A question recommendation method and system based on adaptive total duration of doing questions

Abstract

The application provides a question recommendation method and system based on adaptive total question solving time, which comprises the following steps: obtaining historical answering data, current learning range and total question solving time threshold of a target student, determining the ability level value of the target student according to the historical answering data; generating an initial candidate question set according to the current learning range, and filtering out questions beyond the curriculum and / or repeated questions to obtain a candidate question set; predicting the individualized question solving time of the target student for each candidate question in the candidate question set; calculating the recommendation income of each candidate question, generating a recommended question list based on the recommendation income; pushing the recommended question list to the target student, and updating the subsequent recommendation according to the actual answering time and answering result. The application realizes personalized question sequence recommendation based on total question solving time constraint, dynamically tracks the cognitive load state of the student, effectively controls the learning burden and improves the learning income.

Description

Technical Field

[0001] This invention belongs to the technical field of personalized question recommendation, specifically relating to a question recommendation method and system based on adaptive total question-solving time. Background Technology

[0002] Existing after-school quizzes or online practice systems typically recommend questions based on a uniform number, scope, or difficulty level set by the teacher. While some intelligent question recommendation schemes incorporate student ability values, question difficulty, probability of correct answers, or expected changes in ability, their recommendation goals are mostly focused on selecting a single optimal question or matching questions based on ability values, failing to adequately address the issue of students completing a set of questions within a limited assignment time.

[0003] In scenarios where students' concentration decreases with continuous study time and there are significant differences in problem-solving speed among different students, simply controlling the number of questions cannot reliably control the actual completion time; simply selecting the questions that best match the ability or yield the highest benefit may also lead to the overall recommended question set taking too long, an unbalanced distribution of question types, or an excessive learning burden.

[0004] Compared with existing question recommendation schemes that rely on deep learning to predict correctness, expected changes in ability values, or matching question difficulty, the technical improvements of this scheme are: elevating the total time threshold for answering questions from a general reference parameter to a constraint in the recommendation generation process, and forming adaptive recommendations for question sequences / question packages through a question-level individualized time consumption prediction model, time-budget-based question combination optimization, and a dynamic calibration mechanism after answering.

[0005] Furthermore, existing research rarely considers students' cognitive load during the recommendation process. Cognitive load, as the burden on the learner's cognitive system during learning, has a significant impact on the learner's learning state. As the duration of continuous problem-solving increases, students' cognitive resources are gradually depleted, leading to decreased information processing efficiency and increased error rates, which in turn affects the quality of answers to subsequent questions and learning gains. However, current learner models lack dynamic tracking and utilization of learners' cognitive load.

[0006] On the other hand, existing recommendation algorithms typically do not fully consider the impact of dynamic time intervals between user behavior sequences on recommendation effectiveness. Students' knowledge status and cognitive abilities are constantly changing at different points in time; ignoring the impact of time factors on recommendation results will lead to a decrease in recommendation accuracy. How to effectively integrate time-dimensional information in the recommendation process and achieve dynamic perception and adaptive adjustment of learners' states is a significant challenge currently facing personalized recommendation research. Summary of the Invention

[0007] To address the aforementioned technical issues, this invention proposes a question recommendation method and system based on adaptive total time for answering questions. This method recommends personalized question sequences based on the total time constraint for answering questions, while dynamically tracking the student's cognitive load status, effectively controlling the learning burden and improving learning benefits.

[0008] In a first aspect, the present invention provides the following technical solution: a question recommendation method based on adaptive total question-solving time, comprising: Obtain the target student's historical answer data, current learning scope, and total time threshold for answering questions; determine the target student's ability level value based on the historical answer data. An initial set of candidate questions is generated based on the current learning scope, and questions that are beyond the scope and / or duplicate questions are filtered out to obtain the final set of candidate questions. Based on each candidate question in the candidate question set, predict the individualized time taken by the target student to complete that candidate question; Calculate the recommendation benefit for each candidate question, and generate a list of recommended questions based on the recommendation benefit and under the constraint that the cumulative predicted time spent on answering the candidate questions does not exceed the threshold of the total time spent answering the questions; The recommended question list is pushed to the target student, and subsequent recommendations are updated based on the actual time spent answering and the answer results.

[0009] Compared to existing technologies, the beneficial effects of this invention are as follows: This invention uses a total time threshold for problem-solving as a hard constraint for recommendation generation rather than a post-hoc statistical indicator. This allows for the generation of personalized problem sets while meeting homework time limits, reducing the learning burden on students due to uncontrollable problem quantity or time consumption. Compared to fixed-quantity recommendations, it can adapt to different students' problem-solving speeds, ensuring that students with weaker learning abilities or slower problem-solving speeds are not recommended too many problems. Compared to recommending problems solely based on ability values ​​or single-problem benefits, it can achieve a multi-objective balance within the total time frame, considering knowledge coverage, problem type balance, and learning benefits. By introducing a cognitive load index and a cognitive fatigue penalty term, it achieves dynamic tracking of students' cognitive states. This system automatically adjusts its recommendation strategy based on students' continuous practice time, preventing the push of overly difficult questions when students' cognitive resources are exhausted, thus effectively protecting their learning experience. Its individualized time-consuming prediction model fully considers individual student differences, enabling more accurate adaptation to different students' actual situations. Feedback on actual time consumption is used to calibrate subsequent recommendations, reducing the accumulation of prediction errors and improving the real-time adaptability of the recommendation process. Through filtering for questions beyond the syllabus, duplicate questions, and low-quality questions, it prevents students from practicing unlearned knowledge points or low-value questions within a limited time, ensuring the effectiveness of training within that timeframe. Furthermore, the entire recommendation process relies entirely on learning behavior data, requiring no additional hardware, thus reducing system costs and deployment complexity. This comprehensive approach integrates student profiling, time consumption prediction, benefit assessment, constraint optimization, dynamic calibration, and learning reports, facilitating integration and deployment on online education platforms. It is particularly suitable for smart education scenarios with extremely high requirements for time control and learning efficiency, such as after-school assignments and timed exercises.

[0010] Preferably, the target student's ability level value for: ; In the formula, Indicates target students Ability level value This represents the average difficulty level of the questions that the target students answered correctly. This represents the average difficulty level of the questions that the target students answered incorrectly. This represents the target students' accuracy rate in answering questions within the statistical period. This indicates the average level of knowledge mastery among target students within the current learning scope. This indicates the recent trend of changes in the target students' abilities. to These are the weighting coefficients for the first to fifth capabilities.

[0011] Preferably, the individualized problem-solving time is... The prediction formula is: ; In the formula, Indicates candidate questions, Indicates the target students, Indicates target students Complete the candidate questions The prediction time Indicates candidate questions The difficulty of the questions Indicates candidate questions The question type coding value, Indicate candidate questions The length of the question stem, Indicates candidate questions Historical average response time Indicates target students In candidate questions The average time spent on history questions of this type Indicates target students For candidate questions Related knowledge points The degree of mastery, Indicates target students Ability level value Indicates target students The recent trend in response speed. Indicates target students At any moment Cognitive load index, For bias terms, to These are the prediction weight coefficients for the first to ninth predictions.

[0012] Preferably, the recommended benefits for: ; In the formula, Indicates candidate questions For target students Referral rewards A function representing the fit between the difficulty of the questions and the students' ability levels. This indicates the value of reinforcing weak knowledge points. This indicates the expected change in ability after answering the question. Indicates candidate questions For the current question set Knowledge point coverage contribution This indicates the total time spent answering questions. The cognitive fatigue penalty item below, This indicates a penalty for repetition with recently completed questions. to These are the weighting coefficients for the first to sixth earnings.

[0013] Preferably, the update formula for the cognitive load index is: ; In the formula, For target students At any moment Cognitive load index, This represents the actual time increment for a single question. The cognitive load decay coefficient, This is the load growth factor. and The deadlines are respectively The cumulative number of correct and incorrect answers.

[0014] Preferably, the recommended question list is obtained using the following optimization model: maximize: ; Constraints: ; ; ; ; In the formula, The total number of questions in the candidate question set. As decision variables, Indicates selecting the first The question, Indicates no choice. For the first The question is for the target students Referral rewards For the first The prediction time for the question This is the threshold for the total time required to complete the questions. To allow for a certain duration of fluctuation tolerance, This is the timeout penalty coefficient. To maximize the recommended number of questions, Indicates the first The question types and categories, This is an indicator function; it takes the value 1 when the condition within the parentheses is true, and 0 otherwise. For question type index, The total number of question types. For the first The maximum number of questions of this type allowed.

[0015] Preferably, the step of pushing the recommended question list to the target student and updating subsequent recommendations based on the actual time spent answering and the answer results includes: Obtain the actual time taken by the target student to complete the current question. and the answer result; Update cumulative actual time Update remaining time , This is the total duration threshold; Update the knowledge point mastery level based on the answer results. If the answer is correct, then If the answer is incorrect, ,in This is the learning rate coefficient. The levels of knowledge mastery before and after the update are respectively; when Time-triggered replanning, among which To set a preset deviation threshold, To predict the time consumption; During replanning, based on the updated ability level values, knowledge mastery, and cognitive load index, the remaining time... The internal optimization model is re-solved to generate a new list of recommended questions to replace the incomplete original recommendation sequence; when The problem ends when the time is right, among which This is the minimum effective time threshold for each question.

[0016] Secondly, the present invention provides the following technical solution: a question recommendation system based on adaptive total time for answering questions, the system comprising: The ability module is used to obtain the target student's historical answer data, current learning scope, and total time threshold for answering questions, and to determine the target student's ability level value based on the historical answer data. The processing module is used to generate an initial set of candidate questions based on the current learning scope, and filter out out-of-scope questions and / or duplicate questions to obtain the candidate question set; The prediction module is used to predict the individualized time required for the target student to complete each candidate question based on the candidate question set. The revenue module is used to calculate the recommended revenue for each candidate question, and generate a recommended question list based on the recommended revenue and under the constraint that the cumulative predicted time spent on answering the candidate questions does not exceed the threshold of the total time spent answering the questions. The update module is used to push the recommended question list to the target student and update subsequent recommendations based on the actual answering time and answering results.

[0017] Thirdly, the present invention provides the following technical solution: a computer, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the question recommendation method based on adaptive total time for answering questions as described above.

[0018] Fourthly, the present invention provides the following technical solution: a storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the above-described question recommendation method based on adaptive total time for answering questions. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of the question recommendation method based on adaptive total time for answering questions provided in Embodiment 1 of the present invention; Figure 2 This is a structural block diagram of the question recommendation system based on adaptive total time for answering questions provided in Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of the hardware structure of a computer provided for another embodiment of the present invention.

[0021] The embodiments of the present invention will be further described below with reference to the accompanying drawings. Detailed Implementation

[0022] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain embodiments of the present invention, and should not be construed as limiting the present invention.

[0023] Example 1 In Embodiment 1 of the present invention, as Figure 1 As shown, a question recommendation method based on adaptive total time for answering questions includes: S1. Obtain the target student's historical answer data, current learning scope, and total time threshold for answering questions; determine the target student's ability level value based on the historical answer data. Specifically, obtain the target student's historical answer records, mastery of each knowledge point, accuracy rate for each question type, average time spent on each question type, and recent... The speed trend of each practice session, the current learning scope, the teacher-assigned training objectives, and the total time threshold for completing the exercises. . It can be a fixed value, or it can be automatically generated based on grade level, task type, remaining homework time, student fatigue level, and teacher strategy.

[0024] At the same time, student ability scores are calculated based on the average difficulty of questions answered correctly, the average difficulty of questions answered incorrectly, accuracy rate, knowledge point mastery, and recent performance. : ; In the formula, Indicates target students The ability level value, with a range of values ​​being: ; This represents the average difficulty level of the questions that the target students answered correctly. , This represents the total number of questions answered correctly within the statistical period. For the first The difficulty level of answering the questions correctly; This represents the average difficulty level of the questions that the target students answered incorrectly. , To calculate the total number of questions answered incorrectly within the statistical period. For the first The difficulty level of the questions answered incorrectly; This represents the target students' accuracy rate in answering questions within the statistical period. ; This indicates the average level of knowledge mastery among target students within the current learning scope. , This represents the total number of knowledge points included in the current learning scope. For target students Knowledge points The degree of mastery, ; This indicates the recent trend of changes in the target students' abilities. , This represents the number of practice sessions within the recent sliding window. For the first Ability value after each practice session This represents the average value of the capabilities within the window. to The first to fifth capability weighting coefficients are respectively set to values ​​of , , , , .

[0025] S2. Generate an initial set of candidate questions based on the current learning scope, and filter out out-of-scope questions and / or duplicate questions to obtain the candidate question set; Specifically, an initial set of candidate questions is generated based on the current course, chapter, knowledge points, and the scope specified by the teacher. ; Delete questions containing knowledge points beyond the syllabus based on their relationship to the knowledge points in the questions; Delete duplicate or highly similar questions based on students' recent question-answering records (similarity threshold). ); based on the question quality indicators, remove questions with low discrimination (discrimination). Controversial questions or questions with unusually long prediction times.

[0026] S3. Based on each candidate question in the candidate question set, predict the individualized time required for the target student to complete the candidate question; Specifically, for each candidate question Construct feature vectors and predict target students Individualized time spent on completing this question : ; In the formula, Indicates candidate questions, Indicates the target students; Indicates target students Complete the candidate questions The prediction time is in seconds; Indicates candidate questions The difficulty level of the question, with a range of values. ; Indicates candidate questions The question type coding values ​​are as follows: multiple choice questions are 0.2, true / false questions are 0.1, fill-in-the-blank questions are 0.5, short answer questions are 0.7, problem-solving questions are 0.8, and comprehensive questions are 0.9. Indicates candidate questions The length of the question stem, in characters; Indicates candidate questions The historical average response time, in seconds; Indicates target students In candidate questions The historical average time taken for this question type, in seconds; Indicates target students For candidate questions Related knowledge points The degree of mastery, with a range of values ​​being ; Indicates target students Ability level value; Indicates target students The recent trend in response speed. , This refers to the number of recent speed statistics. For the first Average time spent on each exercise; Indicates target students At any moment The cognitive load index, with a value range of 100%. Initial time ; This is the bias term, and its value is... , to The first to ninth prediction weight coefficients are respectively set to values ​​of , , , , , , , , The above coefficients were obtained by training offline historical data using the least squares method.

[0027] The update formula for the cognitive load index is as follows: ; In the formula, For target students At any moment Cognitive load index, This represents the actual time increment for a single question, in minutes. The cognitive load decay coefficient has a value of [value missing]. ; The load growth factor has a value of [value missing]. ; and The deadlines are respectively The total number of correct and incorrect answers; initial time. ,and .

[0028] S4. Calculate the recommendation benefit for each candidate question, and generate a list of recommended questions based on the recommendation benefit and under the constraint that the cumulative predicted time spent on answering the candidate questions does not exceed the threshold of the total time spent answering the questions. Among them, the recommended benefits for: ; In the formula, Indicates candidate questions For target students Overall referral benefits; A function representing the fit between the difficulty of the questions and the students' ability levels. , This is a bandwidth parameter, with a value of [value to be filled in]. ; This indicates the value of reinforcing weak knowledge points. , For the title A collection of related knowledge points The number of knowledge points For knowledge points Importance weight, default value ; This indicates the expected change in ability after answering the question. ,in ; Indicates candidate questions For the current question set Knowledge point coverage contribution , The set of currently selected topics The set of knowledge points already covered; This indicates the total time spent answering questions. The cognitive fatigue penalty item below, , This is the fatigue accumulation rate coefficient, with a value of [value missing]. ; This indicates a penalty for repetition with recently completed questions. , For target students A collection of questions has been completed recently. This represents the number of questions completed recently. For the title With the topic The similarity is calculated based on the overlap of knowledge points and the consistency of question types, with a value range of... ; to The first to sixth revenue weighting coefficients are respectively set to values ​​of , , , , , .

[0029] The recommended question list is obtained using the following optimization model: maximize: ; Constraints: ; ; ; ; In the formula, The total number of questions in the candidate question set. As decision variables, Indicates selecting the first The question, Indicates no choice. For the first The question is for the target students Referral rewards For the first The prediction time for the question This is the threshold for the total time required to complete the questions. The allowable duration fluctuation tolerance is set at 2.0 minutes. This is the timeout penalty coefficient, with a value of 100. The maximum recommended number of questions is set to 30. Indicates the first The question types and categories, This is an indicator function; it takes the value 1 when the condition within the parentheses is true, and 0 otherwise. For question type index, The total number of question types. For the first The maximum allowed number of question types can be determined using a dynamic programming algorithm in the above model. ) or beam search algorithm ( Please provide a solution.

[0030] S5. Push the recommended question list to the target student, and update subsequent recommendations based on the actual answering time and answering results.

[0031] Step S5 includes: S51. Obtain the actual time taken by the target student to complete the current question. and the answer result; S52, Update cumulative actual time consumed Update remaining time , This is the total duration threshold; S53. Update the knowledge point mastery level based on the answer results. If the answer is correct, then If the answer is incorrect, ,in This is the learning rate coefficient, with a value of 0.15. The levels of knowledge mastery before and after the update are respectively; S54, when Time-triggered replanning, among which The preset deviation threshold is set to 0.30. To predict the time consumption; S55. During replanning, based on the updated ability level, knowledge mastery, and cognitive load index, in the remaining time... The internal optimization model is re-solved to generate a new list of recommended questions to replace the incomplete original recommendation sequence; S56, when The problem ends when the time is right, among which The minimum effective time threshold for each question is 0.5 minutes.

[0032] Specifically, the system stops recommending questions when the cumulative prediction time reaches a threshold, the remaining time is insufficient to recommend effective questions, or the teacher's training objectives have been achieved. The system then outputs a list of recommended questions, estimated completion time, covered knowledge points, training objectives, and a learning report upon completion.

[0033] The question recommendation method based on adaptive total time for answering questions provided in Embodiment 1 of this invention uses a total time threshold as a hard constraint for recommendation generation rather than a post-event statistical indicator. This allows for the generation of personalized question sets while meeting homework time limits, reducing the learning burden on students caused by uncontrollable question quantity or time consumption. Compared to fixed-quantity recommendations, it adapts to different students' problem-solving speeds, ensuring that students with weaker learning abilities or slower problem-solving speeds are not recommended too many questions. Compared to recommending questions solely based on ability values ​​or single-question benefits, it achieves a multi-objective balance within the total time frame, considering knowledge coverage, question type balance, and learning benefits. By introducing a cognitive load index and a cognitive fatigue penalty term, it achieves a better understanding of students' cognitive states. Dynamic tracking automatically adjusts recommendation strategies based on students' continuous practice time, preventing the push of overly difficult questions when students' cognitive resources are exhausted, effectively protecting the learning experience. Individualized time-consuming prediction models fully consider individual student differences, more accurately adapting to different students' actual situations. Feedback from actual time consumption calibrates subsequent recommendations, reducing the accumulation of prediction errors and improving the real-time adaptability of the recommendation process. Over-syllabus filtering, duplicate filtering, and low-quality filtering prevent students from practicing unlearned knowledge points or low-value questions within a limited time, ensuring the effectiveness of training within that timeframe. Furthermore, the entire recommendation process relies entirely on learning behavior data, requiring no additional hardware, reducing system costs and deployment complexity. This comprehensive approach integrates student profiling, time consumption prediction, benefit assessment, constraint optimization, dynamic calibration, and learning reports, facilitating integration and deployment on online education platforms. It is particularly suitable for smart education scenarios with extremely high requirements for time control and learning efficiency, such as homework and timed exercises.

[0034] Example 2 like Figure 2As shown, in Embodiment 2 of the present invention, a question recommendation system based on adaptive total time for answering questions is provided. The system includes: Ability Module 1 is used to obtain the target student's historical answer data, current learning scope, and total time threshold for answering questions, and to determine the target student's ability level value based on the historical answer data; Processing module 2 is used to generate an initial candidate question set based on the current learning scope, and filter out out-of-scope questions and / or duplicate questions to obtain the candidate question set; Prediction module 3 is used to predict the individualized time required for the target student to complete each candidate question based on each candidate question in the candidate question set. The revenue module 4 is used to calculate the recommended revenue for each candidate question, and generate a recommended question list based on the recommended revenue and under the constraint that the cumulative predicted time spent on answering the candidate questions does not exceed the threshold of the total time spent answering the questions. The update module 5 is used to push the recommended question list to the target student and update subsequent recommendations based on the actual answering time and answering results.

[0035] Among them, the target student's ability level value for: ; In the formula, Indicates target students Ability level value This represents the average difficulty level of the questions that the target students answered correctly. This represents the average difficulty level of the questions that the target students answered incorrectly. This represents the target students' accuracy rate in answering questions within the statistical period. This indicates the average level of knowledge mastery among target students within the current learning scope. This indicates the recent trend of changes in the target students' abilities. to These are the weighting coefficients for the first to fifth capabilities.

[0036] Among them, the time spent on individualized problem-solving The prediction formula is: ; In the formula, Indicates candidate questions, Indicates the target students, Indicates target students Complete the candidate questions The prediction time Indicates candidate questions The difficulty of the questions Indicates candidate questions The question type coding value, Indicates candidate questions The length of the question stem, Indicates candidate questions Historical average response time Indicates target students In candidate questions The average time spent on history questions of this type Indicates target students For candidate questions Related knowledge points The degree of mastery, Indicates target students Ability level value Indicates target students The recent trend in response speed. Indicates target students At any moment Cognitive load index, For bias terms, to These are the prediction weight coefficients for the first to ninth predictions.

[0037] Among them, the recommended benefits for: ; In the formula, Indicates candidate questions For target students Referral rewards A function representing the fit between the difficulty of the questions and the students' ability levels. This indicates the value of reinforcing weak knowledge points. This indicates the expected change in ability after answering the question. Indicates candidate questions For the current question set Knowledge point coverage contribution This indicates the total time spent answering questions. The cognitive fatigue penalty item below, This indicates a penalty for repetition with recently completed questions. to These are the weighting coefficients for the first to sixth earnings.

[0038] The update formula for the cognitive load index is as follows: ; In the formula, For target students At any moment Cognitive load index, This represents the actual time increment for a single question. The cognitive load decay coefficient, This is the load growth factor. and The deadlines are respectively The cumulative number of correct and incorrect answers.

[0039] The recommended question list is obtained using the following optimization model: maximize: ; Constraints: ; ; ; ; In the formula, The total number of questions in the candidate question set. As decision variables, Indicates selecting the first The question, Indicates no choice. For the first The question is for the target students Referral rewards For the first The prediction time for the question This is the threshold for the total time required to complete the questions. To allow for a certain duration of fluctuation tolerance, This is the timeout penalty coefficient. To maximize the recommended number of questions, Indicates the first The question types and categories, This is an indicator function; it takes the value 1 when the condition within the parentheses is true, and 0 otherwise. For question type index, The total number of question types. For the first The maximum number of questions of this type allowed.

[0040] The update module 5 is used for: Obtain the actual time taken by the target student to complete the current question. and the answer result; Update cumulative actual time Update remaining time , This is the total duration threshold; Update the knowledge point mastery level based on the answer results. If the answer is correct, then If the answer is incorrect, ,in This is the learning rate coefficient. The levels of knowledge mastery before and after the update are respectively; when Time-triggered replanning, among which To set a preset deviation threshold, To predict the time consumption; During replanning, based on the updated ability level values, knowledge mastery, and cognitive load index, the remaining time... The internal optimization model is re-solved to generate a new list of recommended questions to replace the incomplete original recommendation sequence; when The problem ends when the time is right, among which This is the minimum effective time threshold for each question.

[0041] In other embodiments of the present invention, the present invention provides the following technical solution: a computer, including a memory 102, a processor 101, and a computer program stored in the memory 102 and executable on the processor 101, wherein the processor 101 executes the computer program to implement the question recommendation method based on adaptive total time for answering questions as described above.

[0042] Specifically, the processor 101 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of the present invention.

[0043] The memory 102 may include a large-capacity memory for data or instructions. For example, and not limitingly, the memory 102 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), flash memory, an optical disk drive, a magneto-optical disk drive, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 102 may include removable or non-removable (or fixed) media. Where appropriate, the memory 102 may be internal or external to a data processing device. In a particular embodiment, the memory 102 is non-volatile memory. In a particular embodiment, the memory 102 includes read-only memory (ROM) and random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable read-only memory (PROM), an erasable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), an electrically alterable read-only memory (EAROM), or flash memory, or a combination of two or more of these. Where appropriate, the RAM can be Static Random-Access Memory (SRAM) or Dynamic Random-Access Memory (DRAM). DRAM can be Fast Page Mode Dynamic Random Access Memory (FPMDRAM), Extended Data Out Dynamic Random Access Memory (EDODRAM), Synchronous Dynamic Random-Access Memory (SDRAM), etc.

[0044] The memory 102 can be used to store or cache various data files that need to be processed and / or used for communication, as well as possible computer program instructions executed by the processor 101.

[0045] The processor 101 reads and executes the computer program instructions stored in the memory 102 to implement the above-mentioned question recommendation method based on adaptive total time for answering questions.

[0046] In some embodiments, the computer may further include a communication interface 103 and a bus 100. For example, Figure 3 As shown, the processor 101, memory 102, and communication interface 103 are connected through bus 100 and complete communication with each other.

[0047] The communication interface 103 is used to enable communication between the various modules, devices, units, and / or equipment in the embodiments of the present invention. The communication interface 103 can also enable data communication with other components such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations.

[0048] Bus 100 includes hardware, software, or both, that couples components of a computer device together. Bus 100 includes, but is not limited to, at least one of the following: data bus, address bus, control bus, expansion bus, and local bus. For example, and not as a limitation, bus 100 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 100 may include one or more buses. Although specific buses are described and illustrated in the embodiments of the present invention, the present invention is contemplated by any suitable bus or interconnect.

[0049] The computer can execute the question recommendation method based on adaptive total time for answering questions of this invention based on the question recommendation system based on adaptive total time for answering questions, thereby realizing question recommendation based on adaptive total time for answering questions.

[0050] In some further embodiments of the present invention, in conjunction with the above-described question recommendation method based on adaptive total time for answering questions, the present invention provides the following technical solution: a storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the above-described question recommendation method based on adaptive total time for answering questions.

[0051] Those skilled in the art will understand that the logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can mean any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0052] More specific examples of readable media (a non-exhaustive list) include: electrical connections (electronic devices) with one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0053] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0054] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0055] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A question recommendation method based on adaptive total time for answering questions, characterized in that, include: Obtain the target student's historical answer data, current learning scope, and total time threshold for answering questions; determine the target student's ability level value based on the historical answer data. An initial set of candidate questions is generated based on the current learning scope, and questions that are beyond the scope and / or duplicate questions are filtered out to obtain the final set of candidate questions. Based on each candidate question in the candidate question set, predict the individualized time taken by the target student to complete that candidate question; Calculate the recommendation benefit for each candidate question, and generate a list of recommended questions based on the recommendation benefit and under the constraint that the cumulative predicted time spent on answering the candidate questions does not exceed the threshold of the total time spent answering the questions; The recommended question list is pushed to the target student, and subsequent recommendations are updated based on the actual time spent answering and the answer results.

2. The question recommendation method based on adaptive total time for answering questions according to claim 1, characterized in that, The target student's ability level value for: ； In the formula, Indicates target students Ability level value This represents the average difficulty level of the questions that the target students answered correctly. This represents the average difficulty level of the questions that the target students answered incorrectly. This represents the target students' accuracy rate in answering questions within the statistical period. This indicates the average level of knowledge mastery among target students within the current learning scope. This indicates the recent trend of changes in the target students' abilities. to These are the weighting coefficients for the first to fifth capabilities.

3. The question recommendation method based on adaptive total time for answering questions according to claim 1, characterized in that, The time spent on individualized problem-solving The prediction formula is: ； In the formula, Indicates candidate questions, Indicates the target students, Indicates target students Complete the candidate questions The prediction time Indicates candidate questions The difficulty of the questions Indicates candidate questions The question type coding value, Indicates candidate questions The length of the question stem, Indicates candidate questions Historical average response time Indicates target students In candidate questions The average time spent on history questions of this type Indicates target students For candidate questions Related knowledge points The degree of mastery, Indicates target students Ability level value Indicates target students The recent trend in response speed. Indicates target students At any moment Cognitive load index, For bias terms, to These are the prediction weight coefficients for the first to ninth predictions.

4. The question recommendation method based on adaptive total time for answering questions according to claim 1, characterized in that, The recommended benefits for: ； In the formula, Indicates candidate questions For target students Referral rewards A function representing the fit between the difficulty of the questions and the students' ability levels. This indicates the value of reinforcing weak knowledge points. This indicates the expected change in ability after answering the question. Indicates candidate questions For the current question set Knowledge point coverage contribution This indicates the total time spent answering questions. The cognitive fatigue penalty item below, This indicates a penalty for repetition with recently completed questions. to These are the weighting coefficients for the first to sixth earnings.

5. The question recommendation method based on adaptive total time for answering questions according to claim 3, characterized in that, The update formula for the cognitive load index is: ； In the formula, For target students At any moment Cognitive load index, This represents the actual time increment for a single question. The cognitive load decay coefficient, This is the load growth factor. and The deadlines are respectively The cumulative number of correct and incorrect answers.

6. The question recommendation method based on adaptive total time for answering questions according to claim 1, characterized in that, The recommended question list is obtained using the following optimization model: maximize: ； Constraints: ； ； ； ； In the formula, The total number of questions in the candidate question set. As decision variables, Indicates selecting the first The question, Indicates no choice. For the first The question is for the target students Referral rewards For the first The prediction time for the question This is the threshold for the total time required to complete the questions. To allow for a certain duration of fluctuation tolerance, This is the timeout penalty coefficient. To maximize the recommended number of questions, Indicates the first The question types and categories, This is an indicator function; it takes the value 1 when the condition within the parentheses is true, and 0 otherwise. For question type index, The total number of question types. For the first The maximum number of questions of this type allowed.

7. The question recommendation method based on adaptive total time for answering questions according to claim 1, characterized in that, The steps of pushing the recommended question list to the target student and updating subsequent recommendations based on the actual time spent answering and the answer results include: Obtain the actual time taken by the target student to complete the current question. and the answer result; Update cumulative actual time Update remaining time , This is the total duration threshold; Update the knowledge point mastery level based on the answer results. If the answer is correct, then If the answer is incorrect, ,in This is the learning rate coefficient. The levels of knowledge mastery before and after the update are respectively; when Time-triggered replanning, among which To set a preset deviation threshold, To predict the time consumption; During replanning, based on the updated ability level values, knowledge mastery, and cognitive load index, the remaining time... The internal optimization model is re-solved to generate a new list of recommended questions to replace the incomplete original recommendation sequence; when The problem ends when the time is right, among which This is the minimum effective time threshold for each question.

8. A question recommendation system based on adaptive total time for answering questions, characterized in that, The system includes: The ability module is used to obtain the target student's historical answer data, current learning scope, and total time threshold for answering questions, and to determine the target student's ability level value based on the historical answer data. The processing module is used to generate an initial set of candidate questions based on the current learning scope, and filter out out-of-scope questions and / or duplicate questions to obtain the candidate question set; The prediction module is used to predict the individualized time required for the target student to complete each candidate question based on the candidate question set. The revenue module is used to calculate the recommended revenue for each candidate question, and generate a recommended question list based on the recommended revenue and under the constraint that the cumulative predicted time spent on answering the candidate questions does not exceed the threshold of the total time spent answering the questions. The update module is used to push the recommended question list to the target student and update subsequent recommendations based on the actual answering time and answering results.

9. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the question recommendation method based on adaptive total time for answering questions as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the question recommendation method based on adaptive total time for answering questions as described in any one of claims 1 to 7.

Images