Method and system for evaluating stability of lexical long-term memory based on real-time feedback

Abstract

This invention discloses a method and system for assessing the stability of long-term vocabulary memory based on real-time feedback, relating to the field of foreign language vocabulary learning technology. The method first collects user review and answer data, including time, answers, and results. Based on answer time thresholds, it categorizes ambiguous and regular vocabulary and records the number of reviews. The system dynamically calculates priorities based on vocabulary entry time and length, generating an initial push sequence. During the review process, it monitors cumulative answer time and error counts in real time, dynamically adjusts the sequence order, and triggers secondary review requests for frequently misspelled or ambiguous words. Based on edit distance algorithms and answer matching, it intelligently pushes confusing words for reinforcement practice to weak vocabulary. This invention generates personalized vocabulary review paths driven by real-time feedback data, significantly reducing confusion between similar words, promoting the solid formation of long-term memory, and improving learning efficiency.

Description

Technical Field

[0001] This invention relates to the field of foreign language vocabulary learning technology, specifically to a method and system for assessing the stability of long-term vocabulary memory based on real-time feedback. Background Technology

[0002] In scenarios requiring extensive vocabulary accumulation, such as language learning and exam preparation for specialized knowledge, long-term, stable vocabulary memorization is a crucial prerequisite for ensuring learning effectiveness. However, traditional vocabulary review models generally suffer from problems such as rigid logic in generating study plans and a lack of flexibility in review delivery: on the one hand, the generation of review paths often focuses on single test-taking results, easily misjudging vaguely remembered or ambiguous vocabulary as mastered, thus creating knowledge gaps; on the other hand, the delivery of review content lacks dynamic adaptation to the learner's individual memory state, failing to address the confusion caused by vocabulary morphology or semantic associations, and struggling to adjust the review pace according to the real-time status during the learning process, ultimately leading to low review efficiency and difficulty in supporting long-term, stable vocabulary memorization. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for evaluating the stability of long-term vocabulary memory based on real-time feedback, so as to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: A method for assessing the stability of long-term vocabulary memory based on real-time feedback, comprising the following steps: Step S1: Users review vocabulary by answering questions; collect the user's answering time, answers, and results; when the vocabulary is first pushed to the user, classify the vocabulary type based on the answering time and results, and record the number of times the vocabulary is reviewed; establish a database, associate the vocabulary type with the answer and store it in the vocabulary database, and associate it with the corresponding vocabulary in the vocabulary database, while recording the entry time at this time; Furthermore, step S1 includes: The vocabulary consists of letters and syllables; A database is established, which includes a main vocabulary database, a review vocabulary database, and a simple vocabulary database. The main vocabulary database stores all vocabulary words, and the review vocabulary database stores review vocabulary words for vocabulary review. Vocabulary words other than those in the review vocabulary database and the simple vocabulary database are extracted from the main vocabulary database as learning words, and the user can choose to store the learning words in the review vocabulary database or the simple vocabulary database. The questions are in the form of multiple choice. Each question includes an image and its Chinese definition. The vocabulary corresponding to the question is the vocabulary for this review session. The vocabulary for this review session and several other words are presented as options, and the user selects one word from the options as the answer. The answer results are divided into correct and incorrect selections. When the answer is the same as the vocabulary words reviewed in this session, the answer result is considered to be correct; otherwise, the answer result is considered to be incorrect. When the question and options are pushed to the user, the answering process begins. The time interval from the start of the answering process to the generation of the answer result is recorded as the user's answering time. A fuzzy time threshold is preset. When the answering time exceeds the fuzzy time threshold, the corresponding review vocabulary is considered as fuzzy vocabulary. Otherwise, the corresponding review vocabulary is considered as regular vocabulary. Based on the answer result, regular vocabulary is divided into correctly selected vocabulary and incorrectly selected vocabulary. The vocabulary types include fuzzy vocabulary and regular vocabulary.

[0005] Step S2: Obtain the basic data of the review vocabulary in the review vocabulary database. When a user sends a request to obtain the review vocabulary, extract a preset number of review vocabulary according to the entry time and basic data, generate a push sequence, and push the review vocabulary to the user in sequence according to the push sequence. Furthermore, step S2 includes: The basic data includes the word length of all review words in the review vocabulary, where the word length is the number of letters that make up the word; Calculate the priority of vocabulary review based on vocabulary length and entry time into the database: ; in This indicates the priority of vocabulary review. This represents the time interval between when the vocabulary words were added to the database and when the user sent a request to retrieve them. This represents the length of the vocabulary to be reviewed. This represents the highest priority in the review vocabulary. and These represent preset weighting coefficients; If the number of times a vocabulary word is reviewed is 0, the priority value of that vocabulary word is set to 1; the minimum priority value of a vocabulary word is 0. Set a priority threshold. When extracting review words from the review vocabulary library, prioritize extracting review words with a priority higher than the priority threshold. Then, extract a preset number of review words in order of entry time from earliest to latest. After extraction, randomly shuffle the order of the extracted review words and generate the push order.

[0006] Step S3: When review words with more than 1 review times are pushed to the user, the user's cumulative answer time and answer results are collected, the push sequence of answers is dynamically adjusted according to the cumulative answer time and answer results, and a second review request is sent to the user. Furthermore, step S3 includes: Step S3-1: The user answers questions according to the push sequence and records the number of times each review word is selected correctly. When a review word is selected incorrectly, or selected correctly but the number of correct selections does not exceed the preset correct selection threshold, the review word is reproduced in the subsequent random order of the push sequence, and the subsequent review words are arranged sequentially. Otherwise, the review word is considered to have been reviewed, and its review count in the review word library is incremented by one. Step S3-2: Collect the user's cumulative answer time in real time. The cumulative answer time is the sum of all answer times of the user in the current push sequence. When the cumulative answer time exceeds the preset fatigue time threshold, the subsequent review words in the push sequence are arranged in order of priority so that the user can review the review words in order of priority. Step S3-3: Record the number of times each review word is selected incorrectly. When the number of times a review word is selected incorrectly exceeds the preset error threshold, or when the incorrectly selected review word is a vague word, the review word is used as a second review word, and a second review request is sent to the user. The result of the second review request is obtained. The result of the second review request includes starting the second review and skipping the second review.

[0007] Step S4: Match the confusing words for each word with the answer based on the basic data; obtain the confusing words in the database, extract the confusing words based on the results of the secondary review request, and push secondary review vocabulary to the user for secondary review. Furthermore, step S4 includes: All words in the database are divided into length levels according to word length. The edit distance algorithm is used to calculate the edit distance between different words in the same length level. A preset distance threshold is set, and different words whose edit distance is less than the distance threshold are considered to be confused words. Simultaneously, extract the answers for all words in the database, and distinguish between different words with the same answer as each other; When the result of the second review request is to start the second review, a preset number of confused words from the second review vocabulary are randomly selected, and the second review is conducted by answering questions. When the result of the secondary review request is to skip the secondary review, the review vocabulary will continue to be pushed to the user according to the push sequence.

[0008] A vocabulary long-term memory stability assessment system based on real-time feedback. The system includes a vocabulary basic data module, a review vocabulary push module, a push sequence adjustment module, and a secondary review module. The vocabulary basic data module is used to collect users' answering time, answers and answering results, classify vocabulary types, record the number of reviews and establish a database; The vocabulary review push module is used to obtain the basic data of the vocabulary review in the vocabulary review database, extract the vocabulary review according to the entry time and basic data, generate a push sequence, and push the vocabulary review to the user. The push sequence adjustment module is used to collect the user's cumulative answering time and answering results, dynamically adjust the push sequence, and trigger a second review request based on the number of errors. The secondary review module is used to match the confusing words for each word, extract the confusing words based on the secondary review request results, and provide users with secondary review vocabulary. The output of the vocabulary base data module is connected to the input of the review vocabulary push module; the output of the review vocabulary push module is connected to the input of the push sequence adjustment module; and the output of the push sequence adjustment module is connected to the input of the secondary review module.

[0009] The vocabulary basic data module also includes a question answer data collection unit and a vocabulary data storage unit; The answer data collection unit is used to collect the user's answer time, answer and answer result, classify the vocabulary of this review based on the answer time and answer result, and record the number of times the vocabulary is reviewed; The vocabulary data storage unit is used to establish a database containing a main vocabulary database, a review vocabulary database, and a simple vocabulary database. It associates vocabulary types with answers and stores them in the review vocabulary database, as well as associating them with the corresponding review vocabulary words, and records the entry time.

[0010] The vocabulary review push module also includes a vocabulary data acquisition unit and a push sequence generation unit; The vocabulary data acquisition unit is used to acquire basic data of the review vocabulary in the review vocabulary library. The basic data includes the vocabulary length of all review vocabulary in the review vocabulary library. The push sequence generation unit is used to extract a preset number of review words according to the entry time and basic data when a user sends a request to obtain review words, randomly shuffle the order, generate a push sequence, and push them out in sequence.

[0011] The push sequence adjustment module also includes a cumulative data collection unit and a secondary request triggering unit; The cumulative data acquisition unit is used to collect the user's cumulative answering time and the answering result for each review vocabulary word, and record the number of times the correct answer was selected and the number of times the incorrect answer was selected. The secondary request triggering unit is used to dynamically adjust the push sequence based on the cumulative answering time and answering results. When the number of errors exceeds the error threshold, or the incorrect word is an ambiguous word, a secondary review request is sent to the user.

[0012] The secondary review module also includes a vocabulary confusion matching unit and a secondary review execution unit; The vocabulary confusion matching unit is used to classify words into levels according to word length and match confused words in all words in the database through an edit distance algorithm or answer. The secondary review execution unit is used to randomly select a preset number of confusing words based on the results of the secondary review request, and conduct secondary review by answering questions, or continue to push vocabulary in the original push sequence.

[0013] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention dynamically categorizes vocabulary types and establishes a multi-dimensional database by collecting user answer time and result data in real time. It then calculates priority based on the entry time and vocabulary length, enabling intelligent generation and immediate adjustment of review sequences. This real-time feedback-based review path generation mechanism accurately identifies vaguely remembered words, effectively avoiding the biases caused by traditional methods that rely solely on answer results, and significantly improving the accuracy of vocabulary review path generation.

[0014] 2. This invention monitors cumulative answering time and error frequency, automatically switching to a priority-based review mode when the user shows signs of fatigue, and triggering a secondary review mechanism for frequently misspelled words and ambiguous words. This dynamic adaptation strategy ensures real-time matching between the review pace and the learner's state, while also reinforcing weak points in memory, thereby reducing cognitive load while improving consolidation efficiency.

[0015] 3. This invention automatically generates a confusion word library through edit distance algorithm and answer matching technology. When a user makes consecutive mistakes on a specific word, it intelligently pushes words with easily confused forms or meanings for comparative practice. This targeted secondary review can directly solve the confusion problem in vocabulary memorization, strengthen the ability to distinguish words, reduce the interference of forgetting similar words from the root, and promote the stable construction of long-term memory networks. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the vocabulary long-term memory stability assessment method based on real-time feedback of the present invention. Figure 2 This is a schematic diagram of the structure of the vocabulary long-term memory stability assessment system based on real-time feedback according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1: As Figure 1As shown, this invention provides a technical solution: a method for evaluating the stability of long-term vocabulary memory based on real-time feedback. This method includes the following steps: Step S1: Users review vocabulary by answering questions; collect the user's answering time, answers, and results; when the vocabulary is first pushed to the user, classify the vocabulary type based on the answering time and results, and record the number of times the vocabulary is reviewed; establish a database, associate the vocabulary type with the answer and store it in the vocabulary database, and associate it with the corresponding vocabulary in the vocabulary database, while recording the entry time at this time; Step S1 includes: A database is established, which includes a main vocabulary database, a review vocabulary database, and a simple vocabulary database. The main vocabulary database stores all vocabulary words, and the review vocabulary database stores review vocabulary words for vocabulary review. Vocabulary words other than those in the review vocabulary database and the simple vocabulary database are extracted from the main vocabulary database as learning words, and the user can choose to store the learning words in the review vocabulary database or the simple vocabulary database. The questions are in the form of multiple choice. Each question includes an image and its Chinese definition. The vocabulary corresponding to the question is the vocabulary for this review session. The vocabulary for this review session and several other words are presented as options, and the user selects one word from the options as the answer. The answer results are divided into correct and incorrect selections. When the answer is the same as the vocabulary words reviewed in this session, the answer result is considered to be correct; otherwise, the answer result is considered to be incorrect. When the question and options are pushed to the user, the answering process begins. The time interval from the start of the answering process to the generation of the answer result is recorded as the user's answering time. A fuzzy time threshold is preset. When the answering time exceeds the fuzzy time threshold, the corresponding review vocabulary is considered as fuzzy vocabulary. Otherwise, the corresponding review vocabulary is considered as regular vocabulary. Based on the answer result, regular vocabulary is divided into correctly selected vocabulary and incorrectly selected vocabulary. The vocabulary types include fuzzy vocabulary and regular vocabulary.

[0019] Step S2: Obtain the basic data of the review vocabulary in the review vocabulary database. When a user sends a request to obtain the review vocabulary, extract a preset number of review vocabulary according to the entry time and basic data, generate a push sequence, and push the review vocabulary to the user in sequence according to the push sequence. Step S2 includes: The basic data includes the word length of all review words in the review vocabulary, where the word length is the number of letters that make up the word; Calculate the priority of vocabulary review based on vocabulary length and entry time into the database: ; in This indicates the priority of vocabulary review. This represents the time interval between when the vocabulary words were added to the database and when the user sent a request to retrieve them. This represents the length of the vocabulary to be reviewed. This represents the highest priority in the review vocabulary. and These represent preset weighting coefficients; If the number of times a vocabulary word is reviewed is 0, the priority value of that vocabulary word is set to 1; the minimum priority value of a vocabulary word is 0. Set a priority threshold. When extracting review words from the review vocabulary library, prioritize extracting review words with a priority higher than the priority threshold. Then, extract a preset number of review words in order of entry time from earliest to latest. After extraction, randomly shuffle the order of the extracted review words and generate the push order.

[0020] Step S3: When review words with more than 1 review times are pushed to the user, the user's cumulative answer time and answer results are collected, the push sequence of answers is dynamically adjusted according to the cumulative answer time and answer results, and a second review request is sent to the user. Step S3 includes: Step S3-1: The user answers questions according to the push sequence and records the number of times each review word is selected correctly. When a review word is selected incorrectly, or selected correctly but the number of correct selections does not exceed the preset correct selection threshold, the review word is reproduced in the subsequent random order of the push sequence, and the subsequent review words are arranged sequentially. Otherwise, the review word is considered to have been reviewed, and its review count in the review word library is incremented by one. Step S3-2: Collect the user's cumulative answer time in real time. The cumulative answer time is the sum of all answer times of the user in the current push sequence. When the cumulative answer time exceeds the preset fatigue time threshold, the subsequent review words in the push sequence are arranged in order of priority so that the user can review the review words in order of priority. Step S3-3: Record the number of times each review word is selected incorrectly. When the number of times a review word is selected incorrectly exceeds the preset error threshold, or when the incorrectly selected review word is an ambiguous word, the review word is used as a second review word, and a second review request is sent to the user. The result of the second review request is obtained. The result of the second review request includes starting the second review and skipping the second review.

[0021] Step S4: Match the confusing words for each word with the answer based on the basic data; obtain the confusing words in the database, extract the confusing words based on the results of the secondary review request, and push secondary review vocabulary to the user for secondary review. Step S4 includes: All words in the database are divided into length levels according to word length. The edit distance algorithm is used to calculate the edit distance between different words in the same length level. A preset distance threshold is set, and different words whose edit distance is less than the distance threshold are considered to be confused words. Simultaneously, extract the answers for all words in the database, and distinguish between different words with the same answer as each other; When the result of the second review request is to start the second review, a preset number of confused words from the second review vocabulary are randomly selected, and the second review is conducted by answering questions. When the result of the secondary review request is to skip the secondary review, the review vocabulary will continue to be pushed to the user according to the push sequence.

[0022] For example: Database and preset parameter settings: The main vocabulary database contains 1,000 English words, covering daily learning and basic exam preparation needs; The simple vocabulary database contains 50 basic words, such as a, an, the, etc. These words are generally familiar to users and do not require repeated review. The fuzzy time threshold is set to 3 seconds. For words that are well-memorized, there is no need to hesitate for a long time when answering questions. If the hesitation exceeds 3 seconds, it can be determined that the memory is fuzzy. The threshold for the number of correct answers is set to 2. A single correct answer may be due to chance, while two correct answers are a better proof of proficiency. The fatigue time threshold is set at 15 minutes, which is consistent with the fatigue cycle of most learners who answer questions continuously. The error threshold is set at 2. Two errors indicate weak vocabulary memorization, requiring targeted reinforcement. Maximum priority value R max =10, which makes it easier to quantify priorities and makes the calculation results more intuitive; With a weighting coefficient α=0.1, the time interval between entries has a weaker impact on review priority than the length of vocabulary; with a weighting coefficient β=0.2, longer vocabulary is more difficult to memorize and needs to be given a higher weight. The priority threshold is set to 0.5, and the decimal priority calculated by the formula is used to filter high-urgency words. Each time, 10 vocabulary words are selected for review, balancing review efficiency and memorization effectiveness. Vocabulary length levels are divided into three levels: Level 1 consists of 3 letters or less, Level 2 consists of 4-6 letters, and Level 3 consists of 7 letters or more, which facilitates matching words with similar forms. The edit distance threshold is set to 1. Words with an edit distance of 1 are highly similar in form, which can easily cause memory confusion.

[0023] Initial quiz answer and data storage: Users select "bread" from the main vocabulary as a learning word and store it in the review vocabulary. The system presents a question with a picture of bread and the Chinese definition of "bread". The options are bread, break, cake, and milk. The user has 2.5 seconds to answer. If the user selects bread, the answer is correct. "bread" is categorized as a correct choice word in the regular vocabulary, with a review count of 1; the entry time is recorded as 14:00 on 5th of month X, 2024. Related data will be stored in the review vocabulary database after being associated with this word.

[0024] Priority calculation and push sequence generation: The word length of "bread" is 5; On May 5, 2024 at 14:30, a user sent a review request. The time interval between the entry time and the request was ΔT = 30 minutes. According to the priority formula, the priority of "bread" is 0.6. The review vocabulary currently contains 10 common words: bread, milk, egg, desk, pen, book, tree, bird, fish, and apple. Their priorities are all higher than the threshold of 0.5. After being randomly shuffled from the entry time in the database from morning to night, the following push sequence is generated: egg, bread, pen, desk, milk, book, tree, bird, fish, apple. Among them, the priority of pen is 0.66, the priority of desk is 0.66, and the priority of apple is 0.7.

[0025] Dynamically adjusted push sequence: Users answer questions in sequence. If egg answers correctly, the number of correct answers is recorded as 1. If you answer the bread question correctly and accumulate 2 correct answers, you are considered to have completed the review, and the review count will be updated to 2. The pen was answered incorrectly, and the number of correct answers was recorded as 0. It was then repeated in the 8th position of the sequence, and the subsequent words were shifted to the next position. The new sequence is egg, bread, desk, milk, book, tree, bird, pen, fish, apple; The number of correct answers to the questions on the desk is recorded as 0. Milk answered the question correctly, and the number of correct answers is recorded as 1. The book answered the question correctly, and the number of correct answers is recorded as 1. The tree question was answered incorrectly, and the number of incorrect answers has accumulated to 2, triggering a second review request. At this point, the user's cumulative answering time has reached 17 minutes, exceeding the fatigue time threshold. The subsequent vocabulary words are then sorted by priority (apple-0.7, pen-0.66, fish-0.62), and the second half of the adjusted sequence is tree, apple, pen, fish. The user selects "Start Second Review". Confusion word matching and secondary review: The word "tree" has a length of 4, which is classified as a level 2 length level; the confusion word "free" was matched by the edit distance algorithm; at the same time, the common word "plant" that is easily confused with the answer was extracted; Two confusing words are randomly selected to generate a second review question, which includes a tree image and the Chinese definition of "tree". The options are tree, free, plant, and flower. The user answers by selecting tree, and the result is correct, thus completing the second review.

[0026] The overall process is closed-loop: After the user completes the second review, the system continues to push apple, pen, and fish according to the adjusted push sequence; the user completes the remaining vocabulary questions in sequence, and the system records the answer data of each vocabulary in real time; for words such as pen and desk that do not reach the correct number of times threshold, they are included in the push sequence again in subsequent reviews according to priority, forming a complete process of data collection, sequence generation, dynamic adjustment, secondary consolidation, and cyclical review, realizing the targeted generation of vocabulary memorization paths.

[0027] Example 2: Figure 2 As shown, the present invention provides a vocabulary long-term memory stability assessment system based on real-time feedback. The system includes a vocabulary basic data module, a review vocabulary push module, a push sequence adjustment module, and a secondary review module. The vocabulary basic data module is used to collect users' answering time, answers and answering results, classify vocabulary types, record the number of reviews and establish a database; The vocabulary review push module is used to obtain the basic data of the vocabulary review in the vocabulary review database, extract the vocabulary review according to the entry time and basic data, generate a push sequence, and push the vocabulary review to the user. The push sequence adjustment module is used to collect the user's cumulative answering time and answering results, dynamically adjust the push sequence, and trigger a second review request based on the number of errors. The secondary review module is used to match the confusing words for each word, extract the confusing words based on the secondary review request results, and provide users with secondary review vocabulary. The output of the vocabulary base data module is connected to the input of the review vocabulary push module; the output of the review vocabulary push module is connected to the input of the push sequence adjustment module; and the output of the push sequence adjustment module is connected to the input of the secondary review module.

[0028] The vocabulary basic data module also includes a question answer data collection unit and a vocabulary data storage unit; The answer data collection unit is used to collect the user's answer time, answer and answer result, classify the vocabulary of this review based on the answer time and answer result, and record the number of times the vocabulary is reviewed; The vocabulary data storage unit is used to establish a database containing a main vocabulary database, a review vocabulary database, and a simple vocabulary database. It associates vocabulary types with answers and stores them in the review vocabulary database, as well as associating them with the corresponding review vocabulary words, and records the entry time.

[0029] The vocabulary review push module also includes a vocabulary data acquisition unit and a push sequence generation unit; The vocabulary data acquisition unit is used to acquire basic data of the review vocabulary in the review vocabulary library. The basic data includes the vocabulary length of all review vocabulary in the review vocabulary library. The push sequence generation unit is used to extract a preset number of review words according to the entry time and basic data when a user sends a request to obtain review words, randomly shuffle the order, generate a push sequence, and push them out in sequence.

[0030] The push sequence adjustment module also includes a cumulative data collection unit and a secondary request triggering unit; The cumulative data acquisition unit is used to collect the user's cumulative answering time and the answering result for each review vocabulary word, and record the number of times the correct answer was selected and the number of times the incorrect answer was selected. The secondary request triggering unit is used to dynamically adjust the push sequence based on the cumulative answering time and answering results. When the number of errors exceeds the error threshold, or the incorrect word is an ambiguous word, a secondary review request is sent to the user.

[0031] The secondary review module also includes a vocabulary confusion matching unit and a secondary review execution unit; The vocabulary confusion matching unit is used to classify words into levels according to word length and match confused words in all words in the database through an edit distance algorithm or answer. The secondary review execution unit is used to randomly select a preset number of confusing words based on the results of the secondary review request, and conduct secondary review by answering questions, or continue to push vocabulary in the original push sequence.

[0032] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for assessing the stability of long-term vocabulary memory based on real-time feedback, characterized by: The method includes the following steps: Step S1: Users review vocabulary by answering questions; collect the user's answering time, answers, and results; when the vocabulary is first pushed to the user, classify the vocabulary type based on the answering time and results, and record the number of times the vocabulary is reviewed; establish a database, associate the vocabulary type with the answer and store it in the vocabulary database, and associate it with the corresponding vocabulary in the vocabulary database, while recording the entry time at this time; Step S2: Obtain the basic data of the review vocabulary in the review vocabulary database. When a user sends a request to obtain the review vocabulary, extract a preset number of review vocabulary according to the entry time and basic data, generate a push sequence, and push the review vocabulary to the user in sequence according to the push sequence. Step S3: When review words with more than 1 review times are pushed to the user, the user's cumulative answer time and answer results are collected, the push sequence of answers is dynamically adjusted according to the cumulative answer time and answer results, and a second review request is sent to the user. Step S4: Match the confused words for each word with the answer based on the basic data; obtain the confused words in the database, extract the confused words based on the results of the secondary review request, and push secondary review vocabulary to the user for secondary review.

2. The method for assessing the stability of long-term vocabulary memory based on real-time feedback according to claim 1, characterized in that: Step S1 includes: A database is established, which includes a main vocabulary database, a review vocabulary database, and a simple vocabulary database. The main vocabulary database stores all vocabulary words, and the review vocabulary database stores review vocabulary words for vocabulary review. Vocabulary words other than those in the review vocabulary database and the simple vocabulary database are extracted from the main vocabulary database as learning words, and the user can choose to store the learning words in the review vocabulary database or the simple vocabulary database. The questions are in the form of multiple choice. Each question includes an image and its Chinese definition. The vocabulary corresponding to the question is the vocabulary for this review session. The vocabulary for this review session and several other words are presented as options, and the user selects one word from the options as the answer. The answer results are divided into correct and incorrect selections. When the answer is the same as the vocabulary words reviewed in this session, the answer result is considered to be correct; otherwise, the answer result is considered to be incorrect. When the question and options are pushed to the user, the answering process begins. The time interval from the start of the answering process to the generation of the answer result is recorded as the user's answering time. A fuzzy time threshold is preset. When the answering time exceeds the fuzzy time threshold, the corresponding review vocabulary is considered as fuzzy vocabulary. Otherwise, the corresponding review vocabulary is considered as regular vocabulary. Based on the answer result, regular vocabulary is divided into correctly selected vocabulary and incorrectly selected vocabulary. The vocabulary types include fuzzy vocabulary and regular vocabulary.

3. The method for assessing the stability of long-term vocabulary memory based on real-time feedback according to claim 2, characterized in that: Step S2 includes: The basic data includes the word length of all review words in the review vocabulary, where the word length is the number of letters that make up the word; Calculate the priority of vocabulary review based on vocabulary length and entry time into the database: ； in This indicates the priority of vocabulary review. This represents the time interval between when the vocabulary words were added to the database and when the user sent a request to retrieve them. This represents the length of the vocabulary to be reviewed. This represents the highest priority in the review vocabulary. and These represent preset weighting coefficients; If the number of times a vocabulary word is reviewed is 0, the priority value of that vocabulary word is set to 1; the minimum priority value of a vocabulary word is 0. Set a priority threshold. When extracting review words from the review vocabulary library, prioritize extracting review words with a priority higher than the priority threshold. Then, extract a preset number of review words in order of entry time from earliest to latest. After extraction, randomly shuffle the order of the extracted review words and generate the push order.

4. The method for assessing the stability of long-term vocabulary memory based on real-time feedback according to claim 3, characterized in that: Step S3 includes: Step S3-1: The user answers questions according to the push sequence and records the number of times each review word is selected correctly. When a review word is selected incorrectly, or selected correctly but the number of correct selections does not exceed the preset correct selection threshold, the review word is reproduced in the subsequent random order of the push sequence, and the subsequent review words are arranged sequentially. Otherwise, the review word is considered to have been reviewed, and its review count in the review word library is incremented by one. Step S3-2: Collect the user's cumulative answer time in real time. The cumulative answer time is the sum of all answer times of the user in the current push sequence. When the cumulative answer time exceeds the preset fatigue time threshold, the subsequent review words in the push sequence are arranged in order of priority so that the user can review the review words in order of priority. Step S3-3: Record the number of times each review word is selected incorrectly. When the number of times a review word is selected incorrectly exceeds the preset error threshold, or when the incorrectly selected review word is a vague word, the review word is used as a second review word, and a second review request is sent to the user. The result of the second review request is obtained. The result of the second review request includes starting the second review and skipping the second review.

5. The method for assessing the stability of long-term vocabulary memory based on real-time feedback according to claim 4, characterized in that: Step S4 includes: All words in the database are divided into length levels according to word length. The edit distance algorithm is used to calculate the edit distance between different words in the same length level. A preset distance threshold is set, and different words whose edit distance is less than the distance threshold are considered to be confused words. Simultaneously, extract the answers for all words in the database, and distinguish between different words with the same answer as each other; When the result of the second review request is to start the second review, a preset number of confused words from the second review vocabulary are randomly selected, and the second review is conducted by answering questions. When the result of the secondary review request is to skip the secondary review, the review vocabulary will continue to be pushed to the user according to the push sequence.

6. A vocabulary long-term memory stability assessment system based on real-time feedback, applied to the vocabulary long-term memory stability assessment method based on real-time feedback as described in any one of claims 1-5, characterized in that: The system includes a vocabulary basic data module, a vocabulary review push module, a push sequence adjustment module, and a secondary review module; The vocabulary basic data module is used to collect users' answering time, answers and answering results, classify vocabulary types, record the number of reviews and establish a database; The vocabulary review push module is used to obtain the basic data of the vocabulary review in the vocabulary review database, extract the vocabulary review according to the entry time and basic data, generate a push sequence, and push the vocabulary review to the user. The push sequence adjustment module is used to collect the user's cumulative answering time and answering results, dynamically adjust the push sequence, and trigger a second review request based on the number of errors. The secondary review module is used to match the confusing words for each word, extract the confusing words based on the secondary review request results, and provide users with secondary review vocabulary. The output of the vocabulary base data module is connected to the input of the review vocabulary push module; the output of the review vocabulary push module is connected to the input of the push sequence adjustment module; and the output of the push sequence adjustment module is connected to the input of the secondary review module.

7. The vocabulary long-term memory stability assessment system based on real-time feedback according to claim 6, characterized in that: The vocabulary basic data module also includes a question answer data collection unit and a vocabulary data storage unit; The answer data collection unit is used to collect the user's answer time, answer and answer result, classify the vocabulary of this review based on the answer time and answer result, and record the number of times the vocabulary is reviewed; The vocabulary data storage unit is used to establish a database containing a main vocabulary database, a review vocabulary database, and a simple vocabulary database. It associates vocabulary types with answers and stores them in the review vocabulary database, as well as associating them with the corresponding review vocabulary words, and records the entry time.

8. The vocabulary long-term memory stability assessment system based on real-time feedback according to claim 6, characterized in that: The vocabulary review push module also includes a vocabulary data acquisition unit and a push sequence generation unit; The vocabulary data acquisition unit is used to acquire basic data of the review vocabulary in the review vocabulary library. The basic data includes the vocabulary length of all review vocabulary in the review vocabulary library. The push sequence generation unit is used to extract a preset number of review words according to the entry time and basic data when a user sends a request to obtain review words, randomly shuffle the order, generate a push sequence, and push them out in sequence.

9. The vocabulary long-term memory stability assessment system based on real-time feedback according to claim 6, characterized in that: The push sequence adjustment module also includes a cumulative data collection unit and a secondary request triggering unit; The cumulative data acquisition unit is used to collect the user's cumulative answering time and the answering result for each review vocabulary word, and record the number of times the correct answer was selected and the number of times the incorrect answer was selected. The secondary request triggering unit is used to dynamically adjust the push sequence based on the cumulative answering time and answering results. When the number of errors exceeds the error threshold, or the incorrect word is an ambiguous word, a secondary review request is sent to the user.

10. The vocabulary long-term memory stability assessment system based on real-time feedback according to claim 6, characterized in that: The secondary review module also includes a vocabulary confusion matching unit and a secondary review execution unit; The vocabulary confusion matching unit is used to classify words into levels according to word length and match confused words in all words in the database through an edit distance algorithm or answer. The secondary review execution unit is used to randomly select a preset number of confusing words based on the results of the secondary review request, and conduct secondary review by answering questions, or continue to push vocabulary in the original push sequence.

Images