 71 Thinking difficulty: 93.
[0091] It is also possible for the user to add relevant comments, for example for one particular exam question. The difficulty level is similar to the second most difficult question in China's National Unified College Entrance Examination, and the test places emphasis on how to think about the answer.
[0092] FIG. 4 is an example of a method for generating test questions according to an embodiment of the present disclosure.
[0093] As shown in Fig. 4, a preset condition receiving step is first performed to receive a preset condition. In the example of Fig. 4, the preset condition is to generate a proof question that tests knowledge items such as the relationship between faces, the relationship between lines, and the relationship between lines in solid geometry, the difficulty of the test is medium, and the expected answer time is 2 minutes.
[0094] Next, based on the preset conditions, a test question initial element is generated from at least one preset database. In the example of Fig. 4, the test question initial element is the last answer step 3 of the reference answer 402 of the test question, that is, "line a intersects with line b, and line b is parallel to surface β", and "line a is parallel to surface β", so "surface α is parallel to surface β". This tests the relationship between surfaces in three dimensions.
[0095] Next, an intermediate element generating step is performed to generate an intermediate element from the test question initial element and the at least one preset database based on the preset condition. In the example of FIG. 4, generating an intermediate element in step S203 is answer step 2 and 1. That is, first, answer step 2 is generated, that is, "line a is parallel to plane β" from "line c is in plane β" and "line a is parallel to line c", which tests the relationship between a line and a surface in three dimensions. At this time, in the test question judgment step, it is judged that the test question composed of the test question initial element and the test question intermediate element does not satisfy the preset condition, and for example, the knowledge item to be tested, which is limited by the preset condition of "relationship between lines in two dimensions", is not tested. In this case, return to the test problem intermediate element generation step and generate again the answer step 1, i.e., "line l, line a, and line c are on the same plane, line l is perpendicular to line a, and line l is perpendicular to line c," and "line a is parallel to line c," which tests the relationship between lines in two dimensions. At this time, the test problem determination step determines whether the test problem composed of the test problem initial element and the test problem intermediate element satisfies the preset condition.
[0096] Next, a test problem generation step is executed to generate the test problem from the test problem initial element and the test problem intermediate element. In the example of Fig. 4, the generated problem 401 is as follows: Knowing that "line l, line a, and line c are in the same plane, line l is perpendicular to line a, and line l is perpendicular to line c", "line b is parallel to plane β", "line c is in plane β", and "line a intersects line b", prove that "plane α is parallel to plane β".
[0097] The above describes the process of generating test questions using the test question generation method according to the embodiment of the present disclosure with reference to Figures 2 to 4. Hereinafter, the process of generating test paper using the test question generation method according to the embodiment of the present disclosure will be described with further reference to Figure 5.
[0098] In step S501, a preset condition receiving step is executed to receive the preset conditions.
[0099] Similar to step S201 shown in FIG. 2, in one embodiment of the present disclosure, a preset condition input by a user is received. The preset condition includes one or more of a question type constraint condition, a test content constraint condition, and a test style constraint condition. For example, the question type constraint condition in the preset condition is for setting one or more of whole, answer question, multiple choice question, judgment question, fill-in-the-blank question, and connecting question to be generated. The test content constraint condition in the preset condition is for setting one or more of a knowledge range to be tested, a difficulty level, an expected completion time, and the like. The test style constraint condition in the preset condition is for setting one or more of a question paper configuration, a test question creation style, and the like.
[0100] Alternatively, the preset conditions may be automatically generated based on the recognized user and its related information (e.g., the user's age, gender, grade, region, problem-solving history record, etc.) to match the user's current learning needs and learning progress.
[0101] In step S502, a test question generation step is performed, and a test question is generated from at least one preset database according to the preset conditions. In one embodiment of the present disclosure, step S502 corresponds to steps S201 to S205 described above with reference to FIG. 2, that is, a single test question is generated using the test question generation method according to the embodiment of the present disclosure. That is, the test question generation step includes an initial element generation substep of generating a test question initial element from at least one preset database based on the preset condition, an intermediate element generation substep of generating a test question intermediate element from the test question initial element and the at least one preset database based on the preset condition, and a test question determination substep of determining whether a test question composed of the test question initial element and the test question intermediate element satisfies the preset condition, and if the test question does not satisfy the preset condition, returning to the test question initial element generation substep or the test question intermediate element generation substep, and if the test question satisfies the preset condition, entering the test question generation substep and generating the test question from the test question initial element and the test question intermediate element, and one of the test question initial element and the test question intermediate element generated by executing the test question intermediate element generation step most recently is set as the answer of the test question, and the other of the two is set as the question of the test question. The other test question intermediate elements except the test question intermediate element generated by executing the test question intermediate element generation substep most recently correspond to the test question answer step. If the test question satisfies the preset condition, generating the test question from the test question initial element and the test question intermediate element further includes generating additional elements based on a question format constraint in the preset condition, and combining the test question initial element, the test question intermediate element and the additional element to generate the test question that satisfies the question format constraint. The test question initial element generating substep includes determining the test items from the knowledge base database, the user history database and / or the user group pattern database.And, the test question intermediate element generating sub-step includes determining an answer concept from the test question initial element and the at least one preset database.
[0102] In step S503, a question paper generation step is performed, and the test question generation step is repeatedly performed multiple times to generate a plurality of test questions, and then the plurality of test questions are organized to generate a question paper.
[0103] In step S504, a question paper determining step is performed to determine whether the question paper satisfies the preset conditions. In one embodiment of the present disclosure, determining whether the question paper satisfies the preset conditions includes: generating global constraint parameters from the preset conditions, the global constraint parameters including one or more of a global question format parameter, a parameter for each test question alone, a parameter related to each test question in the whole, a generation time parameter, and a maximum allowable error parameter; and determining whether the plurality of test questions satisfies the global constraint parameters.
[0104] If a negative result is obtained in step S504, ie the question paper does not satisfy the preset conditions, the process returns to the test question generating step S502 or the question paper generating step S503.
[0105] In one embodiment of the present disclosure, the step returns to the test question generation step S502, and one or more of the test questions are generated again. More specifically, for a specific question, a specific step may be generated again.
[0106] In one embodiment of the present disclosure, returning to the test question generation step S503 includes dynamically adjusting the global constraint parameters, returning to the question paper generation step, and regenerating one or more of the plurality of test questions so that a generation error of one or more of the plurality of test questions is smaller than a predetermined threshold.
[0107] If a positive result is obtained in step S504, ie, the question paper satisfies the preset conditions, the process enters step S505, and outputs the question paper.
[0108] The test question generation method according to the embodiment of the present disclosure described with reference to FIG. 5 also realizes, in the absence of a question bank, providing a user with a directional set of question papers that meet different test formats, with knowledge items related to answer steps as the smallest unit, and can flexibly configure the difficulty, type and number of test questions based on the user's needs.
[0109] The above describes the test question generating method according to the embodiment of the present invention with reference to the drawings. In the following, the test question generating device according to the embodiment of the present disclosure will be described.
[0110] FIG. 6 is a block diagram showing a test question generating apparatus according to an embodiment of the present disclosure. As shown in FIG. 6, a test question generating apparatus 600 according to an embodiment of the present disclosure includes a preset condition receiving unit 601, an initial element generating unit 602, an intermediate element generating unit 603, a test question judging unit 604, and a test question generating unit 605. Each of the above modules can perform the process of generating test questions according to the test question generating method according to the embodiment of the present disclosure described with reference to FIG. 2 to FIG. 4. Those skilled in the art can understand that these unit modules can be realized in various ways by only hardware, only software, or a combination thereof, and the present disclosure is not limited to any of them.
[0111] The preset condition receiving unit 601 is configured to receive preset conditions. In one embodiment of the present disclosure, the preset condition receiving unit 601 receives preset conditions input by a user. The preset conditions include one or more of a question type constraint condition, a test content constraint condition, and a test style constraint condition. For example, the question type constraint condition in the preset conditions is for setting one or more of generating whole, answer questions, multiple choice questions, judgment questions, fill-in-the-blank questions, connecting questions, etc. The test content constraint condition in the preset conditions is for setting one or more of a knowledge range to be tested, difficulty level, expected completion time, etc. The test style constraint condition in the preset conditions is for setting one or more of a question paper configuration, a test question creation style, etc. In another embodiment of the present disclosure, the preset condition receiving unit 601 automatically generates preset conditions that are suitable for the user's current learning needs and learning progress based on the recognized user and its related information (for example, the user's age, gender, grade, region, history record of solving questions, etc.).
[0112] The initial element generating unit 602 is configured to generate test question initial elements from at least one preset database based on the preset conditions. In one embodiment of the present disclosure, the at least one preset database includes, for example, a knowledge base database, a test information base, a user group pattern database, and a user history database. Specifically, the knowledge base database is for storing information related to test target knowledge, including one or more of basic knowledge information, common sense information, answering mindset information, answering method information, and easy-to-make-point information. The test information base is for storing information related to the test, including one or more of question paper configuration information, question format information, difficulty distribution information, test time information, and test style information. The user group pattern database is for storing information related to a user group, and the information related to the user group is obtained from the usage record of the user group, and includes one or more of answer speed information, easy-to-make-knowledge item information, answering habit information, and psychological tolerance limit information of a specific user group. The user history database is for storing information related to a specific user, including one or more of answer history information, knowledge acquisition information, and learning progress information of a specific user. It is easy to understand that the at least one preset database according to the embodiment of the present disclosure is not limited to the exemplary database, but may include any other database. Similarly, the information stored in the at least one preset database is not limited to the exemplary information, but may include any other information. Furthermore, the at least one preset database can be dynamically updated.
[0113] On the one hand, the knowledge base database can be dynamically updated according to the educational outline, teaching materials, and progress of educational research, thereby ensuring that it can meet the latest knowledge acquisition requirements in real time, and the test information database can be dynamically updated according to the latest published test outline and test information, thereby ensuring that it can meet the latest test requirements in real time. In other words, the dynamic update of the database realizes the timeliness and standardity of the test question creation.
[0114] On the other hand, the user group pattern database and the user history database dynamically record the relevant information of specific users and specific user groups in real time, thereby ensuring that directional training can be provided to specific users and specific user groups.
[0115] In one embodiment of the present disclosure, taking an answer question as an example, the test question initial element test question generated by the initial element generating unit 602 from at least one preset database based on the preset conditions is a question of the test question. In another embodiment of the present disclosure, taking an answer question as an example, the test question initial element generated by the initial element generating unit 602 is the last answer step n of the reference answer of the test question. That is, the initial element generating unit 602 according to the embodiment of the present disclosure may generate an answer forward from the generation of the question, or may generate a question backward from the generation of the final answer.
[0116] Specifically, the initial element generating unit 602 determines the test contents, question formats, styles, etc. according to preset conditions, and determines the test items to be tested from the relevant information in the knowledge base database, the user group pattern database and the user history database, and ensures that the generated results meet the actual test criteria from the test information database. In addition, for a first-time user or a user with insufficient relevant records in the user history database, the default group user parameters in the user group pattern database can be used.
[0117] The intermediate element generating unit 603 is configured to generate a test question intermediate element from the test question initial element and the at least one preset database according to the preset condition. In one embodiment of the present disclosure, taking an answer question as an example, after the initial element generating unit 602 generates the question 301 of the test question, the intermediate element generating unit 603 generates a test question intermediate element, which is answer step 1. In another embodiment of the present disclosure, taking an answer question as an example, after the initial element generating unit 602 generates the last answer step n of the reference answer 302 of the test question, the intermediate element generating unit 603 generates a test question intermediate element, which is answer step n-1.
[0118] Specifically, the intermediate element generation unit 603 evaluates what answer concept should be used to construct a test question from the test question initial element (e.g., the generation result of answer step n) using related data from the at least one preset database, so that the relationship between answer step n-1 and answer step n is optimized and satisfies the preset condition.
[0119] The test question judgment unit 604 is configured to judge whether the test question composed of the test question initial element and the test question intermediate element satisfies the preset condition, and if the test question does not satisfy the preset condition, return to generate the test question initial element again by the test question initial element generation unit 602, or generate the test question intermediate element again by the test question intermediate element generation unit 603.
[0120] If the test question satisfies the preset condition, the test question generation unit 605 generates the test question from the test question initial element and the test question intermediate element, and one of the test question initial element and the test question intermediate element generated by the most recent test question intermediate element generation unit is for generating an answer to the test question, and the other of the two is for generating a question of the test question. The other test question intermediate elements except for the test question intermediate element generated by executing the most recent test question intermediate element generation step correspond to the answer step of the test question.
[0121] That is, in one embodiment of the present disclosure, taking the answer question as an example, after the initial element generation unit 602 generates the question of the test question, the test question intermediate element generated most recently by the intermediate element generation unit 603 is the last answer step n of the reference answer. Similarly, in another embodiment of the present disclosure, taking the answer question as an example, after the initial element generation unit 602 generates the last answer step n of the reference answer of the test question, the test question intermediate element generated most recently by the intermediate element generation unit 603 is the question of the test question.
[0122] In addition, if the test question satisfies the preset conditions, the test question generation unit 605 also generates additional elements based on the question format constraints in the preset conditions, and combines the test question initial elements, the test question intermediate elements and the additional elements to generate the test question that satisfies the question format constraints.
[0123] In one embodiment of the present disclosure, the additional element may be relevant explanatory text, for example, the directly generated content is: Completing the Square: x 2 +4x-5=(x+5)(x-1) is used. After generating the related additional elements, the final combination is as follows. Although it does not conform to the perfect square method, it conforms to the conditions for using the method of completing the square because -5=-1×5, 4=5-1. Here, the method of completing the square: x 2 Use +4x-5=(x+5)(x-1).
[0124] In addition, generating additional elements based on the question type constraints in the preset conditions further includes generating and combining additional elements based on different question types. For example, if the question type is a multiple choice question, use existing test question initial elements and test question intermediate elements to generate a question part of the multiple choice question and a corresponding correct option, and optionally supplement other incorrect options. If the question type is a judgment question, use existing test question initial elements and test question intermediate elements to generate a question part of the judgment question and a corresponding correct statement, and optionally supplement other incorrect statements. If the question type is a fill-in-the-blank question, use existing test question initial elements and test question intermediate elements to generate a question part of the fill-in-the-blank question and a corresponding missing part. If the question type is a linking question, use existing test question initial elements and test question intermediate elements to generate a corresponding relationship of the linking question, and optionally supplement related figures.
[0125] As described above, the test question generation device 600 according to an embodiment of the present disclosure controls and adjusts the test question generation process with the answer step as the smallest unit, and can adjust the generation parameters of one or more answer steps to meet the final purpose, and each answer step can accurately correspond to the corresponding knowledge item and answer concept, thereby realizing a more accurate and directional question creation service.
[0126] 7 is a block diagram showing another test question generating device according to an embodiment of the present disclosure. As shown in FIG. 7, the test question generating device 700 according to an embodiment of the present disclosure includes a preset condition receiving unit 701, a test question generating unit 702, a question paper generating unit 703, and a test question judging unit 704. Each of the above modules can respectively perform the process of generating a question paper according to the test question generating method according to the embodiment of the present disclosure described above with reference to FIG. 5. Those skilled in the art can understand that these unit modules can be realized in various ways by only hardware, only software, or a combination thereof, and the present disclosure is not limited to any of them.
[0127] The preset condition receiving unit 701 is configured to receive preset conditions. Similar to the preset condition receiving unit 601 shown in FIG. 6, in one embodiment of the present disclosure, the preset condition receiving unit 701 receives preset conditions input by a user. The preset conditions include one or more of a question type constraint condition, a test content constraint condition, and a test style constraint condition. For example, the question type constraint condition in the preset conditions is for setting one or more of generating whole, answer questions, multiple choice questions, judgment questions, fill-in-the-blank questions, connecting questions, etc. The test content constraint condition in the preset conditions is for setting one or more of a knowledge range to be tested, difficulty level, expected completion time, etc. The test style constraint condition in the preset conditions is for setting one or more of a question paper configuration, a test question creation style, etc. Alternatively, the preset condition receiving unit 701 may automatically generate preset conditions that are suitable for the user's current learning needs and learning progress based on a recognized user and its related information (for example, the user's age, gender, grade, region, history record of solving questions, etc.).
[0128] The test question generating unit 702 is configured to generate test questions from at least one preset database based on the preset conditions. In one embodiment of the present disclosure, the test question generating unit 702 correspondingly performs the process performed by the test question generating device 600 described above with reference to FIG. 6.
[0129] The question paper generating unit 703 is configured to, after the test question generating unit generates a plurality of test questions, organize the plurality of test questions to generate a question paper.
[0130] The question paper determining unit 704 is configured to determine whether the question paper satisfies the preset conditions. In one embodiment of the present disclosure, the question paper determining unit 704 generates global constraint parameters from the preset conditions, the global constraint parameters including one or more of a global question format parameter, an individual test question parameter, a related parameter of each test question in the whole, a generation time parameter, and a maximum allowable error parameter, and determines whether the plurality of test questions satisfy the global constraint parameters.
[0131] If the question paper does not satisfy the preset conditions, the test question generating unit 702 generates new test questions, or the question paper generating unit 703 generates new question paper. If the question paper satisfies the preset conditions, the question paper is output.
[0132] In one embodiment of the present disclosure, the test question generation unit 702 may newly generate one or more of the test questions. More specifically, the test question generation unit 702 may newly generate a specific step for a specific question.
[0133] In one embodiment of the present disclosure, the question paper generation unit 703 dynamically adjusts the global constraint parameters to newly generate one or more of the plurality of test questions so that the generation error of one or more of the plurality of test questions is smaller than a predetermined threshold.
[0134] As described above, the test question generation device 700 according to an embodiment of the present disclosure can provide a user with a directional set of question papers that meet different test formats, with the knowledge items related to the answer steps as the smallest unit, in the absence of a question bank, and can flexibly configure the difficulty, type, and number of test questions based on the user's needs.
[0135] The above describes the process of generating test questions or question papers by the test question generation method and device according to the embodiment of the present disclosure. As generally described above with reference to FIG. 1, the test question generation system using the test question generation method according to the embodiment of the present disclosure can realize updating of relevant information and data through user interaction, and provide more accurate question creation services later. Hereinafter, the test question generation system according to the embodiment of the present disclosure and the method of using the same will be described with reference to FIG. 8 and FIG. 9. FIG. 8 is a flowchart showing the method of using the test question generation system according to the embodiment of the present disclosure, and FIG. 9 is a block diagram showing the test question generation system according to the embodiment of the present disclosure.
[0136] As shown in FIG. 8, a method for using the test question generation system according to an embodiment of the present disclosure includes the following steps.
[0137] In step S801, a preset condition receiving step is executed to receive the preset conditions.
[0138] Similar to step S201 shown in FIG. 2 and step S501 shown in FIG. 5, in one embodiment of the present disclosure, step S801 receives preset conditions input by a user. The preset conditions include one or more of question type constraints, test content constraints, and test style constraints. For example, the question type constraints in the preset conditions are for setting the generation of one or more of whole, answer questions, multiple choice questions, judgment questions, fill-in-the-blank questions, and connecting questions. The test content constraints in the preset conditions are for setting one or more of the knowledge range to be tested, difficulty level, expected completion time, and the like. The test style constraints in the preset conditions are for setting one or more of the structure of the question paper, the style of creating test questions, and the like.
[0139] Alternatively, in step S801, the preset conditions may be automatically generated based on the recognized user and its related information (e.g., the user's age, gender, grade, region, history record of solving problems, etc.), to match the user's current learning needs and learning progress.
[0140] In step S802, a test question and question paper generation step is performed to generate test questions and / or question papers. It is easy to understand that in step S802, the test questions may be generated by executing the test question generation method described with reference to Figs. 2 to 4, or the test question generation method described with reference to Fig. 5 may be executed to generate question papers. It should be noted that the test question generation method executed by the test question generation system according to the embodiment of the present disclosure controls and adjusts the test question generation process with the answer step as the smallest unit, and can adjust the generation parameters of one or more answer steps to meet the final purpose, and each answer step can accurately correspond to the corresponding knowledge item and answer concept. Similarly, the test question generation method executed by the test question generation system according to the embodiment of the present disclosure can provide the user with a set of question papers that meet different test formats with directionality, with the knowledge items related to the answer step as the smallest unit, in the absence of a question bank, and can flexibly configure the difficulty level, type and number of test questions based on the needs of the user.
[0141] In step S803, an answer result receiving step is performed to receive the answer result of the user answering the test questions and / or question paper. In one embodiment of the present disclosure, the answer result receiving unit receives the user's online answer to the test questions or question paper. In another embodiment of the present disclosure, the test questions and / or question paper are provided to the user in a manner that can be used offline, for example, provided to the user in the form of a standard question paper. Then, the answered question paper that the user has provided feedback on is collected, and the corresponding answer result is recognized.
[0142] In step S804, an answer result recording step is performed to record the answer result in units of answer steps. The test question / question paper generation method performed by the test question generation system according to the embodiment of the present disclosure controls and adjusts the test question / question paper generation process with answer steps as the smallest unit, so that the user's answer results are also recorded with answer steps as the smallest unit, thereby making it possible to clarify the user's learning status for each answer step.
[0143] In step S805, a database update step is performed to update the at least one preset database from the answer result. In one embodiment of the present disclosure, a user group pattern database and a user history database are updated from the answer result, which records the mastery of a user or a user group for a specific knowledge item corresponding to an answer step, and helps to provide a more accurate question creation service later.
[0144] 9 is a block diagram showing a test question generation system according to an embodiment of the present disclosure. As shown in FIG. 9, the test question generation system 900 according to an embodiment of the present disclosure includes a preset condition receiving unit 901, a test question, question paper generating unit 902, an answer result receiving unit 903, an answer result recording unit 904, and a database updating unit 905. Those skilled in the art can understand that these unit modules can be realized in various ways by only hardware, only software, or a combination thereof, and the present disclosure is not limited to any of them.
[0145] The preset condition receiving unit 901 is configured to receive a preset condition. Like the preset condition receiving unit 601 shown in FIG. 6 and the preset condition receiving unit 701 shown in FIG. 7, in one embodiment of the present disclosure, the preset condition receiving unit 901 receives a preset condition input by a user. The preset condition includes one or more of a question type constraint condition, a test content constraint condition, and a test style constraint condition. For example, the question type constraint condition in the preset condition is for setting one or more of generating a whole, an answer question, a multiple choice question, a judgment question, a fill-in-the-blank question, a connecting question, etc. The test content constraint condition in the preset condition is for setting one or more of a knowledge range to be tested, a difficulty level, an expected completion time, etc. The test style constraint condition in the preset condition is for setting one or more of a question paper configuration, a test question creation style, etc.
[0146] Alternatively, the preset condition receiving unit 901 automatically generates preset conditions that match the user's current learning needs and learning progress based on the recognized user and its related information (e.g., the user's age, gender, grade, region, history record of solving problems, etc.).
[0147] The test question and question paper generating unit 902 is configured to generate test questions and / or question papers. It is easy to understand that the test question and question paper generating unit 902 may generate test questions by executing the test question generating method described with reference to Fig. 2-4, or may generate question papers by executing the test question generating method described with reference to Fig. 5. It should be noted that the test question generating method performed by the test question and question paper generating unit 902 controls and adjusts the test question generation process with an answer step as the smallest unit, and can adjust the generation parameters of one or more answer steps to meet the final purpose, and each answer step can precisely correspond to the corresponding knowledge item and answer idea. Similarly, the test question and question paper generating method performed by the test question and question paper generating unit 902 can provide the user with a set of question papers that meet different test formats with directionality, with the knowledge items related to the answer step as the smallest unit, in the absence of a question bank, and can flexibly configure the difficulty level, type and number of test questions according to the needs of the user.
[0148] The answer result receiving unit 903 is configured to receive the answer result of the user answering the test questions and / or question paper. In one embodiment of the present disclosure, the answer result receiving unit 903 receives the user's online answer to the test questions or question paper. In another embodiment of the present disclosure, the test questions and / or question paper are provided to the user in a manner that can be used offline, for example, provided to the user in the form of a standard question paper. Then, the answered question paper that the user has fed back is collected, and the answer result receiving unit 903 recognizes the corresponding answer result.
[0149] The answer result recording unit 904 is configured to record the answer result in units of answer steps. Since the test question / question paper generating unit 902 controls and adjusts the test question / question paper generating process with answer steps as the smallest unit, the answer result recording unit 904 correspondingly records the user's answer result with answer steps as the smallest unit as well, so as to clarify the user's mastery of each answer step.
[0150] The database updating unit 905 is configured to update the at least one preset database from the answer result. In one embodiment of the present disclosure, the database updating unit 905 updates a user group pattern database and a user history database from the answer result, so as to record the mastery of a user or a user group for a specific knowledge item corresponding to an answer step, and help the test question generation system 900 to provide a more accurate question creation service later.
[0151] 10 is a block diagram showing hardware of an electronic device 1000 according to an embodiment of the present disclosure. The electronic device according to an embodiment of the present disclosure includes at least a processor and a memory for storing computer program instructions. When the computer program instructions are loaded and executed by the processor, the processor executes the neural network model training method and the image processing method.
[0152] Specifically, the electronic device 1000 shown in FIG. 10 includes a central processing unit (CPU) 1001, a graphics processing unit (GPU) 1002, and a main memory 1003. These units are connected to each other by a bus 1004. The central processing unit (CPU) 1001 and / or the graphics processing unit (GPU) 1002 can be used as the processor, and the main memory 1003 can be used as a memory for storing the computer program instructions. The electronic device 1000 may further include a communication unit 1005, a storage unit 1006, an output unit 1007, an input unit 1008, and an external device 1006, which are also connected to the bus 1004.
[0153] FIG. 11 is a schematic diagram showing a computer-readable storage medium according to an embodiment of the present disclosure. As shown in FIG. 11, a computer-readable storage medium 1100 according to an embodiment of the present disclosure stores computer program instructions 1101. When the computer program instructions 1101 are executed by a processor, the neural network model training method and image processing method according to the embodiment of the present disclosure described with reference to the above drawings are executed. The computer-readable storage medium includes, but is not limited to, a volatile memory and / or a non-volatile memory. The volatile memory includes, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory includes, for example, a read-only memory (ROM), a hard disk, a flash memory, an optical disk, a magnetic disk, etc.
[0154] The above describes the test question generation method, test question generation device, test question generation system, method of using the test question generation system, electronic device, computer-readable storage medium and computer program product according to the embodiments of the present disclosure with reference to the drawings, which realizes that in the absence of a question bank, a single test question or a set of question papers that meet different test formats can be provided to the user in a directional manner, with the knowledge items related to the answer step as the smallest unit, and the difficulty, type and number of test questions can be flexibly configured based on the needs of the user, and also meets the demand for education and evaluation reform to provide test questions that are up-to-date and timely, and realizes that related information and data can be updated based on the user's answer status, so as to provide a more accurate question creation service later.
[0155] Those skilled in the art can imagine that each exemplary unit and algorithm step described in the embodiments disclosed herein may be realized by electronic hardware, or a combination of computer software and electronic hardware. Whether the function is performed by hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art may use different methods to realize the described functions for each specific application, but such realization should not be considered as going beyond the scope of the present invention.
[0156] Although the basic principles of the present disclosure are described above using specific embodiments, it should be noted that the advantages, advantages, effects, etc. mentioned in the present disclosure are merely examples and not limitations, and therefore it cannot be considered that each embodiment of the present disclosure must have these advantages, advantages, effects, etc. In addition, the details disclosed above are merely for illustration and understanding and not limitations, and the details above do not necessarily limit the present disclosure to be realized in such a manner.
[0157] Block diagrams of components, devices, equipment, and systems related to this disclosure are merely examples, and are not intended to assert or imply that they must be connected, arranged, or configured in the manner shown in the blocks. Those skilled in the art will appreciate that these components, devices, equipment, and systems may be connected, arranged, or configured in any manner. For example, terms such as "including" and "comprises" are open-ended terms and may be used interchangeably to refer to "including, but not limited to." As used herein, the words "or" and "and" refer to and may be used interchangeably with the word "and / or," unless the context clearly indicates otherwise. As used herein, the word "for example" refers to and may be used interchangeably with the phrase "for example, but not limited to."
[0158] Also, as used herein, "or" when used in a list beginning with "at least one" refers to the disjunctive list, so that, for example, the list "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Also, the term "exemplary" does not imply that the described example is preferred or better than other examples.
[0159] In the system and method of the present disclosure, each component or each step may be disassembled and / or recombined, and such disassembled and / or recombined components should be considered as equivalent configurations to the present disclosure.
[0160] Various changes, substitutions, and amendments may be made to the techniques described herein without departing from the teachings of the technique as defined by the claims. Moreover, the claims of this disclosure are not limited to the particular aspects of the processes, apparatus, manufacture, arrangements of events, means, methods, and acts described above. Any existing or later developed processes, apparatus, manufacture, arrangements of events, means, methods, or acts that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein may be used. The claims therefore include any such processes, apparatus, manufacture, arrangements of events, means, methods, or acts that are within their scope.
[0161] The above description of the disclosed embodiments is provided to enable one skilled in the art to make or use the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Thus, the present disclosure is not limited to the embodiments shown herein, but is intended to be accorded the widest scope consistent with the principles and novel features disclosed herein.
[0162] The above description has been provided for purposes of illustration and description, and is not intended to limit the embodiments of the present disclosure to those skilled in the art. While several exemplary aspects and embodiments have been discussed above, certain variations, modifications, variations, additions, and subcombinations thereof will occur to those skilled in the art.
Claims
1. A method for generating test questions that does not use a question bank in which past test questions are stored, comprising: a preset condition receiving step of receiving a preset condition; an initial element generating step of generating an initial element of a test question from at least one preset database based on the preset conditions; an intermediate element generating step of generating a test question intermediate element from the test question initial element and the at least one preset database based on the preset condition; determining whether a test question constituted by the test question initial element and the test question intermediate element satisfies the preset condition; If the test question does not satisfy the preset condition, the process returns to the test question initial element generation step or the test question intermediate element generation step; If the test question satisfies the preset condition, a test question generation step is entered, and the test question is generated from the test question initial element and the test question intermediate element. A test question generation method in which one of the test question initial element and the test question intermediate element generated by executing the test question intermediate element generation step most recently is for generating an answer to the test question, and the other of the two is for generating a question for the test question.
2. The test question generating method according to claim 1 , wherein the test question intermediate elements other than the test question intermediate element generated by executing the test question intermediate element generating step most recently correspond to the answer step of the test question.
3. 2. The method of claim 1, wherein the at least one preset database includes a knowledge base database, a test information base, a user group pattern database, and a user history database.
4. 4. The test question generation method according to claim 3, wherein the knowledge base database is for storing information related to the knowledge to be tested, including one or more of basic knowledge information, common sense information, information on how to think about answers, information on methods of answering, and information on areas where mistakes are likely to be made.
5. The test question generation method of claim 3, wherein the test information base is for storing information related to the test, including one or more of the following: question paper configuration information, question format information, difficulty distribution information, test time information, and test style information.
6. The test question generation method of claim 3, wherein the user group pattern database is for storing information related to a user group, and the information related to the user group is obtained from the usage records of the user group and includes one or more of answer speed information, knowledge item information that is likely to be mistaken, answering habit information, and psychological tolerance limit information of a particular user group.
7. The test question generation method of claim 3, wherein the user history database is for storing information related to a particular user, including one or more of the particular user's answer history information, knowledge acquisition information, and learning progress information.
8. 8. The method for generating test questions according to claim 1, wherein the preset conditions include one or more of a question type constraint condition, a test content constraint condition, and a test style constraint condition.
9. generating the test question from the test question initial element and the test question intermediate element if the test question satisfies the preset condition, generating additional elements based on the problem type constraints in the preset conditions; The test question generation method according to any one of claims 1 to 7, further comprising: combining the test question initial element, the test question intermediate element, and the additional element to generate the test question that satisfies the question format constraint condition.
10. The test question initial element generating step includes:
4. The method for generating test questions according to claim 3, further comprising determining test items to be tested from said knowledge base database, said user history database and / or said user group pattern database.
11. The test question intermediate element generating step includes: The method of any one of claims 1 to 7, further comprising determining an answer concept from said test question initial elements and said at least one preset database.
12. a test paper generating step of repeatedly executing the test question initial element generating step, the test question intermediate element generating step, the test question determining step, and the test question generating step to generate a plurality of test questions; determining whether the plurality of test questions satisfies the preset condition; If the plurality of test questions does not satisfy the preset condition, returning to the question paper generation step, and generating one or more of the plurality of test questions anew; A test question generation method as described in any one of claims 1 to 7, further comprising: a question paper determination step in which, if the plurality of test questions satisfy the preset conditions, a total generation step is entered and a question paper is generated from the plurality of test questions.
13. The question paper determination step includes: generating global constraint parameters from the preset conditions, the global constraint parameters including one or more of a global question type parameter, a parameter for each test question alone, a related parameter for each test question in the whole, a generation time parameter, a maximum allowable error parameter, and a response time parameter; and determining whether the plurality of test problems satisfies the global constraint parameters.
14. If the plurality of test questions do not satisfy the preset condition, returning to the question paper generation step and generating one or more of the plurality of test questions again, 14. The method of claim 13, further comprising dynamically adjusting the global constraint parameters and returning to the question paper generating step to generate one or more of the plurality of test questions anew.
15. 15. The method of claim 14, wherein a generation error of one or more of the plurality of test questions is less than a predetermined threshold.
16. The test problem initial element and the test problem intermediate element each have a corresponding first parameter range interval; And the test question initial element and the test question intermediate element generated after returning to the test question initial element generating step each have a corresponding second parameter range interval; 2. The method of claim 1, wherein the second parameter range interval is smaller than the first parameter range interval.
17. The test question generating method of claim 1, wherein the at least one preset database does not include answers to the test questions and questions for the test questions.
18. A test question generation device that does not use a question bank in which past test questions are stored, comprising: a preset condition receiving unit configured to receive a preset condition; an initial element generating unit configured to generate test question initial elements from at least one preset database based on the preset conditions; an intermediate element generating unit configured to generate test question intermediate elements from the test question initial elements and the at least one preset database based on the preset conditions; a test question judgment unit configured to determine whether a test question constituted by the test question initial element and the test question intermediate element satisfies the preset condition; configured to generate the test question from the test question initial element and the test question intermediate element; If the test question does not satisfy the preset condition, return to generate the test question initial element again by the test question initial element generating unit, or generate the test question intermediate element again by the test question intermediate element generating unit; a test question generation unit for generating the test question from the test question initial element and the test question intermediate element by the test question generation unit when the test question satisfies the preset condition; A test question generation device, wherein one of the test question initial element and the test question intermediate element generated by the test question intermediate element generation unit most recently is for generating an answer to the test question, and the other of the two is for generating a question for the test question.
19. The test question generation device of claim 18, wherein the at least one preset database does not include answers to the test questions and questions for the test questions.
20. A method of using a test question generation system, comprising: a preset condition receiving step of receiving a preset condition; a test question and / or question paper generation step of generating test questions and / or question papers; an answer result receiving step of receiving an answer result of a user answering the test questions and / or question paper; an answer result recording step of recording the answer result for each answer step; and a database updating step of updating at least one preset database based on the answer result, The test question and question paper generation step includes: an initial element generating step of generating an initial element of a test question from the at least one preset database based on the preset condition; an intermediate element generating step of generating a test question intermediate element from the test question initial element and the at least one preset database based on the preset condition; determining whether a test question constituted by the test question initial element and the test question intermediate element satisfies the preset condition; If the test question does not satisfy the preset condition, the process returns to the test question initial element generation step or the test question intermediate element generation step; If the test question satisfies the preset condition, a test question generation step is entered, and the test question is generated from the test question initial element and the test question intermediate element. A method of using a test question generation system, in which one of the test question initial element and the test question intermediate element generated by executing the test question intermediate element generation step most recently is for generating an answer to the test question, and the other of the two is for generating a question for the test question.
Citation Information
Patent Citations
Test problem distributing system
JP2006126319A
Technique acquisition support device, technique acquisition support program, and technique acquisition support method
JP2014206619A
Method and apparatus for online generation of problem paths from existing problem collections using knowledge graph
JP2018532140A
Customized question paper generation
US20130084554A1
