A vocabulary level test processing method and system based on reading comprehension exercises

Abstract

This invention proposes a vocabulary proficiency testing method and system based on reading comprehension practice, including the following steps: Reading text preprocessing step: For each reading comprehension text, the server calculates the difficulty of each word w in the current text; then, it sorts the words in the text in descending order of difficulty and counts their frequency; providing test content to the user and accumulating difficult words; Vocabulary cognition level assessment step: After the user has accumulated a predetermined number of reading comprehension questions, all the difficult words encountered are averaged using their frequency percentage in the vocabulary frequency table WordFreq as a coefficient to obtain the average frequency percentage information. Then, the corresponding word ranking is found from WordFreq to determine the user's vocabulary size; English text recommendation based on vocabulary assessment step: Based on the user's vocabulary size, the server recommends corresponding vocabulary learning and reading texts, which are sent to the user's terminal via the network.

Description

Technical Field

[0001] This invention belongs to the field of computer-assisted English teaching, and involves related technical fields such as English learner ability assessment, English learning corpus recommendation, and automatic classification of English learning materials. In particular, it relates to a vocabulary level test processing method and system based on reading comprehension exercises. Background Technology

[0002] Vocabulary and grammar are two crucial aspects of English learning and teaching. Typically, individual grammar points can be assessed through specific exercises. While vocabulary can also be tested through multiple-choice and fill-in-the-blank exercises, these usually only assess the test-taker's understanding of one or a few words, making it difficult to measure their overall grasp of English vocabulary. Therefore, this invention does not provide effective guidance for English vocabulary learning. Consequently, this invention aims to provide a solution for assessing a test-taker's overall vocabulary proficiency.

[0003] Currently, vocabulary tests are primarily conducted through dictation or Chinese-English comparison tests. While these methods can reflect a test-taker's vocabulary size to some extent, they only assess their ability to match Chinese and English words, which differs from true English comprehension. Real vocabulary usage occurs within a text. Therefore, testing a reader's vocabulary through reading comprehension exercises would better reflect their vocabulary application skills. Summary of the Invention

[0004] To address the aforementioned technical issues, this invention utilizes several steps, including server-side preprocessing of reading text, accumulation of difficult words, and assessment of vocabulary recognition levels, to achieve student vocabulary testing based on reading comprehension exercises and the delivery of learning content.

[0005] The technical solution of the present invention is as follows: A vocabulary level test processing method based on reading comprehension exercises, comprising the following steps:

[0006] Reading text preprocessing steps: For each reading comprehension text, the server calculates the difficulty of each word w in the current text; then, the words in the text are sorted in descending order of difficulty and their frequency of occurrence is counted.

[0007] Steps for providing test content and accumulating difficult words for users: The server provides users with a predetermined number of reading comprehension questions, marks potential unmastered difficult words based on the user's answers, and accumulates them;

[0008] Vocabulary cognitive level evaluation step: after the user accumulates a predetermined scale of reading comprehension questions, all the difficult words encountered by the user are accumulated and averaged with the word frequency percentage in the word frequency table WordFreq as the coefficient to obtain the average word frequency percentage information, and then the word frequency percentage corresponding to the average word frequency percentage is found in WordFreq to determine the vocabulary size of the user;

[0009] English text recommendation step based on vocabulary evaluation: according to the vocabulary size of the user, the server recommends corresponding vocabulary learning and reading texts for the user, which are sent to the terminal of the user through the network.

[0010] Further, the reading text preprocessing step: for each reading comprehension text text, the difficulty of each word w in the text is calculated; then, the words in the text are sorted in descending order of difficulty and the number of occurrences is counted, as follows:

[0011] Step 1.1 transforms all the words in the text text into their original forms. Word transformations include: capitalization, verb transformations, noun plural transformations, adjective / adverb comparative and superlative transformations, etc. Assume that the set of original forms of all words in the text is WordSet, and go to step 1.2.

[0012] Step 1.2 counts the number of occurrences of each word w and its transformation in the text text, which is Count(w); counts the total number of words in the text text, which is T, and goes to step 1.3.

[0013] Step 1.3 calculates the difficulty of each word w in the text text, where n = Count(w), and goes to step 1.4.

[0014] Step 1.4 sorts all the words w in WordSet according to their difficulty in the text text, and sets the most difficult word in the text as w1, the second difficult word as w2, and so on. The number of occurrences of these words in the text is recorded as n1, n2,..., and so on. These words and their number of occurrences in the text are inserted into the key-value pair sequence sortedWordMap<w i ,n i > in descending order; the key w i represents the word ranked i in the text, and the value n i stores the number of occurrences of the word ranked i in the text.

[0015] Further, the step of providing test content and accumulating difficult words for the user: a predetermined amount of reading comprehension questions are provided for the user, and the user's potential unmastered difficult words are marked and accumulated according to the user's answers; specifically as follows:

[0016] Step 2.1 For the first time user user, set its HardWord accumulation table to be an empty set, and go to Step 2.2;

[0017] Step 2.2 In the subsequent use of the user, the server sends the reading comprehension questions to the client one by one according to the teacher assignment or the way of setting exercises, and the user uploads the answers to the server after completing each reading comprehension exercise, and goes to Step 2.3;

[0018] Step 2.3 According to the reading comprehension text text currently completed by the user user and the uploading of the answers, the hard words are accumulated, assuming that the text has K exercises and the user user answers C questions correctly: if C>0 and C<K, then the first (1-(C-1) / (K-1))*20% words in sortedWordMap<w i ,n i > are selected in order of difficulty from large to small, and set them as w1,...,w m , and go to Step 2.4; otherwise, go to Step 2.5;

[0019] Step 2.4 For each w i , 1<=i<=m, if w i has been in HardWord, i.e. HardWord(w i ) is not zero, then HardWord(w i )=HardWord(w i )+1; otherwise, w i is added to HardWord, and HardWord(w i )=1; go to Step 2.5;

[0020] Step 2.5 Record the current state, and if the user continues to test, go to Step 2.2; otherwise, end Step 2.

[0021] Further, the vocabulary level evaluation step: after the user accumulates a predetermined number of reading comprehension questions, all the difficult words encountered by the user are accumulated and averaged with their frequency percentage in the vocabulary frequency table WordFreq as a coefficient to obtain the average frequency percentage information, and then the corresponding vocabulary ranking of the average frequency percentage is found in WordFreq to determine the vocabulary size of the user, as follows:

[0022] Step 3.1 Frequency weighting of all difficult words accumulated by the user user: assuming that all difficult words in the HardWord corresponding to the user user are w1,w2,...,w H , their cumulative frequencies are N1,N2,...,N HThe corresponding percentage frequencies of these terms in the WordFreq vocabulary frequency table are P1, P2, ..., P. H Calculate the average frequency of difficult words according to the following formula (4), and proceed to step 3.2;

[0023]

[0024] Step 3.2 Sort all words in WordFreq in descending order of frequency percentage. Let the frequency of the Kth word be greater than HP, and the frequency of the (K+1)th word be less than or equal to HP; then K can be regarded as the student's vocabulary size, and the frequency of the Kth word is denoted as K. H .

[0025] Furthermore, the English text recommendation step based on vocabulary assessment includes: Based on the user's vocabulary level, the server recommends corresponding vocabulary learning and reading texts, which are then sent to the user's terminal via the network. Specifically, this includes:

[0026] Based on K, relevant vocabulary learning and reading texts are recommended to users. Specifically, for texts in the corpus, if the frequency of the top 20% of difficult words is calculated according to their difficulty in the text, more than half of them are in K. H Later, or if these words are no longer in K H In the following cases, no content will be pushed; otherwise, the server will push the content to the client.

[0027] According to another aspect of the present invention, a vocabulary proficiency testing system based on reading comprehension exercises is also proposed, comprising a client and a server; the client includes:

[0028] In the user login module, the user enters their username and password on the client side, which are then securely encrypted and sent to the server. The server decrypts the received data and performs other security processing to obtain the username and password. It then compares the data with a database and performs the following operations: If the received username and password match a username-password database stored on the server, the server provides the corresponding service based on the user's learning data; otherwise, the server prompts the user to verify their username and password and try again.

[0029] In the user reading and answering module, users complete reading comprehension texts assigned by the system or teachers and answer questions according to their learning needs. The question information and answer results will be returned to the server.

[0030] The server includes:

[0031] The reading text preprocessing module is used to preprocess all reading comprehension texts stored on the server and generate a sorted list of vocabulary difficulty for each text.

[0032] A user information storage module stores the user's username, password and other information into a database if the user is logging in for the first time, and allocates a special user hard word accumulation table HardWord for the user in the memory, and sets the user's hard word accumulation table HardWord as empty; if the user has logged in before, corresponding operations are performed according to the previous hard word accumulation table record;

[0033] A user answer text providing module assigns corresponding reading comprehension test texts to the user according to the user's needs;

[0034] A user hard word accumulation module accumulates and records the hard words accumulated by the user when reading the text in the user's corresponding hard word accumulation table according to the user's current reading comprehension answer after receiving the user's answer.

[0035] A vocabulary cognitive level evaluation module calculates and returns the user's vocabulary level to the corresponding user according to the user's accumulated answer information.

[0036] Beneficial effects:

[0037] Firstly, the vocabulary level of the reader is tested by reading and answering, which avoids the problem of ignoring the use of the word in the English text by only considering the word itself, and can better evaluate the ability of the tester in the use of vocabulary.

[0038] Secondly, when evaluating the vocabulary cognitive level of the tester, the common Chinese-English comparison method is not used, but the hard words encountered by the tester in reading and answering are considered and evaluated in combination with the hard words in the corpus, which makes the evaluation result more accurate.

[0039] Thirdly, as a whole, the method has good effect and the evaluation result is more accurate. Compared with similar methods, the method has the characteristics of accuracy and efficiency, and is suitable for evaluating the English vocabulary level of various groups of people, greatly reducing the use threshold. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 The flowchart for the implementation of the present application;

[0041] Figure 2 The reading text preprocessing flowchart of the present application;

[0042] Figure 3 The flowchart for providing test content for the user and accumulating hard words of the present application;

[0043] Figure 4 The vocabulary cognitive level evaluation flowchart of the present application;

[0044] Figure 5A reading comprehension exercise-based vocabulary level test processing system block diagram of the present application. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of the present application.

[0046] According to an embodiment of the present application, to implement the reading comprehension exercise-based vocabulary level test processing method and system, a certain scale of vocabulary in a database and English reading comprehension exercises and their answers covering a wide range are needed. Each reading comprehension preferably has multiple exercises, for example, 5 or more. In addition, the database also stores a vocabulary frequency table WordFreq and a hard word cumulative table HardWord for each student. The vocabulary frequency table WordFreq stores all words and their frequency information in the corpus, that is, WordFreq(w) stores the percentage of the number of occurrences of a word w in the corpus to the total number of occurrences of the vocabulary, which is used as a reference when estimating the vocabulary of a student; the hard word cumulative table HardWord(w) of a student records the cumulative number of texts in which the student identifies w as a difficult word of the text when making mistakes in reading the text.

[0047] A reading comprehension exercise-based vocabulary level test processing system according to an embodiment of the present application, as shown in Figure 5 includes a client and a server; the client includes:

[0048] a user login module, in which a user inputs a username and a password on the client, and the client sends the username and the password to the server after security encryption; the server decrypts the received data and obtains the username and the password, and compares the username and the password in the database to perform the following corresponding operations: if the received username and password match a certain username-password stored in the server, the server provides corresponding services according to the user's learning data; otherwise, the user is prompted to check the username and password, and the user tries again;

[0049] a user reading and answering module, in which a user completes a reading comprehension text and answers questions according to the learning needs, and the question information and the answering results are returned to the server;

[0050] the server includes:

[0051] a reading text preprocessing module, which is used for preprocessing all reading comprehension texts stored in the server to generate a sorted list of vocabulary difficulty in each text;

[0052] a user information storage module, if the user is logging in for the first time, storing the user's username, password and other information into a database, and assigning a special user hard word accumulation table HardWord to the user, and setting the user's hard word accumulation table HardWord as empty; if the user has logged in before, performing corresponding operations according to the previous hard word accumulation table record;

[0053] a user answer text providing module, assigning corresponding reading comprehension test texts to the user according to the user's needs;

[0054] a user hard word accumulation module, after receiving the user's answer result, accumulating and recording the hard words accumulated by the user when reading the text according to the user's current reading comprehension answer situation, and storing the hard words into the corresponding hard word accumulation table of the user;

[0055] a vocabulary cognitive level evaluation module, calculating and evaluating the vocabulary level of the user according to the answer information accumulated by the user, and returning the vocabulary level to the corresponding user.

[0056] According to another embodiment of the present application, a vocabulary level test processing method based on reading comprehension exercises is provided, as shown in Figure 1 , comprising the following steps:

[0057] a reading text preprocessing step: the server calculates the difficulty of each word w in the current text in the text for each reading comprehension text text, and then sorts each word in the text according to the difficulty from large to small and counts the number of occurrences;

[0058] a step of providing test content for the user and accumulating hard words: the server provides a predetermined amount of reading comprehension questions for the user, and accumulates and accumulates the potential unmastered hard words of the user according to the user's answer result;

[0059] a vocabulary cognitive level evaluation step: after the user accumulates a predetermined scale of reading comprehension questions, all the hard words encountered by the user are accumulated and averaged with the word frequency percentage in the vocabulary word frequency table WordFreq as a coefficient to obtain average word frequency percentage information, and then the average word frequency percentage corresponding to the vocabulary rank in WordFreq is found to determine the vocabulary of the user;

[0060] an English text recommendation step based on vocabulary evaluation: according to the vocabulary of the user, the server recommends corresponding vocabulary learning and reading texts to the user, and sends the texts to the terminal of the user through the network.

[0061] Specifically, according to the present embodiment, the reading text preprocessing step is as follows, as shown in Figure 2 :

[0062] For each reading comprehension text text, the invention calculates the difficulty of each word w in the current text textDif(w, text); then, the words in the text are sorted in order of difficulty from large to small and the number of occurrences is counted. Among them, it is assumed that the most difficult word in the current text is w1, the second difficult word is w2, and the number of occurrences in the text is n1, n2,..., and so on.

[0063] Suppose the text text contains N words, the probability of the word w appearing k times in the text text can be approximated by a binomial distribution, that is:

[0064]

[0065] where, is the number of combinations of selecting k elements from N elements, N is the total number of words in the text text, P is the number of occurrences of the word w in the corpus, that is, P = WordFreq(w); then, under the condition that the word w appears in the text text, the alpha probability quantile of w appearing n times in the text text is, and the probability of w appearing n times or less is:

[0066]

[0067] And the difficulty of the word w in the text text is composed of the information amount I(w) = -lg(WordFreq(w)) of w in the corpus and the alpha probability quantile a(w, n) of w appearing n times in the text text, that is:

[0068] textDif(w, text) = a(w, n)I(w). (3)

[0069] Specifically, the step is implemented as follows:

[0070] Step 1.1 transforms all the words in the text text into the original form, including capitalizing the first letter, various verb transformations, noun plural transformations, adjective / adverb comparative and superlative transformations, etc. Assume that the set of original forms of all words in the text is WordSet, and go to step 1.2;

[0071] Step 1.2 counts the number of occurrences of each word w and its transformation in the text text in WordSet, which is Count(w); counts the total number of words in text, which is T, and goes to step 1.3;

[0072] Step 1.3. For each word w in WordSet, calculate the difficulty of w in text using the formula textDif(w, text) = a(w, n)I(w), where n = Count(w), go to Step 1.4.

[0073] Step 1.4. Sort all words w in WordSet according to their difficulty in text, set the most difficult word in the current text as w1, the second difficult word as w2, and so on, and record the number of times each word appears in the text as n1, n2,..., and so on. Insert these words and their number of occurrences in the text into a key-value pair sequence sortedWordMap<w i ,n i >.

[0074] After the text vocabulary difficulty evaluation is completed, a key-value pair sequence sortedWordMap<w i ,n i > of words and their number of occurrences in the text is obtained, which is sorted according to difficulty. Among them, the key w i represents the word ranked i-th in difficulty in the text, and the value n i stores the number of times the word ranked i-th in difficulty appears in the text.

[0075] The step of providing test content and accumulating difficult words for the user, specifically includes:

[0076] Since the reading comprehension includes five exercises, the number of questions answered (or answered incorrectly) by the user can be used to mark the potential unmastered difficult words. Specifically, if the user only answers one question correctly, the first 20% of the difficult words in the difficult word sequence of the text are accumulated as vocabulary; if the user answers two questions correctly, the first 15% of the difficult words in the difficult word sequence of the text are accumulated as vocabulary; if the user answers three questions correctly, the first 10% of the difficult words in the difficult word sequence of the text are accumulated as vocabulary; if the user answers four questions correctly, the first 5% of the difficult words in the difficult word sequence of the text are accumulated as vocabulary; if the user answers all five questions correctly or incorrectly, no difficult word accumulation is performed. The specific implementation is as follows:

[0077] Step 2.1. For the first-time user user, set the difficult word accumulation table HardWord to be an empty set, go to Step 2.2.

[0078] Step 2.2. In the subsequent use of the user, the server can send reading comprehension questions to the client one by one according to the teacher's assignment or exercise arrangement, etc. for the user user to practice. After the user completes each reading comprehension exercise, the answer is uploaded to the server, go to Step 2.3.

[0079] Step 2.3. Accumulate difficult words according to the current completed reading comprehension text and the uploaded answer of the user user. Assume that the text has K questions and the user user answers C questions correctly: if C>0 and C<K, select the first (1-(C-1) / (K-1))*20% words in sortedWordMap<w i ,n i > in descending order of difficulty, and suppose they are w1,...,w m , go to step 2.4; otherwise, go to step 2.5.

[0080] Step 2.4. For each w i (1<=i<=m): if w i has been in HardWord, i.e., HardWord(w i ) is not zero, then let HardWord(w i ) = HardWord(w i )+1; otherwise, add w i to HardWord and let HardWord(w i ) = 1; go to step 2.5.

[0081] Step 2.5. Record the current state. If the user continues the test, go to step 2.2; otherwise, end step 2.

[0082] The accumulation of difficult words is a gradual process, and the user needs to complete a certain number of reading comprehension questions to accumulate new words. If the reading comprehension is completely correct, the text vocabulary difficulty is too small; if the reading comprehension is completely wrong, the text vocabulary difficulty is too large; neither of these two cases is suitable for accurate vocabulary accumulation.

[0083] The vocabulary recognition level evaluation step includes: after the user accumulates a certain number of reading comprehension questions, all the difficult words encountered by the user are accumulated and averaged with their word frequency percentage in the word frequency table WordFreq as a coefficient to obtain average word frequency percentage information, and then the word ranking corresponding to the average word frequency percentage is found in WordFreq to determine the vocabulary size of the user.

[0084] Specifically, this step is implemented as follows:

[0085] Step 3.1. Frequency weighting of all difficult words accumulated by the user user: assume that all difficult words in the HardWord corresponding to the user user are w1,w2,...,w H , and their cumulative frequencies are N1,N2,...,N Hand the corresponding word frequency percentage in the word frequency list WordFreq is P1, P2,..., P H The average word frequency of difficult words is calculated according to the following formula (4) and the process goes to step 3.2.

[0086]

[0087] Step 3.2: All the words in WordFreq are sorted according to the frequency percentage from large to small. Let the frequency of the Kth word be greater than HP, and the frequency of the K+1th word be less than or equal to HP. Then K can be regarded as the vocabulary size of the student. The frequency of the Kth word is recorded as K H .

[0088] English text recommendation based on vocabulary evaluation: Based on K, the corresponding vocabulary learning and reading text can be recommended for the student. Specifically, for a text text in the corpus, if more than half of the word frequencies of the top 20% difficult words in the text are greater than K H , it is considered that the text is not suitable for reading; if none of these words is greater than K H , it is considered that the text is too simple; otherwise, the text is pushed.

[0089] Although the above describes the specific embodiments of the present application in detail, so that those skilled in the art can understand the present application, it should be clear that the present application is not limited to the scope of the specific embodiments, and for those skilled in the art, as long as various changes are within the spirit and scope of the present application defined and determined by the appended claims, all the inventions utilizing the concept of the present application are within the scope of protection.

Claims

1. A vocabulary level test processing method based on reading comprehension exercises, characterized by, The method comprises the following steps: The reading text preprocessing step: for each reading comprehension text text, calculate the difficulty of each word w in the current text; then, sort the words in the text according to the difficulty from large to small and count the number of occurrences, as follows: Step 1.1 Transform all the words in the text text into the original form, including: capitalization, verb transformation, noun plural transformation, adjective / adverb comparative and superlative transformation, assuming that the set of original forms of all words in the text is WordSet, go to step 1.2; Step 1.2 Count the number of occurrences of each word w and its transformation in the text text, denoted as Count(w); count the total number of words in the text, denoted as T, go to step 1.3; Step 1.3 For each word w in WordSet, calculate the difficulty of w in the text, where n=Count(w), go to step 1.4; The step of providing test content for the user and accumulating difficult words: providing a predetermined amount of reading comprehension questions for the user, and marking the user's potential unmastered difficult words according to the user's answer results and accumulating them; as follows: Step 2.1 For the first user user, set the difficult word accumulation table HardWord to be empty, go to step 2.2; Step 2.2 In the subsequent use of the user, the server sends reading comprehension questions to the client according to the teacher assignment or exercise arrangement for the user user to practice, and the user uploads the answers to the server after completing each reading comprehension exercise, go to step 2.3; Go to step 2.5; Step 1.4 Sort all the words w in the WordSet according to their difficulty in the text text, set the most difficult word in the current text as w1, the second difficult word as w2, and the number of times they appear in the text as n1, n2,..., and so on; insert these words and their number of times of appearance in the text into the key-value pair sequence sortedWordMap<w i , n i > in descending order; the key w i represents the word ranked i-th in terms of difficulty in the text, and the value n i stores the number of times the word ranked i-th in terms of difficulty appears in the text; The difficulty of a word w in a text text is composed of the information content of w in the corpus I(w) = -lg(WordFreq(w)) and the alpha probability fractile of n occurrences of w in text text , i.e. textDif(w, text) = textDif(w, text) .

2. The vocabulary level test processing method based on reading comprehension exercise according to claim 1, wherein, Step 2.5 Record the current state, if the user continues to test, go to step 2.2; otherwise, end step 2. ​ ​ Step 2.

3. Accumulate difficult words according to the current completed reading comprehension text text and the uploaded answer situation of user user. Assume that text has K exercises and user answers C questions correctly: if C > 0 and C < K, select the first (1- (C-1) / (K-1)) words in sortedWordMap<w i ,n i > in order of difficulty from large to small, and go to step 2.4; otherwise, go to step 2.

5. 20% of the words, set them as w1,...,w m , and go to step 2.4; otherwise, go to step 2.

5. Step 2.4 For each w i , 1<=i<=m, if w i has been in HardWord, i.e., HardWord(w i ) is not zero, then let HardWord(w i )=HardWord(w i )+1; Otherwise, set w i Add HardWord, let HardWord(w i )=1; ​ ​ 3.The vocabulary level test processing method based on reading comprehension exercise according to claim 1, wherein, The vocabulary cognitive level evaluation step: after the user accumulates enough reading comprehension questions of a predetermined scale, all difficult words encountered by the user are accumulated and averaged with the word frequency percentage in the word frequency table WordFreq as a coefficient to obtain average word frequency percentage information, and then the vocabulary ranking corresponding to the average word frequency percentage is found from the WordFreq to determine the vocabulary size of the user, specifically as follows: Step 3.1 Perform frequency-weighted analysis on all difficult words accumulated by user: Assume that all difficult words in the HardWord corresponding to user are w1, w2, ..., w H The cumulative number of occurrences are N1, N2, ..., N H The corresponding percentage frequencies of these terms in the WordFreq vocabulary frequency table are P1, P2, ..., P. H Calculate the average frequency of difficult words according to the following formula (4), and proceed to step 3.2; （4） Step 3.2 Sort all the words in WordFreq by the percentage of frequency in descending order. Let the frequency of the Kth word be greater than HP, while the frequency of the K+1th word be less than or equal to HP. Then K is regarded as the vocabulary of the student. Let the frequency of the Kth word be K H .

4. The vocabulary level test processing method based on reading comprehension exercise according to claim 3, wherein, The English text recommendation step based on vocabulary evaluation: according to the vocabulary size of the user, the server recommends corresponding vocabulary learning and reading texts to the user's terminal through the network, specifically including: Based on K, the corresponding vocabulary learning and reading text are recommended to the user. Specifically, for the text in the corpus, if it is calculated according to the difficulty of the vocabulary in the text, more than half of the word frequency of the first 20% difficult words are in K H Afterwards, or if none of these words are in K H , then no content is pushed; in addition to the two cases, the server pushes the content to the client, where the frequency of the Kth vocabulary is recorded as K H . 5.A vocabulary level test processing system based on reading comprehension exercises, comprising a client and a server; characterized in that, The client includes: The user login module: the user inputs the username and password on the client, which are sent to the server after secure encryption processing by the client; the server decrypts the received data and performs corresponding security processing to obtain the username and password, and performs the following corresponding operations in the library: if the received username and password match a certain username-password stored in the server, the server provides corresponding services according to the user's learning data; otherwise, the user is prompted to check the username and password and try again; The user reading and answering module: the user completes the reading comprehension text and answers the questions according to the learning needs, and the question information and answer results are returned to the server; The server includes: The reading text preprocessing module: used for preprocessing all reading comprehension texts stored in the server to generate a sorted list of vocabulary difficulty in each text; The user information storage module: if the user is logging in for the first time, the user's username and password information are stored in the database, and a user-specific hard word accumulation table HardWord is allocated to the user in the storage; if the user has logged in before, the corresponding operation is performed according to the previous hard word accumulation table record; The user answering text providing module: the system assigns the user with corresponding reading comprehension test texts according to the user's needs; The user hard word accumulation module: after receiving the user's answer results, the server accumulates and records the hard words accumulated by the user in reading the text according to the user's current reading comprehension answer situation in the user's corresponding hard word accumulation table; The vocabulary cognitive level evaluation module: the server calculates and evaluates the user's vocabulary level according to the user's accumulated answer information and returns it to the corresponding user; For each reading comprehension text text, the difficulty of each word w in the current text textDif(w, text) is calculated; then, the words in the text are sorted in descending order of difficulty and the number of occurrences is counted; where w1 is the most difficult word in the current text, w2 is the second difficult word, and n1, n2,... are the number of occurrences of w1 and w2 in the text, and so on; Assuming that the text text contains N words, the probability of the occurrence of a word w in the text text is approximated by a binomial distribution, that is: （1） wherein, is the number of combinations of choosing k elements from N elements, N is the total number of vocabulary of the text text, P is the percentage of the occurrence of the word w in the corpus, i.e. P = WordFreq(w); then, the α probability quantile of the occurrence of w n times in the text text under the condition that the word w appears in the text text, is the cumulative probability of the occurrence of w less than or equal to n times, i.e. （2） The difficulty of a word w in a text text is composed of the information content of w in the corpus I(w) = -lg(WordFreq(w)) and the alpha probability fractile of n occurrences of w in text text , i.e. textDif(w, text) = textDif(w, text) .

Images