Practice effect evaluation method of English learning simulation system based on artificial intelligence

Abstract

The invention discloses a practice effect evaluation method for an English learning simulation system based on artificial intelligence, and particularly relates to the technical field of English practice effect evaluation, and the method comprises the steps: 1, obtaining user information, 2, analyzing the concentration degree of a user, 3, carrying out the practice analysis of each skill of the user, 4, carrying out the improvement index analysis of each skill of the user, and 5, carrying out the evaluation of the practice effect of the English learning simulation system. 5, user practice effect response; and 6, user practice effect comprehensive evaluation. Starting from four skills of English learning, each data influencing the exercise score of the user in each skill exercise process is comprehensively considered, the concentration coefficient of the user in each skill exercise process is combined, the exercise score of the user in each skill exercise process is obtained through analysis, the exercise score of English learning simulation of the user is accurately reflected, and the user experience is improved. Therefore, the user can find the defects in the practice process timely and make up the defects timely, thereby improving the English level.

Description

Technical Field

[0001] The present invention relates to the technical field of English practice effect evaluation, and in particular to a practice effect evaluation method of an English learning simulation system based on artificial intelligence. Background Art

[0002] The artificial intelligence-based English learning simulation system refers to the use of artificial intelligence technology to create a virtual, interactive learning environment to help students improve their English language skills. By combining natural language processing, speech recognition, machine learning and other technologies, this system can provide learners with a personalized, immersive language learning experience, simulate various real communication scenarios, and help students learn and use English more effectively. However, feedback on learning effects and improvement methods are essential for students to improve their English proficiency. Therefore, it is necessary to study the practice effect evaluation method of the English learning simulation system.

[0003] Prior art, such as the invention patent application CN117350904A, discloses a method for evaluating the practice effect of an English practice system based on artificial intelligence, which includes: obtaining English learning behavior data and calculating an English learning interest index; obtaining English practice question data and calculating index data; obtaining English learning goal data and calculating an English learning goal achievement index; and calculating an English practice effect comprehensive evaluation index. The present invention comprehensively evaluates the learning effect of users in English practice through an English practice effect comprehensive evaluation index formula by combining an English learning interest index, an English practice correctness index, an English practice difficulty index, an English learning efficiency index, an English practice consolidation index, and an English learning goal achievement index.

[0004] Prior art, such as the invention patent application CN114547154A, discloses an intelligent interactive English training method, system and computer-readable medium, the method comprising: flexibly adjusting a suitable training plan according to the user's ability, training starting point and training goal, starting from vocabulary training, using reading, listening, writing and speaking to improve the user's comprehensive English ability based on vocabulary mastery, and learners continuously strengthen their ability to master new words in each link of listening, speaking, reading and writing, rather than just rigidly memorizing words. At the same time, the difficulty of training changes dynamically with the user's training situation, so that the user can train at a difficulty suitable for his or her own situation, and the user's training situation is recorded, so that the effect after training can be more clearly understood.

[0005] In combination with the above scheme, it is found that the evaluation of the English practice effect in the prior art usually takes English practice questions as the starting point, and considers the number and accuracy of English practice questions for evaluation, ignoring the fact that there are four requirements for English learning: listening, speaking, reading and writing. The different data behind each type of learning reflects the user's English practice effect. This rough evaluation method cannot provide the user's actual performance in the English learning process. Regarding the four requirements for English learning, the prior art focuses on how to improve the English learning effect by strengthening word learning. This single training method cannot meet the diversified learning methods in reality, and cannot provide corresponding solutions to improve the user's English learning effect in a targeted manner, resulting in certain limitations in the English simulation learning system and failure to meet actual usage needs. Summary of the invention

[0006] The purpose of the present invention is to provide a practice effect evaluation method for an English learning simulation system based on artificial intelligence, which solves the problems existing in the background technology.

[0007] To solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides a practice effect evaluation method for an English learning simulation system based on artificial intelligence, including: Step 1, obtaining user information: extracting user information from a database, including the user's practice time for each skill and the practice score of each skill in the previous stage.

[0008] Step 2: User concentration analysis: Obtain facial image videos and body posture image videos of users during the practice of each skill, and analyze the user's concentration coefficient during the practice of each skill.

[0009] Step 3. Analysis of the user's skill practice: obtain the data during the user's skill practice, and calculate the user's listening skill practice score based on the data during the user's listening skill practice; calculate the user's speaking skill practice score based on the data during the user's speaking skill practice; calculate the user's reading skill practice score based on the data during the user's reading skill practice; and calculate the user's writing skill practice score based on the data during the user's writing practice.

[0010] Step 4: Analysis of the improvement index of each skill of the user: obtain the practice score dataset of each skill of the user in each historical stage, calculate the improvement index of each skill of the user based on the practice score dataset of each skill of the user in the last stage, obtain the improvement index of each skill of the user at each stage, and calculate the fluctuation value of the improvement index of each skill of the user.

[0011] Step 5: Response to user practice effects: Implement response measures for the level ranges corresponding to the fluctuation values ​​of each user's skill improvement index.

[0012] Step 6: Comprehensive evaluation of user practice results: Based on the practice scores and improvement indexes of each skill, calculate the comprehensive score of the user's practice results.

[0013] The beneficial effects of the present invention are: 1. The present invention targets the four skills of English learning, including different characteristic attributes in the process of listening, speaking, reading and writing practice, and divides them into prominent characteristics and general characteristics to analyze the concentration coefficient of users in the process of practicing each skill, thereby improving the accuracy of the concentration coefficient of users in the process of practicing each skill.

[0014] 2. The present invention starts from the four skills of English learning, comprehensively considers various data that affect the practice scores of users during the practice of each skill, and combines the concentration coefficient of users during the practice of each skill to analyze and obtain the practice scores of users during the practice of each skill. This comprehensive consideration method can accurately reflect the practice scores of users' English learning simulations, which is convenient for users to promptly discover defects in the practice process and make up for them in time to improve their English level.

[0015] 3. The present invention calculates the improvement index of each skill of the user based on the practice score of each skill at the current stage and the practice score of each skill at the previous stage, analyzes the improvement index of each skill of the user at each stage, and obtains the fluctuation value of the improvement index of each skill of the user. It can reflect whether the improvement effect of each skill of the user over a period of time is highly stable or highly unstable, so that the user can make timely adjustments to meet the needs of practicing various English skills, thereby improving the level of various English skills. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0018] 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.

[0019] For ease of understanding, some professional terms in this application are explained below: The vowel part in the audio waveform graph refers to the waveform graph, which is usually shown as a longer and uniform waveform. It is relatively stable in the graph, has a longer sound wave period, and lasts longer.

[0020] The consonant part in the audio waveform graph refers to the shorter and more prominent fluctuations in the waveform graph, showing a short and rapidly changing waveform, such as the instantaneous burst of a plosive sound.

[0021] Reference Figure 1 As shown, the present invention provides a practice effect evaluation method for an English learning simulation system based on artificial intelligence, the method comprising: step one, obtaining user information: extracting user information from a database, including the user's practice time for each skill and the practice score of each skill in the previous stage.

[0022] Step 2: User concentration analysis: Obtain facial image videos and body posture image videos of users during the practice of each skill, and analyze the user's concentration coefficient during the practice of each skill.

[0023] Step 3. Analysis of the user's skill practice: obtain the data during the user's skill practice, and calculate the user's listening skill practice score based on the data during the user's listening skill practice; calculate the user's speaking skill practice score based on the data during the user's speaking skill practice; calculate the user's reading skill practice score based on the data during the user's reading skill practice; and calculate the user's writing skill practice score based on the data during the user's writing practice.

[0024] Step 4: Analysis of the improvement index of each skill of the user: obtain the practice score dataset of each skill of the user in each historical stage, calculate the improvement index of each skill of the user based on the practice score dataset of each skill of the user in the last stage, obtain the improvement index of each skill of the user at each stage, and calculate the fluctuation value of the improvement index of each skill of the user.

[0025] Step 5: Response to user practice effects: Implement response measures for the level ranges corresponding to the fluctuation values ​​of each user's skill improvement index.

[0026] Step 6: Comprehensive evaluation of user practice results: Based on the practice scores and improvement indexes of each skill, calculate the comprehensive score of the user's practice results.

[0027] It should be noted that, in a specific embodiment, the database is used to store facial videos and body posture videos of users during the practice of each skill, store the practice time of each skill of the user, store the total number of exercises completed and the total number of correct exercises during the practice of the user's listening skills, store the number of words correctly marked in the user's listening content vocabulary list, store the standard reaction time of each exercise in the process of each user's listening skills practice, store the audio, standard speaking speed, allowed number of freezes, and standard audio waveform of each word in the process of the user's oral skills practice, store the standard reading speed, content retelling text of the reading material, and standard content text in the process of the user's reading skills practice, store the writing sentence pattern list, each sentence segment of the writing text, and the standard total number of sentence patterns in the writing text that are included in the writing sentence pattern list during the user's writing skills practice, store the practice score data set of each skill of the user in the previous stage, store the first interval and the second interval of the fluctuation value of each skill improvement index of the user, and store the first interval, the second interval, and the third interval of the comprehensive score of the user's practice effect.

[0028] In a specific embodiment of the present invention, the analysis of the concentration coefficient of the user during each skill practice process is specifically performed by: for the prominent feature reflecting concentration during the listening skill practice process: blinking action, the frame processing interval of the facial video during the user's listening skill practice process is set to , and obtain sample slices of each blinking action during the user's listening skill practice.

[0029] Aiming at the general features that reflect concentration during listening skill practice: facial expressions and head postures, the frame processing interval of the facial video and body posture video during the user's listening skill practice is set to , and obtain sample slices of facial expressions and head postures during the user's listening skill practice.

[0030] For the prominent features that reflect the concentration during the user's oral skill practice: mouth movements, general features: eyebrow movements, hand movements; for the prominent features that reflect the concentration during the user's reading skill practice: eye focus points, general features: mouth movements, hand movements; for the prominent features that reflect the concentration during the user's writing skill practice: body postures, general features: eye focus points, head postures, the facial video and body posture video during the user's listening skill practice are set to a frame processing interval of 100 according to the general features. , highlight the feature setting frame processing interval is processed in this way.

[0031] Get the sample slices of the user's skill practice process, input them into the convolutional neural network through computer vision technology, and get the user's concentration time during the skill practice process after processing. , , Indicates the number of each user's English skill, through the formula: , calculate the user's concentration coefficient during each skill practice ,in Represents the user number extracted from the database Skill practice time.

[0032] The present invention targets four skills of English learning, including different characteristic attributes in the process of listening, speaking, reading and writing practice, which are divided into prominent characteristics and general characteristics to analyze the concentration coefficient of users in the process of practicing each skill, and improve the accuracy of the concentration coefficient of users in the process of practicing each skill.

[0033] In a specific embodiment of the present invention, the user's listening skill practice score is calculated based on the data in the user's listening skill practice process. The specific analysis method is: extract the total number of exercises completed by the user in the listening skill practice process from the database. , Total number of correct exercises , through the timing module, obtain the reaction time of each exercise during the user's listening skill practice ,in , Indicates the number of each exercise in the user's listening skills practice process. Indicates the total number of exercises during the user's listening skill practice.

[0034] Extract the user's listening content vocabulary list from the database and mark the number of correct vocabulary The listening vocabulary list contains the total number of words in the listening content. .

[0035] It should be noted that, in a specific embodiment, the number of correctly marked words in the user's listening content vocabulary list is extracted from the database. The specific analysis method is to first extract the vocabulary list containing the listening content from the database, and the user marks the words he hears. Then the system counts the number of correctly marked words in the user's listening content vocabulary list, and finally stores it in the database.

[0036] Based on the concentration coefficient of the user's listening skills practice process , through the formula: , calculate the user's listening skill practice score ,in Indicates the standard reaction time of each exercise extracted from the database during the user's listening skills practice Indicates the pre-set weight factor affecting the correctness of exercises during the user's listening skills practice. Indicates the influence weight factor of the marked correct vocabulary in the preset user listening skill practice process, Represents the preset weight factor affecting the response efficiency during the user's listening skill practice.

[0037] It should be noted that, in a specific embodiment, the pre-set weight factors affecting the data in the user's listening skill practice process are pre-set based on a comprehensive evaluation by relevant English teaching experts based on three factors affecting the user's listening skill practice score: answer accuracy, content repetition, and reaction time. For example, , , .

[0038] In a specific embodiment of the present invention, the user's oral skill practice score is calculated based on the data during the user's oral skill practice. The specific analysis method is as follows: the audio during the user's oral skill practice is extracted from the database, and the speech speed during the user's oral skill practice is first obtained by using an audio analysis program. , Number of freezes , through the formula: , calculate the speed compliance index of the user during the oral skills practice ,in It represents the standard speaking speed of the user during the oral skill practice extracted from the database.

[0039] It should be noted that, in a specific embodiment, the standard speaking speed during the user's oral skill practice is preset by English teaching experts based on the speaking speed of a person during normal oral communication.

[0040] By formula: , calculate the fluency index of the user's oral skills practice process ,in Indicates the allowed number of pauses during the user's oral skill practice extracted from the database.

[0041] Then, the audio during the user's oral skill practice is sliced ​​according to each word to obtain the actual audio waveform of each word during the user's oral skill practice, which is recorded as ,in , Indicates the number of each word during the user's oral skills practice. Represents the total number of words in the user's oral skills practice process, and obtains the standard audio waveform of each word in the user's oral skills practice process from the database, which is recorded as .

[0042] Compare the actual audio waveform of each word during the user's oral skill practice with the standard audio waveform, and count the total number of vowel parts in the actual audio waveform of each word during the user's oral skill practice , the total number of vowels in the standard audio waveform , analyze the similarity between the waveform of each vowel part in the actual audio waveform of each word and the waveform of each vowel part in the standard audio waveform during the user's oral skill practice, and record it as ,in , Indicates the number of each vowel part in the audio waveform of a word during the user's oral skill practice. Indicates the total number of vowel parts in the actual audio waveform of a word during the user's speaking skill practice.

[0043] By formula: , calculate the vowel matching coefficient during the user's oral skills practice .

[0044] According to the analysis method of vowel matching coefficient, the consonants in the user's oral skill practice process are analyzed to obtain the consonant matching coefficient in the user's oral skill practice process. .

[0045] It should be noted that, in a specific embodiment, the consonant matching coefficient of the user during the oral skill practice is The analysis method is: count the total number of consonant parts in the actual audio waveform of each word during the user's oral skill practice , the total number of consonant parts in the standard audio waveform , analyze the similarity between the waveform of each consonant part in the actual audio waveform of each word and the waveform of each consonant part in the standard audio waveform during the user's oral skill practice, and record it as ,in , Indicates the number of each consonant part in the audio waveform during the user's oral skill practice. Indicates the total number of consonant parts in the actual audio waveform of each word during the user's oral skill practice.

[0046] By formula: , calculate the consonant matching coefficient during the user's oral skills practice .

[0047] It should be noted that, in a specific embodiment, the similarity of the waveforms is determined by a time domain analysis method. First, two waveforms are sampled, and the Pearson correlation coefficient between the two waveforms is calculated for the sampling results to determine the similarity of the two waveforms.

[0048] By formula: , calculate the pronunciation accuracy index of the user during the oral skill practice .

[0049] Based on the speed conformity index, fluency conformity coefficient, and pronunciation accuracy index of the user's oral skills practice, the formula is: , calculate the user's speaking skill practice score ,in It is the user's concentration coefficient during the oral skill practice process.

[0050] In a specific embodiment of the present invention, the user's reading skill practice score is calculated based on the data in the user's reading skill practice process. The specific analysis method is: extracting the total number of words in the reading material during the user's reading skill practice process from the database Total time spent reading , through the formula: , calculate the reading speed of the user during reading skill practice , through the formula: , the reading speed index of the user during reading skill practice is calculated, where Represents the standard reading speed of the user during reading skill practice stored in the database.

[0051] It should be noted that, in a specific embodiment, the standard reading speed of the user during the reading skill practice is analyzed by: the number of first choice is The reading speed of each tester is obtained and averaged to obtain the average reading speed of each tester. , according to the reading speed of each user in the reading skill practice process, the average reading speed of each user in history is obtained , and finally through the formula: , calculate the standard reading speed of the user during reading skill practice .

[0052] Extract the content retelling text and standard content text of the reading materials during the user's reading skill practice from the database, and count the total number of sentences in the standard content text , extract each sentence in the content retelling text of the reading material during the user's reading skill practice, compare the semantics with each sentence in the standard content text, and count the total number of sentences with similar semantics to the sentences in the standard content text, recorded as , through the formula: , calculate the comprehension index of users during reading skill practice .

[0053] Reading speed compliance index based on user reading skill practice , Comprehension Compliance Index , through the formula: , calculate the user's reading skills practice score ,in The concentration coefficient of users during reading skill practice.

[0054] It should be noted that, in a specific embodiment, the semantic similarity between two texts is determined by analyzing the sentence structure and syntax tree through a syntactic analysis tool, and using methods such as the shortest path or depth to calculate the structural similarity between the sentences.

[0055] In a specific embodiment of the present invention, the practice score of the user's writing skills is calculated based on the data in the user's writing practice process. The specific analysis method is as follows: extracting a list of writing sentence patterns and each sentence segment of the writing text in the user's writing skill practice process from a database, and recording each sentence segment of the writing text in the user's writing skill practice process as , , Indicates the number of each sentence segment of the writing text during the user's writing skill practice. Indicates the total number of sentences in the text written by the user during the writing skill practice. Match each sentence segment of the writing text during the user's writing skill practice with the writing sentence pattern list, and count the number of sentence patterns in the writing sentence pattern list contained in each sentence segment of the writing text during the user's writing skill practice, recorded as , through the formula: , calculate the sentence richness index during the user's skill practice ,in Represents the total number of standards of the sentence patterns in the writing sentence pattern list contained in the writing texts during the writing skill practice of the user stored in the database.

[0056] It should be noted that, in a specific embodiment, the standard total number of sentence patterns in the writing sentence pattern list included in the writing text during the user's writing skill practice is preset according to the sentence pattern requirements of the writing skill practice by English teaching experts, for example .

[0057] Extract each word in the text written by the user during the writing skill practice, compare it with the spelling of each standard word, and count the total number of words in the text written by the user during the writing skill practice , total number of wrong words , through the formula: , calculate the word correctness index of the text written during the user's skill practice .

[0058] Based on the sentence richness index and word correctness index of the writing text during the user's writing practice, the formula is: , calculate the user's writing skills practice score ,in The concentration coefficient during user writing skill practice.

[0059] The present invention starts from the four skills of English learning, comprehensively considers various data that affect the practice scores of users in the process of practicing each skill, and combines the concentration coefficient of users in the process of practicing each skill to analyze and obtain the practice scores of users in the process of practicing each skill. This comprehensive consideration method can accurately reflect the practice scores of users' English learning simulation, which is convenient for users to promptly discover defects in the practice process and make up for them in time to improve their English level.

[0060] In a specific embodiment of the present invention, the specific analysis method for calculating the improvement index of each skill of the user is as follows: based on the practice score of the user's listening skills , oral skills practice scoring , Reading Skills Practice Scoring , Writing Skills Practice Scoring , encapsulates the practice score data set of each user's skills , through the formula: , calculate the improvement index of each skill of the user at the current stage , , Indicates the number of each skill of the user, where Indicates the previous stage user extracted from the database A training and scoring dataset for 3 skills.

[0061] It should be noted that in a specific embodiment, for example, the user's listening skill practice score , the practice score of the user’s listening skills in the previous stage , then by the formula , calculate the improvement index of the user's listening skills at the current stage .

[0062] In a specific embodiment of the present invention, the fluctuation value of the skill improvement index of each user is calculated, and the specific analysis method is: based on the improvement index of each skill of the user at each historical stage and the improvement index of each skill of the user at the current stage, the improvement index of each skill of the user at each stage is combined to obtain , , Indicates the number of each stage of the user, Indicates the total number of user stages.

[0063] By formula: , calculate the fluctuation value of each skill improvement index of the user .

[0064] It should be noted that, in a specific embodiment, for example, the improvement index of each skill of the user at each stage is obtained. , , , , , through the formula: , calculate the fluctuation value of the user's listening skill improvement index .

[0065] In a specific embodiment of the present invention, the response measures for implementing the fluctuation values ​​of each skill improvement index of the user corresponding to the level interval are specifically analyzed as follows: based on the user's Fluctuation value of skill improvement index ,like In user The first interval of fluctuation value of the skill improvement index is used to determine the user's The fluctuation level of the skill improvement index is high, and the user's The difficulty level of practicing a skill is a response measure of one level.

[0066] It should be noted that, in a specific embodiment, the user The fluctuation value of the skill improvement index is high, indicating that the user If a skill is difficult to practice at certain stages, it is necessary to reduce the difficulty level of the practice by one level in the subsequent simulation practice.

[0067] The exercise difficulty coefficient is set according to an English learning simulation system based on artificial intelligence, and simulation practice modes with exercise difficulty coefficients of various levels can be selected.

[0068] like In user If the fluctuation value of the skill improvement index of the user is in the second interval, the fluctuation value level of each skill improvement index of the user is judged to be low fluctuation, and the user's skill improvement index is improved. The difficulty level of practicing a skill is a response measure of one level.

[0069] It should be noted that, in a specific embodiment, the user The fluctuation value of the skill improvement index is low, indicating that the user The progress of each skill is relatively stable. At some stages, the practice difficulty is too low and the improvement of practice effect is not satisfactory. It is necessary to increase the practice difficulty coefficient by one level in the subsequent simulation practice process.

[0070] It should be noted that, in a specific embodiment, the first interval and the second interval of the fluctuation value of each skill improvement index do not specifically refer to the importance level of the relationship, but are for the convenience of describing and distinguishing the fluctuation value intervals of each skill improvement index.

[0071] The first interval and the second interval of the fluctuation value of each skill improvement index are preset by English teaching experts based on the historical learning data of each user of the artificial intelligence English learning simulation system and the subsequent learning effects.

[0072] According to The response measures for the level intervals corresponding to the fluctuation values ​​of each skill improvement index of the users are implemented.

[0073] The present invention calculates the improvement index of each skill of the user based on the practice score of each skill at the current stage and the practice score of each skill at the previous stage, analyzes the improvement index of each skill of the user at each stage, and obtains the fluctuation value of the improvement index of each skill of the user. It can reflect whether the improvement effect of each skill of the user in a period of time is highly stable or highly unstable, so that the user can make timely adjustments to meet the needs of practicing various English skills, thereby improving the level of various English skills.

[0074] In a specific embodiment of the present invention, the comprehensive score of the user's practice effect is calculated by a specific analysis method: based on the practice score data set of each skill of the user , the improvement index of each user's skills , through the formula: , calculate the comprehensive score of the user's practice effect .

[0075] like If the comprehensive score of the user's practice effect is in the first interval, the user's practice effect is determined to be excellent.

[0076] like If the comprehensive score of the user's practice effect is in the second interval, it is determined that the user's practice effect is good.

[0077] like If the comprehensive score of the user's practice effect is in the third interval, the user's practice effect is judged to be unqualified.

[0078] The user's practice effect level is displayed through an artificial intelligence-based English learning simulation system.

[0079] It should be noted that, in a specific embodiment, the first interval, the second interval, and the third interval of the comprehensive score of the user's exercise effect do not specifically refer to the importance level in the relationship, but are for the convenience of describing and distinguishing the intervals of the comprehensive score of the user's exercise effect.

[0080] The intervals of the comprehensive scores of the user's practice effects are divided by English teaching experts according to the comprehensive scores of the historical practice effects of each user and the subsequent learning situation, and are stored in the database.

[0081] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A method for evaluating the practice effect of an English learning simulation system based on artificial intelligence, characterized in that: The method comprises: Step 1: Obtain user information: extract user information from the database, including the user's skill practice time and the practice score of each skill in the previous stage; Step 2: User concentration analysis: Obtain facial image videos and body posture image videos of users during the practice of each skill, and analyze the user's concentration coefficient during the practice of each skill; Step 3, analysis of the user's skill practice: obtaining various data during the user's skill practice process, and calculating the user's listening skill practice score based on the various data during the user's listening skill practice process; calculating the user's speaking skill practice score based on the various data during the user's speaking skill practice process; calculating the user's reading skill practice score based on the various data during the user's reading skill practice process; and calculating the user's writing skill practice score based on the various data during the user's writing practice process; Step 4: Analysis of the improvement index of each skill of the user: obtain the practice score dataset of each skill of the user in each historical stage, calculate the improvement index of each skill of the user based on the practice score dataset of each skill of the user in the last stage, obtain the improvement index of each skill of the user in each stage, and calculate the fluctuation value of the improvement index of each skill of the user; Step 5: Response to user practice effect: Implement response measures for the level intervals corresponding to the fluctuation values ​​of each skill improvement index of the user; Step 6: Comprehensive evaluation of user practice results: Based on the practice scores and improvement indexes of each skill, calculate the comprehensive score of the user's practice results.

2. The method for evaluating the practice effect of an English learning simulation system based on artificial intelligence according to claim 1, characterized in that: The specific analysis method for analyzing the concentration coefficient of the user during each skill practice is as follows: In view of the prominent feature of blinking that reflects concentration during listening skill practice, the frame processing interval of the facial video during the user's listening skill practice is set to , obtain each sample slice of the blinking action during the user's listening skill practice; Aiming at the general features that reflect concentration during listening skill practice: facial expressions and head postures, the frame processing interval of the facial video and body posture video during the user's listening skill practice is set to , obtaining sample slices of facial expressions and sample slices of head postures during the user's listening skill practice; For the prominent features that reflect the user's concentration during oral skill practice: mouth movements, general features: eyebrow movements, hand movements; for the prominent features that reflect the user's concentration during reading skill practice: eye focus points, general features: mouth movements, hand movements; for the prominent features that reflect the user's concentration during writing skill practice: body postures, general features: eye focus points, head postures, the facial videos and body posture videos of the user during each skill practice are processed at a frame processing interval of 100 according to the general features. , highlight the feature setting frame processing interval to be processed in the following manner; Get the sample slices of the user's skill practice process, input them into the convolutional neural network through computer vision technology, and get the user's concentration time during the skill practice process after processing. , , Indicates the number of each user's English skill, through the formula: , calculate the user's concentration coefficient during each skill practice ,in Represents the user number extracted from the database Skill practice time.

3. The method for evaluating the practice effect of an English learning simulation system based on artificial intelligence according to claim 2, characterized in that: The specific analysis method for calculating the listening skill practice score of the user based on the data during the listening skill practice of the user is as follows: Extract the total number of exercises completed by the user during the listening skill practice from the database , Total number of correct exercises , through the timing module, obtain the reaction time of each exercise during the user's listening skill practice ,in , Indicates the number of each exercise in the user's listening skills practice process. Indicates the total number of exercises during the user's listening skill practice; Extract the user's listening content vocabulary list from the database and mark the number of correct vocabulary The listening vocabulary list contains the total number of words in the listening content. ; Based on the concentration coefficient of the user's listening skills practice process , through the formula: , calculate the user's listening skill practice score ,in represents the standard reaction time of each exercise extracted from the database, Indicates the pre-set weight factor affecting the correctness of exercises during the user's listening skills practice. Indicates the influence weight factor of the marked correct vocabulary in the preset user listening skill practice process, Represents the preset weight factor affecting the response efficiency during the user's listening skill practice.

4. The method for evaluating the practice effect of an English learning simulation system based on artificial intelligence according to claim 3, characterized in that: The specific analysis method for calculating the practice score of the user's oral skills based on the data during the user's oral skills practice is as follows: Extract the audio of the user's oral skills practice from the database, first use the audio analysis program to obtain the user's speaking speed during the oral skills practice , Number of freezes , through the formula: , calculate the speed compliance index of the user's oral skills practice process ,in It represents the standard speaking speed of the user during the oral skill practice process extracted from the database; By formula: , calculate the fluency index of the user's oral skills practice process ,in Indicates the allowed number of pauses during the user's oral skills practice extracted from the database; Then, the audio during the user's oral skill practice is sliced ​​according to each word to obtain the actual audio waveform of each word during the user's oral skill practice, which is recorded as ,in , Indicates the number of each word during the user's oral skills practice. Represents the total number of words in the user's oral skills practice process, and obtains the standard audio waveform of each word in the user's oral skills practice process from the database, which is recorded as ; Compare the actual audio waveform of each word during the user's oral skill practice with the standard audio waveform, and count the total number of vowel parts in the actual audio waveform of each word during the user's oral skill practice , the total number of vowels in the standard audio waveform , analyze the similarity between the waveform of each vowel part in the actual audio waveform of each word and the waveform of each vowel part in the standard audio waveform during the user's oral skill practice, and record it as ,in , Indicates the number of each vowel part in the audio waveform of a word during the user's oral skill practice. Indicates the total number of vowel parts in the actual audio waveform of a word during the user's oral skill practice; By formula: , calculate the vowel matching coefficient during the user's oral skills practice ; According to the analysis method of vowel matching coefficient, the consonants in the user's oral skill practice process are analyzed to obtain the consonant matching coefficient in the user's oral skill practice process. ; By formula: , calculate the pronunciation accuracy index of the user during the oral skill practice ; Based on the speed conformity index, fluency conformity coefficient, and pronunciation accuracy index of the user's oral skills practice, the formula is: , calculate the user's speaking skill practice score ,in It is the user's concentration coefficient during the oral skill practice process.

5. The practice effect evaluation method of the English learning simulation system based on artificial intelligence according to claim 4 is characterized in that: The specific analysis method for calculating the practice score of the user's reading skills based on the data during the user's reading skills practice is as follows: Extract the total number of words in the reading material during the user's reading skill practice from the database Total time spent reading , through the formula: , calculate the reading speed of the user during reading skill practice , through the formula: , calculate the reading speed index of the user during reading skill practice ,in Indicates the standard reading speed of the user during reading skill practice stored in the database; Extract the content retelling text and standard content text of the reading materials during the user's reading skill practice from the database, and count the total number of sentences in the standard content text , extract each sentence in the content retelling text of the reading material during the user's reading skill practice, compare the semantics with each sentence in the standard content text, and count the total number of sentences with similar semantics to the sentences in the standard content text, recorded as , through the formula: , calculate the comprehension index of users during reading skill practice ; Reading speed compliance index based on user reading skill practice , Comprehension Compliance Index , through the formula: , calculate the user's reading skills practice score ,in The concentration coefficient of users during reading skill practice.

6. The method for evaluating the practice effect of an English learning simulation system based on artificial intelligence according to claim 5, characterized in that: The specific analysis method for calculating the practice score of the user's writing skills based on the data in the user's writing practice process is as follows: Extract the writing sentence pattern list and each sentence segment of the writing text during the user's writing skill practice from the database, and record each sentence segment of the writing text during the user's writing skill practice as , , Indicates the number of each sentence segment of the writing text during the user's writing skill practice. Indicates the total number of sentences in the writing text during the user's writing skill practice; Match each sentence segment of the writing text during the user's writing skill practice with the writing sentence pattern list, and count the number of sentence patterns in the writing sentence pattern list contained in each sentence segment of the writing text during the user's writing skill practice, recorded as , through the formula: , calculate the sentence richness index during the user's skill practice ,in It represents the standard total number of sentence patterns in the writing sentence pattern list contained in the writing text during the writing skill practice of the user stored in the database; Extract each word in the text written by the user during the writing skill practice, compare it with the spelling of each standard word, and count the total number of words in the text written by the user during the writing skill practice , total number of wrong words , through the formula: , calculate the word correctness index of the text written during the user's skill practice ; Sentence richness index based on user writing practice , Word Correctness Index of Writing Text , through the formula: , calculate the user's writing skills practice score ,in The concentration coefficient during user writing skill practice.

7. The method for evaluating the practice effect of an English learning simulation system based on artificial intelligence according to claim 6, characterized in that: The specific analysis method for calculating the improvement index of each skill of the user is as follows: Practice scoring based on user listening skills , oral skills practice scoring , Reading Skills Practice Scoring , Writing Skills Practice Scoring , encapsulates the practice score data set of each user's skills , through the formula: , calculate the improvement index of each skill of the user at the current stage , , Indicates the number of each skill of the user, where Indicates the previous stage user extracted from the database A training and scoring dataset for 3 skills.

8. The method for evaluating the practice effect of an English learning simulation system based on artificial intelligence according to claim 7, characterized in that: The specific analysis method for calculating the fluctuation value of each skill improvement index of the user is as follows: Based on the improvement index of each skill at each stage of the user's history and the improvement index of each skill at the current stage, the improvement index of each skill at each stage of the user is combined to obtain the improvement index of each skill at each stage of the user , , Indicates the number of each stage of the user, Indicates the total number of user stages; By formula: , calculate the fluctuation value of each skill improvement index of the user .

9. The method for evaluating the practice effect of an English learning simulation system based on artificial intelligence according to claim 8, characterized in that: The specific analysis method of implementing the response measures for the level intervals corresponding to the fluctuation values ​​of each skill improvement index of the user is as follows: Based on user Fluctuation value of skill improvement index ,like In user The first interval of fluctuation value of the skill improvement index is used to determine the user's The fluctuation level of the skill improvement index is high, and the user's The difficulty level of practicing a skill is a level of response measure; like In user If the fluctuation value of the skill improvement index of the user is in the second interval, the fluctuation value level of each skill improvement index of the user is judged to be low fluctuation, and the user's skill improvement index is improved. The difficulty level of practicing a skill is a level of response measure; According to The response measures for the level intervals corresponding to the fluctuation values ​​of each skill improvement index of the users are implemented.

10. The method for evaluating the practice effect of an English learning simulation system based on artificial intelligence according to claim 9, characterized in that: The specific analysis method for calculating the comprehensive score of the user's exercise effect is as follows: Practice rating dataset based on user skills , the improvement index of each user's skills , through the formula: , calculate the comprehensive score of the user's practice effect ; like If the user's practice effect is in the first interval of the comprehensive score, the user's practice effect level is determined to be excellent; like If the user's practice effect is in the second interval of the comprehensive score of the user's practice effect, the user's practice effect level is determined to be good; like If the user's practice effect is in the third range of the comprehensive score, the user's practice effect level is judged to be unqualified; The user's practice effect level is displayed through an artificial intelligence-based English learning simulation system.

Images