English teaching system based on artificial intelligence

Abstract

The invention provides an English teaching system based on artificial intelligence, and the system determines an English learning target selected by a user when detecting that the user carries out an English learning activity; outputting initial English learning content corresponding to the English learning target; detecting learning environment information, physiological parameters and an emotional state analysis result of the user; the emotional state analysis result represents the current anxiety degree of the user; acquiring a learning result of the user on the initial English learning content; determining a current target coefficient of the user according to the physiological parameters, the learning result and the emotional state analysis result; according to the learning environment information and the target coefficient, inputting the learning environment information and the target coefficient into a pre-trained classification model to obtain a current classification result of the user, the prediction result comprising a learning material category of the English learning target; and outputting the English learning material corresponding to the learning material category. By applying the system provided by the embodiment of the invention, personalized English learning experience can be provided for the user.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to an English teaching system based on artificial intelligence. Background Art

[0002] In traditional English teaching, teachers mainly rely on experience and standardized teaching materials to arrange course content and progress. Although this method can provide systematic knowledge transfer, it is incapable of dealing with individual differences among students. With the development of information technology and artificial intelligence, intelligent teaching tools have gradually emerged, aiming to optimize the learning experience in a data-driven way. However, existing technologies still face many challenges in achieving truly personalized education.

[0003] Most existing teaching systems use a unified teaching schedule and content setting, which is difficult to meet the personalized needs of different students. Each student's learning foundation, learning ability and learning style are significantly different. A unified teaching method will inevitably lead to some students finding it difficult to learn, while others find the course content too simple. This unified teaching method not only reduces the overall learning efficiency, but may also dampen students' enthusiasm for learning, affecting their long-term learning motivation and effectiveness. Summary of the invention

[0004] The technical problem to be solved by this application is to provide an English teaching system based on artificial intelligence, which can provide users with a personalized English learning experience. The specific solution is as follows:

[0005] An English teaching system based on artificial intelligence, comprising:

[0006] A first determining unit, configured to determine an English learning goal selected by the user when detecting that the user is performing an English learning activity;

[0007] A first output unit, configured to output initial English learning content corresponding to the English learning goal;

[0008] A first detection unit is used to detect the user's learning environment information, physiological parameters and emotional state analysis results; the emotional state analysis results represent the user's current anxiety level;

[0009] A first acquisition unit, configured to acquire the learning result of the user on the initial English learning content;

[0010] A second determining unit, configured to determine a current target coefficient of the user according to the physiological parameter, the learning result and the emotional state analysis result, wherein the target coefficient represents a degree of information processing burden borne by the user when performing English learning activities;

[0011] An execution unit, configured to input the learning environment information and the target coefficient into a pre-trained classification model to obtain the current classification result of the user, wherein the prediction result includes the learning material category of the English learning target;

[0012] The second output unit is used to output the English learning materials corresponding to the learning material category.

[0013] In the above system, optionally, the second determining unit includes:

[0014] A calculation subunit, configured to calculate a physiological score of the user according to the physiological parameters;

[0015] A non-linear conversion subunit, used for performing non-linear conversion on the physiological score, the learning result and the emotional state analysis result;

[0016] The execution subunit is used to input the physiological score, learning result and emotional state analysis result after nonlinear conversion into a pre-built multi-layer perceptron to obtain the current target coefficient of the user.

[0017] In the above system, optionally, the nonlinear conversion subunit comprises:

[0018] a calculation module, configured to perform weighted processing on the physiological score, the learning result, and the emotional state analysis result to obtain weighted items corresponding to the physiological score, the learning result, and the emotional state analysis result, respectively;

[0019] The nonlinear conversion module is used to perform nonlinear conversion on each of the weighted items using a nonlinear conversion function corresponding to each of the weighted items.

[0020] In the above system, optionally, the first detection unit includes:

[0021] The environment detection subunit is used to collect environmental noise, environmental light intensity, environmental temperature and environmental humidity; and obtain the learning environment information of the user according to the environmental noise, environmental light intensity, environmental temperature and environmental humidity;

[0022] The emotion detection subunit is used to recognize the user's expression and voice to obtain the user's emotional state analysis result;

[0023] The physiological detection subunit is used to collect at least one psychological parameter of the user's heart rate, skin conductivity, breathing information and electroencephalogram information.

[0024] In the above system, optionally, the first determining unit includes:

[0025] A recommendation subunit, configured to recommend a plurality of candidate learning objectives to the user according to the user's historical learning performance, learning preferences, and learning progress;

[0026] The receiving subunit is used to receive a learning goal selected by a user from among the candidate learning goals.

[0027] In the above system, optionally, the first determining unit further includes:

[0028] A planning subunit, configured to receive available time resources and a learning outline set by a user, and generate a learning plan for the user according to the available time resources and the learning outline;

[0029] The time management subunit is used to determine the learning progress of the user according to the historical learning situation of the user, the learning plan and the current time.

[0030] In the above system, optionally, the first acquisition unit includes:

[0031] A learning status detection subunit, used to obtain the user's learning behavior data on the initial English learning content, and obtain the user's online test results, interactive exercise evaluation results, and self-report questionnaire results based on the learning behavior data;

[0032] The evaluation subunit is used to obtain the learning results of the user on the initial English learning content according to the online test results, interactive exercise evaluation results and self-report questionnaire results of the user.

[0033] In the above system, optionally, the second output unit includes:

[0034] A matching subunit, used for acquiring English learning materials corresponding to the learning material category in a preset learning resource library;

[0035] The output subunit is used to output the English learning materials corresponding to the learning material category.

[0036] The above system may optionally further include:

[0037] The second acquisition unit is used to obtain the user's learning results of the English learning material, and return to trigger the second determination unit to re-execute the process of determining the user's current target coefficient based on the physiological parameters, the learning results and the emotional state analysis results.

[0038] The above system may optionally further include:

[0039] A first detection unit, used to detect whether the target coefficient is greater than a preset target coefficient threshold;

[0040] The third output unit is used to output a prompt message when it is detected that the target coefficient is greater than the target coefficient threshold and the number of consecutive times reaches a preset number threshold; the prompt message is used to prompt the user to take a rest.

[0041] An English teaching method based on artificial intelligence, comprising:

[0042] When detecting that a user is performing an English learning activity, determining an English learning goal selected by the user;

[0043] Outputting initial English learning content corresponding to the English learning goal;

[0044] Detecting the user's learning environment information, physiological parameters, and emotional state analysis results; the emotional state analysis results represent the user's current anxiety level;

[0045] Acquiring the user's learning results of the initial English learning content;

[0046] Determining the current target coefficient of the user according to the physiological parameter, the learning result and the emotional state analysis result, wherein the target coefficient represents the degree of information processing burden borne by the user when performing English learning activities;

[0047] According to the learning environment information and the target coefficient, the pre-trained classification model is input to obtain the current classification result of the user, wherein the prediction result includes the learning material category of the English learning target;

[0048] Output English learning materials corresponding to the learning material category.

[0049] Based on the implementation of the present application, an artificial intelligence-based English teaching system is provided, the system includes: a first determination unit, which is used to determine the English learning goal selected by the user when detecting that the user is performing an English learning activity; a first output unit, which is used to output the initial English learning content corresponding to the English learning goal; a first detection unit, which is used to detect the learning environment information, physiological parameters and emotional state analysis results of the user; the emotional state analysis results represent the current anxiety level of the user; a first acquisition unit, which is used to obtain the learning results of the user on the initial English learning content; a second determination unit, which is used to determine the current target coefficient of the user according to the physiological parameters, the learning results and the emotional state analysis results, and the target coefficient represents the degree of information processing burden borne by the user when performing English learning activities; an execution unit, which is used to input the learning environment information and the target coefficient into a pre-trained classification model to obtain the current classification result of the user, and the prediction result includes the learning material category of the English learning goal; a second output unit, which is used to output the English learning material corresponding to the learning material category. The system provided by the embodiment of the present application can provide users with a personalized English learning experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0051] Figure 1 A schematic diagram of the structure of an English teaching system based on artificial intelligence provided in this application;

[0052] Figure 2 A schematic diagram of the structure of a second determination unit provided in this application;

[0053] Figure 3 A system flow chart of an artificial intelligence-based English teaching method provided for this application;

[0054] Figure 4 A flowchart of a process for determining a user's current target coefficient provided by the present application. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0056] In this application, the terms "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, system, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, system, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.

[0057] Most existing teaching systems use a unified teaching schedule and content setting, which is difficult to meet the personalized needs of different students. Each student's learning foundation, learning ability and learning style are significantly different. A unified teaching method will inevitably lead to some students finding it difficult to learn, while others find the course content too simple. This unified teaching method not only reduces the overall learning efficiency, but may also dampen students' enthusiasm for learning, affecting their long-term learning motivation and effectiveness.

[0058] Although some existing teaching systems are also trying to provide personalized teaching, the existing teaching systems usually rely on single-dimensional data to evaluate students' learning status, resulting in inaccurate evaluation of students' learning status. The evaluation mechanism of the existing system is usually static, and recommendations are made based on preset rules or models, while ignoring real-time dynamic changes. For example, when the information processing burden of students is too high, the system fails to detect and adjust the recommended content in time, resulting in the recommended learning materials exceeding the students' ability to bear, or being too simple to effectively challenge the students' abilities.

[0059] Based on this, the embodiment of the present application provides an English teaching system based on artificial intelligence, which can be applied to electronic devices, and the electronic devices can be controllers, personal computers, servers, tablet devices, smart phones, smart wearable devices, etc. The system flow chart of the system is as follows: Figure 1 As shown, specifically including:

[0060] A first determining unit 101 , a first output unit 102 , a first detecting unit 103 , a first acquiring unit 104 , a second determining unit 105 , an executing unit 106 and a second output unit 107 .

[0061] The first determining unit 101 is configured to determine an English learning goal selected by the user when it is detected that the user is performing an English learning activity.

[0062] In this embodiment, the first determination unit can synchronize the user's course progress in real time by connecting to the API of the education platform or teaching material system. At the same time, sensors (such as keyboard input frequency, mouse click mode) or in-application events (such as opening learning materials, starting exercises) are used to detect whether the user is performing English learning activities. When it is detected that the user is performing English learning activities, the system will start the recommendation mechanism and provide the user with multiple alternative learning goals. The system receives the user's selection from each alternative learning goal, thereby determining the English learning goal selected by the user.

[0063] In an embodiment provided in the present application, based on the above implementation process, optionally, the first determining unit 101 includes:

[0064] The recommendation subunit is used to recommend a plurality of candidate learning objectives to the user according to the user's historical learning performance, learning preferences and learning progress.

[0065] The receiving subunit is used to receive a learning goal selected by a user from among the candidate learning goals.

[0066] In this embodiment, the learning goal can be a specific learning purpose or task that the user hopes to achieve in the process of learning English. These goals can be short-term (such as mastering a specific grammar point, completing a reading comprehension exercise) or long-term (such as passing a certain level of English test, improving overall English proficiency).

[0067] For example, when the system detects that the user is learning English, a learning prompt window pops up, asking the user what he wants to learn today and providing the following options: Option 1, review yesterday's content (based on the user's historical learning data); Option 2, learn new grammar points (based on the course outline); Option 3, improve listening skills (based on the results of the user's most recent test). When the user selects Option 2, the goal is further refined, asking the user which part of grammar he wants to learn, and the user selects "past perfect tense" as the learning goal.

[0068] Xiao Ming chose "Learn new grammar points". At this point, the system further refines this goal and asks him which part of grammar he wants to learn. Xiao Ming chose "Past perfect tense".

[0069] In an embodiment provided in the present application, based on the above implementation process, optionally, the first determining unit 101 further includes:

[0070] A planning subunit, configured to receive available time resources and a learning outline set by a user, and generate a learning plan for the user according to the available time resources and the learning outline;

[0071] The time management subunit is used to determine the learning progress of the user according to the historical learning situation of the user, the learning plan and the current time.

[0072] In this embodiment, the planning subunit can generate a detailed study plan based on the available time resources and study outline set by the user. The user can enter the specific time period available for study every day or every week through the system interface, and the system can also automatically detect the user's free time through calendar integration. In addition, the user can upload or select a preset study outline, such as a textbook catalog or an exam outline. Based on this information, the planning subunit will generate a reasonable and personalized schedule to ensure that the time allocation of learning tasks is scientific and reasonable, avoiding excessive concentration or dispersion, thereby providing users with an efficient learning framework.

[0073] Optionally, the time management subunit can dynamically adjust and determine the user's learning progress based on the user's historical learning situation, current learning plan and current time. Collect and analyze the user's past learning data, including completed tasks, test scores and interactive exercise results, to evaluate the user's learning efficiency and weak links. Combined with the current learning plan, the time management subunit compares historical data, evaluates the rationality of the plan, and recommends adjustments when appropriate. The system monitors the current time in real time and flexibly adjusts learning tasks to ensure that users complete high-priority tasks within the appropriate time period, avoid learning interruptions caused by time conflicts, and dynamically adjust the learning progress to ensure that the learning burden is moderate.

[0074] For example, when a user logs into the system and starts a new learning session, the system automatically detects that the user is doing an English learning activity and pops up a prompt window to ask him what he wants to learn today. The user sets 7 to 9 pm every day as his English learning time and selects "CET-4 preparation" as the learning outline. Based on this information, the system generates a detailed study plan, including specific tasks for each day, such as grammar review, vocabulary accumulation, and mock tests. Subsequently, the system analyzes Xiao Ming's learning data for the past week and finds that the user is weak in the listening part, so it is recommended to increase the time for listening practice and arrange a complete mock test on the weekend. The system also reminds users that on the upcoming weekdays, due to other arrangements, they can shorten their study time in the evening and focus on efficient task completion.

[0075] The first output unit 102 is used to output the initial English learning content corresponding to the English learning goal.

[0076] In this embodiment, after determining the English learning goal of the user, the first output unit 102 outputs the initial English learning content corresponding to the goal. It can ensure that the user can immediately obtain learning materials that match their learning goals, such as explanations of specific grammar points, exercises of related vocabulary, listening or reading materials, etc. Through precise content output, the first output unit not only provides the user with a clear starting point for learning, but also enhances the pertinence and efficiency of learning, allowing the user to enter an effective learning state in the shortest time.

[0077] The first detection unit 103 is used to detect the learning environment information, physiological parameters and emotional state analysis results of the user; the emotional state analysis results represent the current anxiety level of the user.

[0078] In this embodiment, the user's learning environment information, physiological parameters and emotional state analysis results are detected, especially the emotional state used to characterize the user's current anxiety level. Through multi-dimensional data collection and analysis, the unit can provide a scientific basis for subsequent personalized learning adjustments to ensure that users learn in the best state.

[0079] In an embodiment provided in the present application, based on the above implementation process, optionally, the first detection unit 103 includes:

[0080] The environment detection subunit is used to collect environmental noise, environmental light intensity, environmental temperature and environmental humidity; and obtain the learning environment information of the user according to the environmental noise, environmental light intensity, environmental temperature and environmental humidity;

[0081] The emotion detection subunit is used to recognize the user's expression and voice to obtain the user's emotional state analysis result;

[0082] The physiological detection subunit is used to collect at least one psychological parameter of the user's heart rate, skin conductivity, breathing information and electroencephalogram information.

[0083] In this embodiment, the environment detection subunit collects a variety of physical parameters related to the user's learning environment, including ambient noise, ambient light intensity, ambient temperature, and ambient humidity. Among them, too high or too low noise levels may affect the user's concentration. Appropriate light helps protect eyesight and improve reading comfort. Comfortable temperature and humidity conditions can improve learning efficiency and avoid distraction due to discomfort. By combining these data, the environment detection subunit can evaluate whether the user's learning environment is suitable, and provide optimization suggestions or automatically adjust the difficulty and type of learning content accordingly.

[0084] Optionally, the emotion detection subunit obtains the analysis results of the user's emotional state mainly by identifying the user's expression and voice, paying special attention to the user's anxiety level. Specifically, the user's facial expression can be captured by a camera, and the emotional changes can be analyzed in combination with a machine learning algorithm. The user's voice is recorded by a microphone, and features such as intonation and speaking speed are analyzed to determine the user's emotional state. This non-invasive emotion monitoring method not only improves the convenience of data collection, but also reflects the user's psychological state in real time, helping the system to take corresponding support measures in a timely manner, such as providing relaxation techniques or adjusting the difficulty of learning tasks. In this embodiment, the user's emotional state analysis results can be obtained in an existing manner, which will not be repeated here.

[0085] The first acquisition unit 104 is configured to acquire the learning result of the user on the initial English learning content.

[0086] In this embodiment, the first acquisition unit can evaluate the user's understanding and mastery of the initial English learning content, and comprehensively acquire the user's learning results by collecting the user's learning behavior data, such as online test results, interactive practice evaluations, and self-report questionnaires. This unit can not only quantify the user's knowledge mastery, but also identify the strengths and weaknesses in the learning process, providing a basis for subsequent personalized adjustments and feedback, ensuring the continuous optimization of the learning path and maximization of learning effects.

[0087] In an embodiment provided in the present application, based on the above implementation process, optionally, the first acquisition unit includes:

[0088] A learning status detection subunit, used to obtain the user's learning behavior data on the initial English learning content, and obtain the user's online test results, interactive exercise evaluation results, and self-report questionnaire results based on the learning behavior data;

[0089] The evaluation subunit is used to obtain the learning results of the user on the initial English learning content according to the online test results, interactive exercise evaluation results and self-report questionnaire results of the user.

[0090] In this embodiment, the learning status detection subunit can obtain the user's learning behavior data for the initial English learning content, and obtain the user's online test results, interactive exercise evaluation results, and self-report questionnaire results based on these data. Specifically: the online test results include test results in the form of multiple-choice questions, fill-in-the-blank questions, true-or-false questions, etc., which quantify the user's memory and understanding of knowledge points. The interactive exercise evaluation results include evaluation results in interactive forms such as listening exercises, oral exercises, and writing exercises. The self-report questionnaire results can be obtained through subjective feedback from the user. The learning status detection subunit uses a variety of evaluation tools to ensure that the user's learning behavior data is comprehensively collected from different angles.

[0091] Optionally, the assessment subunit can be based on aggregating the online test results, interactive exercise evaluation results, and self-report questionnaire results to form a comprehensive data set. According to the importance of different types of assessment results (such as tests, exercises, questionnaires), corresponding weights are assigned. For example, the test score may account for 60%, the interactive exercise evaluation account for 30%, and the self-report questionnaire account for 10%. Based on the weights, a comprehensive learning score, i.e., the learning outcome, is calculated.

[0092] The second determining unit 105 is used to determine the current target coefficient of the user according to the physiological parameter, the learning result and the emotional state analysis result, wherein the target coefficient represents the degree of information processing burden borne by the user when performing English learning activities.

[0093] In an embodiment provided in the present application, based on the above implementation process, optionally, the second determining unit, such as Figure 2 As shown, including:

[0094] A calculation subunit 201 is used to calculate the physiological score of the user according to the physiological parameters;

[0095] A non-linear conversion subunit 202, configured to perform non-linear conversion on the physiological score, the learning result and the emotional state analysis result;

[0096] The execution subunit 203 is used to input the physiological score, learning result and emotional state analysis result after nonlinear conversion into a pre-built multi-layer perceptron to obtain the current target coefficient of the user.

[0097] In this embodiment, the calculation subunit can calculate the user's physiological score S according to the user's physiological parameters. phys Specifically, the user's heart rate (HR), skin conductivity (GSR), breathing information (BR) and electroencephalogram (EEG) information can be obtained from the physiological detection subunit. Using the preset algorithm or model, these physiological parameters are converted into a comprehensive physiological score S phys . For example, the following formula can be used:

[0098] S phys =w1*f(HR)+w2·f(GSR)+w3·f(BR)+w4·f(EEG)

[0099] Among them, w1, w2, w3, and w4 are the weights of each physiological parameter, and f(·) is a function that standardizes or normalizes each parameter to ensure that different parameters are compared in the same dimension.

[0100] In an embodiment provided in the present application, based on the above implementation process, optionally, the nonlinear conversion subunit includes:

[0101] a calculation module, configured to perform weighted processing on the physiological score, the learning result, and the emotional state analysis result to obtain weighted items corresponding to the physiological score, the learning result, and the emotional state analysis result, respectively;

[0102] The nonlinear conversion module is used to perform nonlinear conversion on each of the weighted items using a nonlinear conversion function corresponding to each of the weighted items.

[0103] In this embodiment, the calculation module calculates the physiological score S phys , learning result S learn And the emotional state analysis results S emo They are weighted separately and assigned different weights w according to their importance in evaluating the user's information processing burden. phys , w learn , w emo ,For example:

[0104] S′ phys =w phys *S phys

[0105] S′ learn =w learn *S learn

[0106] S′ emo =w emo *S emo

[0107] Calculate the weighted value of each dimension to form the corresponding weighted term S′ phys , S′ learn and S′ emo .

[0108] Optionally, a pre-set nonlinear transformation function (such as the Sigmoid function) can be used. Or the ReLU function ReLU(x)=max(0,x)), which performs a nonlinear transformation on each weighted item. For example:

[0109] S″ phys =σ(S′ phys )

[0110] S″ learn =σ(S′ learn )

[0111] S″ emo =σ(S′emo )

[0112] Optionally, the nonlinear transformation functions corresponding to different weighting items can be the same or different.

[0113] In this embodiment, the execution subunit can use the data after nonlinear transformation as the input feature x=[S″ phys , S″ learn , S″ emo ] is sent to the multilayer perceptron. The multilayer perceptron performs complex calculations and analyses on the input data through its internal neural network structure and outputs a comprehensive target coefficient T. Specifically, assuming that the output layer of the multilayer perceptron uses a linear activation function, the target coefficient T can be expressed as:

[0114] T=MLP(x)=W L *σ(W L-1 *σ(...σ(W (1) *x+b (1) )...+b (L-1) )+b (L) )

[0115] Among them, W i and b (i) are the weight matrix and bias vector of the i-th layer, L is the number of network layers, and σ(·) is the activation function.

[0116] In this embodiment, the training process of the multilayer perceptron may include: obtaining an initial multilayer perceptron and a first training data set, the first training data set including multiple first training data samples and a sample label corresponding to each first training sample, the first training sample including historical physiological scores, historical learning results, and historical emotional state analysis results, the sample label may be a target coefficient determined by an expert system or a rule set, and the initial multilayer perceptron is trained using the first training sample in the first training data set until the number of training times of the initial multilayer perceptron is greater than a preset number threshold, and the initial multilayer perceptron is determined as a trained multilayer perceptron.

[0117] In this embodiment, the target coefficient T is a quantitative indicator that indicates the degree of information processing burden that the user bears when learning English in the current state. A higher target coefficient means that the user may face a greater cognitive load and needs to adjust the learning content or take a break appropriately.

[0118] For example, the user is using the improved English teaching system. After the user has completed a period of learning about "past perfect tense", the system performs the following operations through the second determining unit 105:

[0119] The computing subunit calculates the physiological score S based on the user's heart rate, skin conductivity and breathing information.phys .

[0120] Nonlinear transformation of subunits to physiological score S phys , learning results S learn And the emotional state analysis results S emo Perform weighted processing.

[0121] Use the Sigmoid function to perform nonlinear transformation on the weighted items to obtain S″ phys , S″ learn , S″ emo .

[0122] Execution subunit: Input the nonlinearly transformed data into the multilayer perceptron and calculate the target coefficient T.

[0123] If the target coefficient T is high, the system will remind Xiao Ming to take a break or simplify the learning task to reduce his cognitive load; if T is low, the system may add some challenging exercises to improve the learning effect.

[0124] The execution unit 106 is used to input the learning environment information and the target coefficient into a pre-trained classification model to obtain the current classification result of the user, and the prediction result includes the learning material category of the English learning target.

[0125] In this embodiment, the classification model may be a multi-layer perceptron model, a random forest model, a support vector machine model, or the like.

[0126] Optionally, the model training process may be to obtain an initial model and a second training data set, wherein the second training data set includes a plurality of second training samples and a classification label corresponding to each sample, historical environment information of the second training samples, and a target coefficient. The initial model is trained using the second training samples in the second training data set until the prediction accuracy of the initial model is greater than a preset accuracy threshold, and the initial model having a prediction accuracy greater than the accuracy threshold is determined as a trained classification model.

[0127] In this embodiment, the learning material categories may include one or more of basic listening materials, basic reading materials, basic writing materials, basic vocabulary materials, intermediate listening materials, intermediate reading materials, intermediate writing materials, intermediate vocabulary materials, advanced listening materials, advanced reading materials, advanced writing materials, and advanced vocabulary materials.

[0128] The second output unit 107 is used to output the English learning materials corresponding to the learning material category.

[0129] In an embodiment provided in the present application, based on the above implementation process, optionally, the second output unit includes:

[0130] A matching subunit, used for acquiring English learning materials corresponding to the learning material category in a preset learning resource library;

[0131] The output subunit is used to output the English learning materials corresponding to the learning material category.

[0132] In an embodiment provided in the present application, based on the above implementation process, optionally, it further includes:

[0133] The second acquisition unit is used to obtain the user's learning results of the English learning material, and return to trigger the second determination unit to re-execute the process of determining the user's current target coefficient based on the physiological parameters, the learning results and the emotional state analysis results.

[0134] In an embodiment provided in the present application, based on the above implementation process, optionally, it further includes:

[0135] A first detection unit, used to detect whether the target coefficient is greater than a preset target coefficient threshold;

[0136] The third output unit is used to output a prompt message when it is detected that the target coefficient is greater than the target coefficient threshold and the number of consecutive times reaches a preset number threshold; the prompt message is used to prompt the user to take a rest.

[0137] This system integrates multi-dimensional data (including physiological parameters, learning outcomes, emotional states, and environmental information) and uses advanced machine learning models to dynamically evaluate the user's learning burden and personalize recommendations for the most suitable learning materials. It not only improves learning efficiency and user experience, but also ensures that users can learn efficiently in the most appropriate state, truly realizing intelligent education that teaches students in accordance with their aptitude and adapts to the situation.

[0138] The present application also provides an artificial intelligence-based English teaching method, which can be applied to electronic devices. The method flow chart of the method is as follows: Figure 3 As shown, specifically including:

[0139] S301: When it is detected that a user is performing an English learning activity, determining an English learning goal selected by the user.

[0140] S302: Outputting the initial English learning content corresponding to the English learning goal.

[0141] S303: Detecting the learning environment information, physiological parameters and emotional state analysis results of the user; the emotional state analysis results represent the current anxiety level of the user.

[0142] S304: Obtaining the learning result of the user on the initial English learning content.

[0143] S305: Determine the current target coefficient of the user according to the physiological parameter, the learning result and the emotional state analysis result, wherein the target coefficient represents the degree of information processing burden borne by the user when performing English learning activities.

[0144] S306: Input the learning environment information and the target coefficient into a pre-trained classification model to obtain the current classification result of the user, wherein the prediction result includes the learning material category of the English learning target.

[0145] S307: Output the English learning materials corresponding to the learning material category.

[0146] In an embodiment provided in the present application, based on the above scheme, optionally, the process of determining the current target coefficient of the user according to the physiological parameters, the learning results and the emotional state analysis results is as follows: Figure 4 As shown, including:

[0147] S401: Calculate the physiological score of the user according to the physiological parameters.

[0148] S402: Perform nonlinear transformation on the physiological score, the learning result, and the emotional state analysis result.

[0149] S403: Inputting the physiological score, learning result and emotional state analysis result after nonlinear conversion into a pre-built multi-layer perceptron to obtain the current target coefficient of the user.

[0150] It should be noted that, in this document, relational terms such as first and second, etc. are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0151] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0152] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server, etc.) to execute the system described in each embodiment of the present application or some parts of the embodiments.

[0153] The above is a detailed introduction to an artificial intelligence-based English teaching system provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the system of the present application and its core ideas. At the same time, for general technical personnel in this field, according to the ideas of the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. An English teaching system based on artificial intelligence, characterized in that: include: A first determining unit, configured to determine an English learning goal selected by the user when detecting that the user is performing an English learning activity; A first output unit, configured to output initial English learning content corresponding to the English learning goal; A first detection unit, used to detect the user's learning environment information, physiological parameters and emotional state analysis results; The emotional state analysis result represents the current anxiety level of the user; A first acquisition unit, configured to acquire the learning result of the user on the initial English learning content; A second determining unit, configured to determine a current target coefficient of the user according to the physiological parameter, the learning result and the emotional state analysis result, wherein the target coefficient represents a degree of information processing burden borne by the user when performing English learning activities; An execution unit, configured to input the learning environment information and the target coefficient into a pre-trained classification model to obtain the current classification result of the user, wherein the prediction result includes the learning material category of the English learning target; The second output unit is used to output the English learning materials corresponding to the learning material category.

2. The system according to claim 1, characterized in that The second determining unit includes: A calculation subunit, configured to calculate a physiological score of the user according to the physiological parameters; A non-linear conversion subunit, used for performing non-linear conversion on the physiological score, the learning result and the emotional state analysis result; The execution subunit is used to input the physiological score, learning result and emotional state analysis result after nonlinear conversion into a pre-built multi-layer perceptron to obtain the current target coefficient of the user.

3. The system according to claim 2, characterized in that The nonlinear conversion subunit comprises: a calculation module, configured to perform weighted processing on the physiological score, the learning result, and the emotional state analysis result to obtain weighted items corresponding to the physiological score, the learning result, and the emotional state analysis result, respectively; The nonlinear conversion module is used to perform nonlinear conversion on each of the weighted items using a nonlinear conversion function corresponding to each of the weighted items.

4. The system according to claim 1, characterized in that The first detection unit comprises: The environment detection subunit is used to collect environmental noise, environmental light intensity, environmental temperature and environmental humidity; and obtain the learning environment information of the user according to the environmental noise, environmental light intensity, environmental temperature and environmental humidity; The emotion detection subunit is used to recognize the user's expression and voice to obtain the user's emotional state analysis result; The physiological detection subunit is used to collect at least one psychological parameter of the user's heart rate, skin conductivity, breathing information and electroencephalogram information.

5. The system according to claim 1, characterized in that The first determining unit includes: A recommendation subunit, configured to recommend a plurality of candidate learning objectives to the user according to the user's historical learning performance, learning preferences, and learning progress; The receiving subunit is used to receive a learning goal selected by a user from among the candidate learning goals.

6. The system according to claim 1, characterized in that The first determining unit further includes: A planning subunit, configured to receive available time resources and a learning outline set by a user, and generate a learning plan for the user according to the available time resources and the learning outline; The time management subunit is used to determine the learning progress of the user according to the historical learning situation of the user, the learning plan and the current time.

7. The system according to claim 1, characterized in that The first acquisition unit includes: A learning status detection subunit, used to obtain the user's learning behavior data on the initial English learning content, and obtain the user's online test results, interactive exercise evaluation results, and self-report questionnaire results based on the learning behavior data; The evaluation subunit is used to obtain the learning results of the user on the initial English learning content according to the online test results, interactive exercise evaluation results and self-report questionnaire results of the user.

8. The system according to claim 1, characterized in that The second output unit comprises: A matching subunit, used for acquiring English learning materials corresponding to the learning material category in a preset learning resource library; The output subunit is used to output the English learning materials corresponding to the learning material category.

9. The method according to claim 1, characterized in that: Also includes: The second acquisition unit is used to obtain the user's learning results of the English learning material, and return to trigger the second determination unit to re-execute the process of determining the user's current target coefficient based on the physiological parameters, the learning results and the emotional state analysis results.

10. The system according to claim 9, characterized in that Also includes: A first detection unit, used to detect whether the target coefficient is greater than a preset target coefficient threshold; A third output unit is used to output prompt information when it is detected that the target coefficient is greater than the target coefficient threshold for a continuous number of times reaching a preset number threshold; The prompt information is used to prompt the user to take a rest.

Images