English vocabulary intelligent learning path planning system and method based on reinforcement learning

Abstract

The invention discloses an English vocabulary intelligent learning path planning system and method based on reinforcement learning, and relates to the technical field of education learning, and the system comprises a learning path preliminary planning module, a learning equipment feedback module, a learning effect analysis module and a learning path updating module. According to the English vocabulary intelligent learning path planning system and method based on reinforcement learning, the basic vocabulary test result, the planned learning time and the target of the learner are obtained, and personalized planning of the learning path is achieved in combination with dynamic lexicon matching and learning starting point setting. A reinforcement learning model is adopted to dynamically divide vocabulary tasks, the vocabulary difficulty gradient, the use frequency and a multi-modal exercise mode are fused into a stage task, the problems that in a traditional method, a learning path is rigid, and task allocation and user ability are disjointed are solved, and therefore the learning efficiency is improved while the learning target achievement rate is guaranteed. And the pertinence and the time resource utilization rate of vocabulary practice are obviously improved.

Description

Technical Field

[0001] The present invention relates to the field of educational learning technology, and in particular to an English vocabulary intelligent learning path planning system and method based on reinforcement learning. Background Art

[0002] Driven by the acceleration of globalization and the innovation of educational concepts, English vocabulary learning is undergoing a paradigm shift from mechanical memorization to ability training. Cognitive science research reveals the deep mechanism of human memory and understanding, emphasizing the improvement of vocabulary activation rate through semantic network construction, contextualized synesthesia and distributed practice, rather than simply relying on repeated recitation. At the same time, through a systematic grading system, real context simulation and task-based output design, learners can achieve a natural transformation from knowledge accumulation to communication ability in a dynamic language environment, and ultimately form a sustainable autonomous learning ecology.

[0003] For example, the invention patent with announcement number CN112149994B is an English personal ability tracking learning system based on statistical analysis, including an English personal ability test and analysis module, a learning path planning module, a student ability and course matching module, a tutoring resource module, a student ability tracking module, and a real question verification module; the application is based on the statistical analysis of student test samples, takes English proficiency testing as the starting point, and is guided by path planning for setting learning goals for students. It tracks and evaluates the learning process and progress of students at each step, calculates the students' progress value and the distance from the learning goal, and uses real question verification to continuously cyclically correct the accuracy of the assessed progress value, ability value, and learning path calculation; the system has detailed labeling instructions for courses and tutoring resources, and the data generated during student learning will enter the student's personal database in the system, reflecting the daily learning progress.

[0004] For example, the invention patent with announcement number CN112734142B is a resource learning path planning method and device based on deep learning, which involves artificial intelligence technology, including first collecting the learning data click records of virtual objects in real time and storing them in the virtual object database; then obtaining the target data from the virtual object database according to the screening conditions to form a sample set for model training to obtain a prediction model; finally, if the initial input features of the current course sub-trajectory information are received, they are input into the prediction model for calculation to obtain the output result.

[0005] However, in the process of implementing the technical solution of the present application, the present application discovered that the above technology has at least the following technical problems: the current learning effect of the learners is only reflected by the test results, but the learners may be disturbed by external factors in actual learning, resulting in the test results being unable to accurately reflect the learners' actual learning situation, thereby affecting the dynamic adjustment of the personalized learning path and reducing the learners' learning efficiency. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides an English vocabulary intelligent learning path planning system and method based on reinforcement learning, which can effectively solve the problems involved in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: In a first aspect, the present invention provides an English vocabulary intelligent learning path planning system based on reinforcement learning, including: a learning path preliminary planning module, which is used to obtain learner individual feature information and perform preliminary planning of vocabulary learning paths, wherein the preliminary planning of vocabulary learning paths includes vocabulary library matching, learning starting point setting and learning task division; a learning device feedback module, which is used to collect learner learning device performance data in real time, and quantify the interference degree of the learning device performance data on the learning effect to obtain a learning device interference quantitative index, judge whether the learning device performance will interfere with the learning effect and provide feedback; a learning effect analysis module, which is used to monitor the learner's learning environment, collect learner interaction data in real time, and evaluate the reference value of the learner's vocabulary learning effect detection result in combination with the learning device interference quantitative index; a learning path update module, which is used to obtain the vocabulary learning effect, and judge whether the vocabulary learning task is completed according to the reference value of the vocabulary learning effect detection result, and update the learning path according to the completion of the vocabulary learning task.

[0008] As a further method, the preliminary planning of vocabulary learning path is specifically carried out as follows: the individual characteristic information of the learner includes the basic vocabulary test results of the learner, the set learning time and learning goals; the learning vocabulary is matched according to the learning goals set by the learner; the learning starting point of the learner is set according to the basic vocabulary test results of the learner; the learning task includes each vocabulary to be learned and each vocabulary practice method.

[0009] As a further method, the degree of interference of learning device performance data on learning effect is quantified, and the specific process is: the learning device performance data includes CPU occupancy, memory occupancy and vocabulary refresh response delay; for CPU occupancy and memory occupancy, the maximum theoretical occupancy is set as the benchmark value according to the device hardware specifications, and the actual occupancy is converted into a ratio relative to the benchmark value; for vocabulary refresh response delay, a dynamic benchmark range is set based on the lowest response delay and the highest response delay of the learning device response in historical learning tasks, the vocabulary refresh response delay is mapped to a standardized value within the benchmark range, and weighted processing is performed according to a preset device interference weight coefficient to generate a learning device interference quantification index, and the learning device interference quantification index is used to quantitatively evaluate the degree of interference of learning device performance on learners' learning effects.

[0010] As a further method, the method of judging whether the performance of the learning device will interfere with the learning effect and providing feedback specifically comprises the following steps: obtaining a preset learning device interference quantification index threshold from a vocabulary learning database; comparing the learning device interference quantification index with the learning device interference quantification index threshold; if the learning device interference quantification index is less than the learning device interference quantification index threshold, marking the learning device interference as negligible interference and not including it in the reference value evaluation of the vocabulary learning effect detection result; if the learning device interference quantification index is not less than the learning device interference quantification index threshold, marking the learning device interference as significant interference, and pushing a learning device interference prompt to the learner, and including the learning device interference in the reference value evaluation of the vocabulary learning effect detection result.

[0011] As a further method, the reference value of the vocabulary learning effect detection results of the learners is evaluated, and the specific process is: processing the learner interaction data to obtain the comprehensive interaction anomaly evaluation parameters of each vocabulary of the learners; by integrating the learning environment monitoring data, the learning device interference quantitative index and the comprehensive interaction anomaly evaluation parameters of each vocabulary of the learners, a multidimensional credibility evaluation model of each vocabulary is constructed to obtain the multidimensional credibility evaluation value of each vocabulary, and the multidimensional credibility evaluation value of the vocabulary is used to quantitatively evaluate the reference value of the vocabulary learning effect detection results; the learning environment monitoring data includes the light intensity, noise decibel value and equipment tilt angle at each monitoring time point.

[0012] As a further method, the learner interaction data is processed to obtain the comprehensive interaction anomaly evaluation parameters of each vocabulary of the learner. The specific process is: based on the interaction data of all users in the historical tasks, a group benchmark data set of vocabulary learning time, exercise type time ratio and interaction frequency is established; for the current learner's interaction data, the deviation ratio of each vocabulary learning time and the benchmark time, the deviation amplitude of the time ratio of each exercise type and the benchmark ratio, and the difference coefficient of the interaction frequency of each vocabulary learning process and the benchmark frequency are obtained respectively; the above deviations are aggregated by dynamic weighting to generate the comprehensive interaction anomaly evaluation parameters of each vocabulary of the learner, and the comprehensive interaction anomaly evaluation parameters of each vocabulary of the learner are used to characterize the overall deviation degree of the current learner's interaction behavior from the group standard; the learner interaction data includes the learning time of each vocabulary, the time ratio of each vocabulary and each exercise type, and the interaction frequency of each vocabulary learning process.

[0013] As a further method, the method of judging whether the vocabulary learning task is completed and updating the learning path according to the completion of the vocabulary learning task is as follows: the vocabulary learning effect includes the accuracy rate of each vocabulary exercise; matching each vocabulary multidimensional credibility evaluation value with the vocabulary exercise accuracy rate threshold value corresponding to the multidimensional credibility evaluation value interval of each vocabulary preset in the vocabulary learning database, and comparing each vocabulary exercise accuracy rate with each vocabulary exercise accuracy rate threshold value; if the accuracy rate of a vocabulary exercise is not less than the vocabulary exercise accuracy rate threshold value, marking the vocabulary as a valid learning vocabulary, otherwise, marking the vocabulary as an invalid learning vocabulary, and counting the number of invalid learning vocabulary; according to the number of invalid learning vocabulary and the multidimensional credibility evaluation value of each vocabulary, processing is performed to obtain the completion degree of the vocabulary learning task, and updating the learning path according to the completion degree of the vocabulary learning task.

[0014] As a further method, the learning path is updated according to the completion of the vocabulary learning task, including: comparing the completion of the vocabulary learning task with a vocabulary learning task completion threshold preset in a vocabulary learning database; if the completion of the vocabulary learning task is less than the vocabulary learning task completion threshold, marking the vocabulary learning task as an unfinished learning task, and reinforcing the learning model to obtain the unfinished learning task and re-assign the practice task.

[0015] As a further method, the updating of the learning path according to the completion of the vocabulary learning task also includes: if the completion of the vocabulary learning task is not less than the completion threshold of the vocabulary learning task, the reinforcement learning model assigns review tasks to the vocabulary learning task, obtains invalid learning vocabulary in the vocabulary learning task, and re-assigns practice tasks for the invalid learning vocabulary.

[0016] The second aspect of the present invention provides an English vocabulary intelligent learning path planning method based on reinforcement learning, including: S1. Acquire learner individual feature information and perform preliminary planning of vocabulary learning path, wherein the preliminary planning of vocabulary learning path includes vocabulary library matching, learning starting point setting and learning task division; S2. Collect learner learning device performance data in real time, and quantify the interference degree of learning device performance data on learning effect to obtain learning device interference quantitative index, judge whether learning device performance will interfere with learning effect and provide feedback; S3. Monitor learner learning environment, collect learner interaction data in real time, and evaluate the reference value of learner vocabulary learning effect detection result in combination with learning device interference quantitative index; S4. Acquire vocabulary learning effect, and judge whether vocabulary learning task is completed according to the reference value of vocabulary learning effect detection result, and update learning path according to the completion of vocabulary learning task.

[0017] Compared with the prior art, the embodiments of the present invention have at least the following beneficial effects:

[0018] (1) The present invention provides an English vocabulary intelligent learning path planning system and method based on reinforcement learning, obtains learners' basic vocabulary test results, planned learning time and goals, and combines dynamic vocabulary matching with learning starting point setting to achieve personalized learning path planning. The reinforcement learning model is used to dynamically divide vocabulary tasks, and the vocabulary difficulty gradient, usage frequency and multimodal practice methods are integrated into staged tasks, which solves the problems of rigid learning paths and disconnection between task allocation and user capabilities in traditional methods, thereby significantly improving the pertinence of vocabulary practice and the utilization of time resources while ensuring the achievement rate of learning goals.

[0019] (2) The present invention analyzes the interference of learning devices and compares it with preset thresholds to trigger a hierarchical response strategy, thereby filtering low-reliability learning effect data caused by device performance degradation at the data source, thereby achieving accurate perception and dynamic compensation of device interference, and helping to maintain the consistency and stability of learning path planning in complex hardware environments. In particular, in scenarios with high concurrency tasks or low-performance devices, it can help reduce the misjudgment rate caused by device freezes and improve the reliability and anti-interference fault tolerance of personalized learning path updates.

[0020] (3) The present invention constructs a multi-dimensional credibility evaluation model by integrating learning environment monitoring data, learning device interference quantitative indicators, and learners' comprehensive interaction anomaly evaluation parameters of each vocabulary, thus solving the limitation of traditional solutions that rely solely on user behavior or device indicators. At the same time, it improves the accuracy of determining the reference value of learning effects, especially in complex scenarios where device and environmental interference is not directly reflected in the user's explicit behavior, ensuring the robustness and anti-interference of dynamic updates of learning paths.

[0021] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of system module connection of the present invention.

[0023] Figure 2 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0026] Reference Figure 1 As shown, the first aspect of the present invention provides an English vocabulary intelligent learning path planning system based on reinforcement learning, including: a learning path preliminary planning module, which is used to obtain learner individual feature information and perform preliminary planning of vocabulary learning paths, wherein the preliminary planning of vocabulary learning paths includes vocabulary library matching, learning starting point setting and learning task division; a learning device feedback module, which is used to collect learner learning device performance data in real time, and quantify the interference degree of learning device performance data on learning effects, obtain learning device interference quantitative indicators, judge whether learning device performance will interfere with learning effects and provide feedback; a learning effect analysis module, which is used to monitor learner learning environment, collect learner interaction data in real time, and evaluate the reference value of learner vocabulary learning effect detection results in combination with learning device interference quantitative indicators; a learning path update module, which is used to obtain vocabulary learning effects, and judge whether vocabulary learning tasks are completed according to the reference value of vocabulary learning effect detection results, and update the learning path according to the completion of vocabulary learning tasks. A vocabulary learning database, which is used to store data in the vocabulary learning process.

[0027] Specifically, a preliminary planning of vocabulary learning path is carried out, and the specific process is as follows: the individual characteristic information of the learner includes the learner's basic vocabulary test results, the set learning time and learning goals.

[0028] The results of the learner's basic vocabulary test include vocabulary size, vocabulary difficulty distribution and common error types. The learner can be tested for vocabulary in advance and automatically obtained through the vocabulary test system. The learning time refers to the time period that the learner plans to use for vocabulary learning, and the learning goals include the vocabulary size and mastery level that the learner expects to achieve.

[0029] The learning vocabulary is obtained by matching the learning objectives set by the learners.

[0030] It needs to be explained that a learning vocabulary bank refers to a collection of vocabulary covering different topics and different levels of difficulty. The learning vocabulary bank can be automatically matched based on the learner's basic vocabulary test results and learning goals, or it can be manually selected by the learner.

[0031] Set the learner's learning starting point according to the learner's basic vocabulary test results.

[0032] The learning starting point refers to the starting point of the personalized learning path determined according to the learner's current vocabulary mastery level. The learner's vocabulary and vocabulary difficulty distribution are obtained through basic vocabulary tests. Based on the vocabulary size and difficulty distribution, a vocabulary set for the initial learning task that matches the learner's basic vocabulary test results is selected from the preset learning vocabulary library.

[0033] The learning time set by the learner includes multiple learning time periods. The vocabulary to be mastered in each learning time period is marked as each learning task. For a single vocabulary, multiple exercises may be performed in the same learning task, and each exercise corresponds to a vocabulary practice method, such as listening, speaking, reading, etc. Learning task division refers to the allocation of vocabulary to be learned and different vocabulary practice methods in each learning time period, and the allocation of learning tasks is performed based on the reinforcement learning model. For example, according to factors such as the difficulty of the vocabulary, the frequency of use, and the correlation between the vocabulary, the vocabulary is associated with the same learning task, and then the corresponding learning tasks are assigned in sequence according to the user's current learning stage and goals.

[0034] The reinforcement learning model is a vocabulary learning method based on reinforcement learning theory, which regards vocabulary learning as a process in which an intelligent agent interacts with vocabulary learning tasks and related contexts. Reinforcement learning (RL) has been widely used in the field of artificial intelligence, involving robot control, game intelligence, etc. Common reinforcement learning algorithms include Q learning, deep Q network, policy gradient method and actor-critic method. The intelligent agent completes the vocabulary learning task by taking different vocabulary practice methods, and adjusts the vocabulary practice method in time according to the feedback of the vocabulary learning task, so as to optimize the vocabulary learning effect.

[0035] The learning task includes various words to be learned and various vocabulary practice methods.

[0036] In a specific embodiment, the present invention realizes personalized planning of learning paths by obtaining the learner's basic vocabulary test results, planned learning time and goals, combined with dynamic vocabulary matching and learning starting point setting. Based on the learner's vocabulary, difficulty distribution and error type characteristics, the initial learning vocabulary matching its current level is adaptively selected, and according to the preset learning goals and time constraints, the vocabulary tasks are dynamically divided using a reinforcement learning model, and the vocabulary difficulty gradient, frequency of use and multimodal practice methods are integrated into the staged tasks, solving the problems of rigid learning paths and disconnection between task allocation and user capabilities in traditional methods, thereby significantly improving the pertinence of vocabulary exercises and the utilization of time resources while ensuring the achievement rate of learning goals.

[0037] It should be understood that the present invention evaluates the learning effect of vocabulary and the reference value of the learning effect detection result based on the multiple vocabulary practice data, and the multiple vocabulary practices are not necessarily continuous in the monitoring time period.

[0038] Specifically, the interference degree of the learning device performance data on the learning effect is quantified, and the specific process is: the learning device performance data includes CPU occupancy rate, memory occupancy rate and vocabulary refresh response delay.

[0039] In this embodiment, the CPU occupancy rate, memory occupancy rate and vocabulary refresh response delay can be directly obtained through the corresponding interface or tool provided by the device operating system. The vocabulary refresh response delay refers to the average time it takes for the device to respond to vocabulary display when the learner interacts with the device.

[0040] For CPU occupancy and memory occupancy, the maximum theoretical occupancy is set as the benchmark value according to the device hardware specifications, and the actual occupancy is converted into a ratio relative to the benchmark value. For vocabulary refresh response delay, a dynamic benchmark range is set based on the lowest response delay and the highest response delay of the learning device in historical learning tasks. The vocabulary refresh response delay is mapped to a standardized value within the benchmark range, and weighted processing is performed according to a preset device interference weight coefficient to generate a learning device interference quantification index, which is used to quantitatively evaluate the degree of interference of the learning device performance on the learner's learning effect.

[0041] In a specific embodiment, the method for obtaining the quantitative index of interference of the learning device is:

[0042] ;

[0043] ;

[0044] In the formula, represents the quantitative index of learning device interference, Indicates the device interference weight coefficient corresponding to the preset CPU usage. Indicates the device interference weight coefficient corresponding to the preset memory usage. Indicates the device interference weight coefficient corresponding to the preset vocabulary refresh response delay, Indicates the current CPU usage. Indicates the maximum CPU usage of the device. Indicates the current memory usage. Indicates the maximum memory usage of the device. A normalized value representing the latency of the lexical refresh response, represents the vocabulary refresh response delay, YCmin is the lowest response delay in the historical task, and YCmax is the highest response delay in the historical task. , and The value range of is between 0 and 1, reflecting the importance of different types of performance data on the interference of learning effects. The device interference weight coefficient can be obtained according to the specific device type. For example, a mapping relationship is established between the device type and the interference weight coefficient of each device and stored in the vocabulary learning database. The weight coefficient suitable for the learning device is selected from the vocabulary learning database through the mapping relationship.

[0045] The quantitative index of learning device interference is obtained by processing CPU occupancy, memory occupancy and vocabulary refresh response delay. The greater the CPU occupancy, memory occupancy and vocabulary refresh response delay, the greater the degree of interference of the learning device on the learner's learning effect. CPU occupancy, memory occupancy and vocabulary refresh response delay affect each other. High CPU occupancy will lead to scheduling delays of vocabulary refresh instructions, insufficient memory resources will aggravate vocabulary loading jams, and the cumulative effect of refresh delays will further reversely push up CPU and memory resource consumption, forming a positive feedback loop of device performance degradation.

[0046] The present invention analyzes CPU occupancy, memory occupancy and vocabulary refresh response delay, maps the three to a unified interference assessment dimension, and combines preset weight coefficients to quantify the comprehensive interference degree of device performance on the learning process, thereby breaking through the limitations of traditional single indicator evaluation. This solution can adapt to the learning scenarios of devices with different hardware configurations, while ensuring the accuracy of learning path planning, significantly reducing the interference tolerance threshold of device performance fluctuations on learning effect evaluation, and providing a data basis for the reliable update of subsequent learning paths.

[0047] Specifically, it is determined whether the performance of the learning device will interfere with the learning effect and feedback is given. The specific process is: obtaining a preset learning device interference quantitative index threshold from the vocabulary learning database; comparing the learning device interference quantitative index with the learning device interference quantitative index threshold. If the learning device interference quantitative index is less than the learning device interference quantitative index threshold, the learning device interference is marked as negligible interference and is not included in the reference value evaluation of the vocabulary learning effect detection results.

[0048] If the learning device interference quantitative index is not less than the learning device interference quantitative index threshold, the learning device interference will be marked as significant interference, and a learning device interference prompt will be pushed to the learner, and the learning device interference will be included in the reference value evaluation of the vocabulary learning effect detection results.

[0049] In a specific embodiment, by analyzing the interference of the learning device and comparing it with the preset threshold, a hierarchical response strategy is triggered, and low-reliability learning effect data caused by device performance degradation is filtered at the data source, thereby achieving accurate perception and dynamic compensation of device interference, which helps to maintain the consistency and stability of learning path planning in complex hardware environments, especially in high-concurrency tasks or low-performance device scenarios. It can help reduce the misjudgment rate caused by device freezes and improve the reliability and anti-interference fault tolerance of personalized learning path updates.

[0050] Specifically, the reference value of the test results of the learners' vocabulary learning effect is evaluated, and the specific process is: the learners' interaction data is processed to obtain the comprehensive interaction anomaly evaluation parameters of each vocabulary of the learners.

[0051] By integrating learning environment monitoring data, quantitative indicators of learning device interference and comprehensive interaction anomaly assessment parameters of learners' vocabularies, a multidimensional credibility assessment model for each vocabulary is constructed to obtain a multidimensional credibility assessment value for each vocabulary. The multidimensional credibility assessment value for each vocabulary is used to quantitatively assess the reference value of vocabulary learning effect detection results. The learning environment monitoring data includes light intensity, noise decibel value and device tilt angle at each monitoring time point.

[0052] The light intensity is collected in real time by an illuminance meter, the noise decibel value can be measured by a sound level meter, and the device tilt angle is continuously tracked and recorded by a built-in gyroscope. The device tilt angle refers to the degree of inclination of the learning device relative to the standard horizontal plane during use, that is, the angle between the plane where the learning device screen is located and the horizontal plane.

[0053] In a specific embodiment, the calculation formula of the vocabulary multi-dimensional credibility evaluation model is:

[0054] ;

[0055] In the formula, represents the multidimensional credibility evaluation value of the i-th word, Indicates the credibility weight coefficient corresponding to the preset learning device interference quantitative index, Indicates the credibility weight coefficient corresponding to the preset comprehensive interaction anomaly assessment parameter, Indicates the credibility weight coefficient corresponding to the preset environmental factors, represents the quantitative index of learning device interference, represents the comprehensive interaction anomaly evaluation parameter of the i-th word, represents the light intensity of the i-th word at the t-th monitoring time point, Indicates the reference to the appropriate light intensity. Indicates the allowable deviation value of light intensity. represents the noise decibel value of the i-th word at the t-th monitoring time point, represents the critical maximum noise decibel value, represents the device tilt angle of the i-th word at the t-th monitoring time point, represents the preset learning device interference quantitative index threshold, i represents the number of each word, i=1,2,3,...,n, n represents the total number of words, t represents the number of each monitoring time point, t=1,2,3,...,s, s represents the total number of monitoring time points.

[0056] The reference suitable light intensity and the critical maximum noise decibel value are preset in the vocabulary learning database. The reference suitable light intensity refers to the optimal light intensity for vocabulary learning under an ideal environment, and the critical maximum noise decibel value is the maximum noise threshold to ensure that the learning effect is not disturbed. , and The value range is between 0 and 1, and is dynamically adjusted according to the actual learning environment and user behavior data. For example, a mapping relationship is established between the comprehensive interaction anomaly assessment parameter type and each credibility weight coefficient, and stored in a vocabulary learning database, and a credibility weight coefficient is selected from the vocabulary learning database through the mapping relationship.

[0057] The multidimensional credibility evaluation value of vocabulary is calculated by weighted calculation of factors such as the quantitative index of learning device interference, the comprehensive interaction anomaly evaluation parameter, the light intensity deviation value, the noise decibel value and the device tilt angle. The larger the quantitative index of learning device interference, the comprehensive interaction anomaly evaluation parameter, the light intensity deviation value and the noise decibel value, or the greater the fluctuation of the device tilt angle, the lower the multidimensional credibility evaluation value of the vocabulary, indicating that the learning effect of the vocabulary during practice testing is worse.

[0058] The parameters in the multidimensional credibility evaluation value of vocabulary are interrelated and affect each other. For example, device performance and environmental factors will directly affect the learner's interactive behavior pattern. The present invention constructs a multidimensional credibility evaluation model by integrating learning environment monitoring data, quantitative indicators of learning device interference, and comprehensive interaction anomaly evaluation parameters of each vocabulary of learners, which solves the limitations of traditional solutions that rely solely on user behavior or device indicators. Users may maintain a pseudo-normal state due to behavioral inertia. Through nonlinear correlation modeling of multi-source heterogeneous data, the combined effects of device performance degradation, environmental interference, and user behavior deviations are incorporated into a unified evaluation framework, which improves the accuracy of determining the reference value of learning effects, especially in complex scenarios where device and environmental interference is not directly reflected in the user's explicit behavior, ensuring the robustness and anti-interference of dynamic updates of learning paths.

[0059] Specifically, the learner interaction data is processed to obtain the comprehensive interaction anomaly evaluation parameters of each vocabulary of the learner. The specific process is: based on the interaction data of all users in the historical tasks, a group benchmark data set of vocabulary learning time, time ratio of exercise type and interaction frequency is established, and the benchmark learning time of each vocabulary, the benchmark time ratio of each exercise type and the benchmark interaction frequency of each vocabulary learning are obtained by taking the average of the historical data.

[0060] It should be understood that the exercise type time ratio refers to the proportion of different exercise types in the total learning time, and the vocabulary learning interaction frequency refers to the frequency of interaction between learners and devices during a specific vocabulary learning process.

[0061] The interaction data of the current learner are processed to obtain the deviation ratio of each vocabulary learning time and the benchmark time, the deviation amplitude of each exercise type time ratio and the benchmark ratio, and the difference coefficient of the interaction frequency and the benchmark frequency of each vocabulary learning process; the above deviations are aggregated using dynamic weighting to generate comprehensive interaction anomaly evaluation parameters of each vocabulary of the learner, and the comprehensive interaction anomaly evaluation parameters of each vocabulary of the learner are used to characterize the overall deviation degree of the current learner's interaction behavior from the group standard.

[0062] In a specific embodiment, the comprehensive interaction anomaly evaluation parameter of each vocabulary of the learner is obtained as follows:

[0063] ;

[0064] In the formula, represents the comprehensive interaction anomaly evaluation parameter of the i-th word, Indicates the interaction anomaly evaluation weight coefficient corresponding to the preset vocabulary learning time. Indicates the interaction anomaly evaluation weight coefficient corresponding to the preset exercise type time ratio, Indicates the interaction anomaly assessment weight coefficient corresponding to the preset interaction frequency, represents the learning time of the i-th vocabulary of the current learner, represents the learning benchmark duration of the i-th word, Indicates the allowed deviation value of the learning time of the i-th vocabulary, It represents the time consumption ratio of the rth exercise type of the i-th vocabulary. represents the benchmark ratio of time spent on the rth exercise type of the i-th vocabulary, It represents the allowable deviation value of the time consumption ratio of the rth exercise type of the i-th vocabulary. represents the interaction frequency of the i-th vocabulary learning, represents the interaction benchmark frequency of the ith vocabulary learning, represents the allowable deviation value of the interaction frequency of the i-th vocabulary learning, i represents the number of each vocabulary, i=1,2,3,...,n, n represents the total number of vocabulary, r represents the number of each exercise type, r=1,2,3,...,h, h represents the total number of exercise types.

[0065] , and The value range of is between 0 and 1, and a mapping relationship between different words and each interaction anomaly evaluation weight coefficient can be established according to the learning experiment of different words. For example, a mapping relationship is established between each word and each interaction anomaly evaluation weight coefficient and stored in a vocabulary learning database, and the interaction anomaly evaluation weight coefficient is selected from the vocabulary learning database through the mapping relationship.

[0066] The comprehensive interaction anomaly assessment parameter is obtained by jointly processing the vocabulary learning time, the time ratio of exercise type and the vocabulary learning interaction frequency. The greater the deviation of the vocabulary learning time, the time ratio of exercise type and the vocabulary learning interaction frequency from the corresponding benchmark values, the larger the comprehensive interaction anomaly assessment parameter is, indicating that the learner's interactive behavior during vocabulary learning is more abnormal.

[0067] The present invention constructs multi-dimensional abnormal evaluation parameters by dynamically associating the deviation characteristics of vocabulary learning time, exercise type time ratio and interaction frequency, and accurately captures the hidden abnormal patterns in user behavior. Vocabulary learning time, exercise type time ratio and interaction frequency are interrelated, which can comprehensively reflect the abnormal behavior of users. A benchmark is established based on historical group data, and the current user's vocabulary learning time deviation, exercise type distribution offset and interaction frequency anomaly are non-linearly aggregated. It helps to distinguish between the increase in passive interaction frequency caused by equipment performance fluctuations and active learning behavior, reduce the misjudgment rate, and provide a high-fidelity behavior analysis basis for learning path updates.

[0068] The learner interaction data includes the learning time of each vocabulary, the time consumption ratio of each vocabulary and each exercise type, and the interaction frequency of each vocabulary learning process.

[0069] Specifically, it is determined whether the vocabulary learning task is completed, and the learning path is updated according to the completion of the vocabulary learning task. The specific process is: the vocabulary learning effect includes the accuracy of each vocabulary exercise; each vocabulary multidimensional credibility evaluation value is matched with the vocabulary exercise accuracy threshold corresponding to the multidimensional credibility evaluation value interval of each vocabulary preset in the vocabulary learning database, and each vocabulary exercise accuracy is compared with each vocabulary exercise accuracy threshold.

[0070] If the correct rate of a vocabulary exercise is not less than the correct rate threshold of the vocabulary exercise, the vocabulary is marked as a valid learning vocabulary. Otherwise, the vocabulary is marked as an invalid learning vocabulary, and the number of invalid learning vocabulary is counted.

[0071] According to the number of invalid learning words and the multi-dimensional credibility evaluation value of each word, the completion degree of the vocabulary learning task is obtained, and the learning path is updated according to the completion degree of the vocabulary learning task.

[0072] In a specific embodiment, the degree of completion of the vocabulary learning task is obtained in the following manner:

[0073] ;

[0074] In the formula, W represents the completion of vocabulary learning tasks, e represents a natural constant, Indicates the task completion weight coefficient corresponding to the preset number of invalid learning words, represents the task completion weight coefficient corresponding to the preset vocabulary multidimensional credibility evaluation value, N represents the number of invalid learning vocabulary, represents the multi-dimensional credibility evaluation value of the i-th word, i represents the number of each word, i=1,2,3,...,n, and n represents the total number of words.

[0075] and The value range of is between 0 and 1, and can be obtained according to the number of words and the difficulty of words in the learning task. For example, the number of words and the difficulty of words are quantified, and a mapping relationship is established between the quantification results and the weight coefficient of each task completion, which is stored in the vocabulary learning database. The task completion weight coefficient is selected from the vocabulary learning database through the mapping relationship.

[0076] In this embodiment, the vocabulary multidimensional credibility evaluation value and the number of invalid learning vocabulary affect each other. The higher the vocabulary multidimensional credibility evaluation value, the lower the interference degree of the learning task, and the number of invalid learning vocabulary. The present invention breaks through the traditional single accuracy threshold judgment method by integrating the composite calculation model of the number of invalid learning vocabulary and the multidimensional credibility evaluation value of each vocabulary, and realizes the accurate quantification of the task completion quality. At the same time, it helps to reduce the task completion error rate and provide an adaptive optimization basis for the dynamic learning path.

[0077] Specifically, the learning path is updated according to the completion of the vocabulary learning task, including: comparing the completion of the vocabulary learning task with a vocabulary learning task completion threshold preset in the vocabulary learning database; if the completion of the vocabulary learning task is less than the vocabulary learning task completion threshold, marking the vocabulary learning task as an unfinished learning task, and reinforcing the learning model to obtain the unfinished learning task and re-assign the practice task.

[0078] In a specific embodiment, the present invention analyzes the vocabulary learning task as a whole, breaks through the limitation of traditional solutions that only provide feedback on the effect of a single vocabulary learning, and integrates the systematic impact of environmental interference factors on the learning process from a task-level global perspective. When environmental interference causes the overall effectiveness of the learning task to decline, cluster analysis based on the multidimensional credibility evaluation value of the vocabulary can be used to identify the cumulative damage to attention and cognitive load caused by the environment, rather than isolating it from the lack of individual vocabulary mastery. At the same time, analyzing the vocabulary learning task as a whole helps to identify ineffective task learning caused by environmental or equipment interference, and to update the learning path in a timely manner.

[0079] Specifically, updating the learning path according to the completion of the vocabulary learning task also includes: if the completion of the vocabulary learning task is not less than the completion threshold of the vocabulary learning task, the reinforcement learning model assigns review tasks to the vocabulary learning task, obtains invalid learning vocabulary in the vocabulary learning task, and re-assigns practice tasks for the invalid learning vocabulary.

[0080] Reference Figure 2 As shown, the second aspect of the present invention provides an English vocabulary intelligent learning path planning method based on reinforcement learning, including: S1. Acquire learner individual feature information and perform preliminary planning of vocabulary learning path, wherein the preliminary planning of vocabulary learning path includes vocabulary library matching, learning starting point setting and learning task division; S2. Collect learner learning device performance data in real time, and quantify the interference degree of learning device performance data on learning effect to obtain learning device interference quantitative index, judge whether learning device performance will interfere with learning effect and provide feedback; S3. Monitor learner learning environment, collect learner interaction data in real time, and evaluate the reference value of learner vocabulary learning effect detection result in combination with learning device interference quantitative index; S4. Acquire vocabulary learning effect, and judge whether vocabulary learning task is completed according to the reference value of vocabulary learning effect detection result, and update learning path according to the completion of vocabulary learning task.

[0081] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that technicians in the relevant technical field can understand and use the present invention well. As long as they do not deviate from the structure of the present invention or exceed the scope defined by the present invention, they should all belong to the protection scope of the present invention.

Claims

1. An English vocabulary intelligent learning path planning system based on reinforcement learning, characterized by: include: A learning path preliminary planning module is used to obtain learner individual characteristic information and perform preliminary planning of vocabulary learning paths, wherein the preliminary planning of vocabulary learning paths includes vocabulary matching, setting of learning starting points, and division of learning tasks; The learning device feedback module is used to collect the learner's learning device performance data in real time, and quantify the degree of interference of the learning device performance data on the learning effect to obtain the learning device interference quantitative index, judge whether the learning device performance will interfere with the learning effect and provide feedback; The learning effect analysis module is used to monitor the learner's learning environment and collect learner interaction data in real time, and evaluate the reference value of the learner's vocabulary learning effect test results in combination with the quantitative indicators of learning device interference; The learning path update module is used to obtain the vocabulary learning effect, and judge whether the vocabulary learning task is completed according to the reference value of the vocabulary learning effect detection result, and update the learning path according to the completion status of the vocabulary learning task.

2. The English vocabulary intelligent learning path planning system based on reinforcement learning according to claim 1 is characterized by: The specific process of preliminary planning of vocabulary learning path is as follows: The learner's individual characteristic information includes the learner's basic vocabulary test results, set learning time and learning goals; The learning vocabulary is obtained by matching the learning goals set by the learners; Set the learner's learning starting point according to the learner's basic vocabulary test results; The learning task includes various words to be learned and various vocabulary practice methods.

3. The English vocabulary intelligent learning path planning system based on reinforcement learning according to claim 1 is characterized by: The interference degree of the learning device performance data on the learning effect is quantified, and the specific process is as follows: The learning device performance data includes CPU occupancy, memory occupancy and vocabulary refresh response delay; For CPU occupancy and memory occupancy, the maximum theoretical occupancy is set as the benchmark value according to the device hardware specifications, and the actual occupancy is converted into a ratio relative to the benchmark value. For vocabulary refresh response delay, a dynamic benchmark range is set based on the lowest response delay and the highest response delay of the learning device in historical learning tasks. The vocabulary refresh response delay is mapped to a standardized value within the benchmark range, and weighted processing is performed according to a preset device interference weight coefficient to generate a learning device interference quantification index, which is used to quantitatively evaluate the degree of interference of the learning device performance on the learner's learning effect.

4. The English vocabulary intelligent learning path planning system based on reinforcement learning according to claim 1 is characterized by: The specific process of judging whether the performance of the learning device will interfere with the learning effect and providing feedback is as follows: Obtaining a preset learning device interference quantitative index threshold from a vocabulary learning database; The learning device interference quantitative index is compared with the learning device interference quantitative index threshold. If the learning device interference quantitative index is less than the learning device interference quantitative index threshold, the learning device interference is marked as negligible interference and is not included in the reference value evaluation of the vocabulary learning effect test result. If the learning device interference quantitative index is not less than the learning device interference quantitative index threshold, the learning device interference will be marked as significant interference, and a learning device interference prompt will be pushed to the learner, and the learning device interference will be included in the reference value evaluation of the vocabulary learning effect detection results.

5. The English vocabulary intelligent learning path planning system based on reinforcement learning according to claim 1 is characterized by: The specific process of evaluating the reference value of the vocabulary learning effect test results of learners is as follows: Processing the learner interaction data to obtain comprehensive interaction anomaly evaluation parameters of each vocabulary of the learner; By integrating the learning environment monitoring data, the quantitative indicators of learning equipment interference and the comprehensive interaction anomaly evaluation parameters of each vocabulary of the learner, a multidimensional credibility evaluation model of each vocabulary is constructed to obtain the multidimensional credibility evaluation value of each vocabulary, which is used to quantitatively evaluate the reference value of the vocabulary learning effect detection results; The learning environment monitoring data includes light intensity, noise decibel value and device tilt angle at each monitoring time point.

6. The English vocabulary intelligent learning path planning system based on reinforcement learning according to claim 5 is characterized by: The learner interaction data is processed to obtain the comprehensive interaction anomaly evaluation parameters of each vocabulary of the learner, and the specific process is as follows: Based on the interaction data of all users in historical tasks, a group benchmark dataset of vocabulary learning time, time ratio of exercise types and interaction frequency is established; The interaction data of the current learners are processed to obtain the deviation ratio between the learning time of each vocabulary and the benchmark time, the deviation range between the time ratio of each exercise type and the benchmark ratio, and the difference coefficient between the interaction frequency of each vocabulary learning process and the benchmark frequency. The above deviations are aggregated by dynamic weighting to generate comprehensive interaction anomaly evaluation parameters of each vocabulary of the learner, wherein the comprehensive interaction anomaly evaluation parameters of each vocabulary of the learner are used to characterize the overall deviation degree of the current learner's interaction behavior from the group standard; The learner interaction data includes the learning time of each vocabulary, the time consumption ratio of each vocabulary and each exercise type, and the interaction frequency of each vocabulary learning process.

7. The English vocabulary intelligent learning path planning system based on reinforcement learning according to claim 1 is characterized by: The specific process of determining whether the vocabulary learning task is completed and updating the learning path according to the completion of the vocabulary learning task is as follows: The vocabulary learning effect includes the accuracy rate of each vocabulary exercise; Matching each vocabulary multidimensional credibility evaluation value with a vocabulary practice accuracy threshold value corresponding to each vocabulary multidimensional credibility evaluation value interval preset in a vocabulary learning database, and comparing each vocabulary practice accuracy rate with each vocabulary practice accuracy threshold value; If the correct rate of a vocabulary exercise is not less than the correct rate threshold of the vocabulary exercise, the vocabulary is marked as a valid learning vocabulary, otherwise, the vocabulary is marked as an invalid learning vocabulary, and the number of invalid learning vocabulary is counted; According to the number of invalid learning words and the multi-dimensional credibility evaluation value of each word, the completion degree of the vocabulary learning task is obtained, and the learning path is updated according to the completion degree of the vocabulary learning task.

8. The English vocabulary intelligent learning path planning system based on reinforcement learning according to claim 1 is characterized by: The updating of the learning path according to the completion of the vocabulary learning task includes: The completion degree of the vocabulary learning task is compared with the completion degree threshold of the vocabulary learning task preset in the vocabulary learning database. If the completion degree of the vocabulary learning task is less than the completion degree threshold of the vocabulary learning task, the vocabulary learning task is marked as an unfinished learning task, and the reinforcement learning model obtains the unfinished learning task and re-assigns the practice task.

9. The English vocabulary intelligent learning path planning system based on reinforcement learning according to claim 1 is characterized by: The updating of the learning path according to the completion of the vocabulary learning task also includes: If the completion degree of the vocabulary learning task is not less than the completion degree threshold of the vocabulary learning task, the reinforcement learning model will assign review tasks to the vocabulary learning task, obtain invalid learning vocabulary in the vocabulary learning task, and re-assign practice tasks for the invalid learning vocabulary.

10. A method for the English vocabulary intelligent learning path planning system based on reinforcement learning as claimed in any one of claims 1 to 9, characterized in that: include: S1. Obtain learner individual characteristic information and conduct preliminary planning of vocabulary learning path, wherein the preliminary planning of vocabulary learning path includes vocabulary matching, learning starting point setting and learning task division; S2. Collect the learner's learning device performance data in real time, and quantify the degree of interference of the learning device performance data on the learning effect to obtain the learning device interference quantitative index, determine whether the learning device performance will interfere with the learning effect and provide feedback; S3. Monitor the learner's learning environment and collect learner interaction data in real time, and evaluate the reference value of the learner's vocabulary learning effect test results by combining the quantitative indicators of learning device interference; S4. Obtain the vocabulary learning effect, and judge whether the vocabulary learning task is completed based on the reference value of the vocabulary learning effect test result, and update the learning path based on the completion status of the vocabulary learning task.

Images