Method, system, storage medium and device for correcting answer data of English-Chinese translation test questions

Abstract

The present invention discloses a method, system, storage medium and device for correcting answer data of English-to-Chinese test questions, the method comprising: obtaining a batch of English words and English phrases, establishing a basic vocabulary library according to the batch of English words and English phrases; obtaining user answer data of English-to-Chinese test questions, slicing the user answer data to obtain answer phrases to be corrected; obtaining Chinese interpretations corresponding to the English words and English phrases required to be tested recorded in the basic vocabulary library to obtain reference answer data, slicing the reference answer data to obtain reference answer phrases; comparing the answer phrases to be corrected with the reference answer phrases, using a preset hierarchical correction algorithm to evaluate the answer phrases to be corrected according to multiple evaluation indicators, so as to output the correction results of the user answer data. The present invention solves the problem that the accuracy of the existing automatic correction of English-to-Chinese vocabulary is general and correction errors are easily caused.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method, system, storage medium and device for correcting answer data of English-Chinese translation test questions. Background Art

[0002] At present, the grading technology in the field of English teaching in actual application is mainly automatic grading of English composition and automatic grading of Chinese-English translation, and both are for grading English handwritten answers, while there are relatively few studies and applications on English-Chinese grading technology. For automatic grading of vocabulary English-Chinese translation, the score is usually determined by directly comparing whether the student's answer matches the correct answer or by the similarity between the two.

[0003] The calculation of Chinese text similarity is mainly based on semantic feature representation. However, in the automatic English-Chinese translation correction scenario for words and phrases, short texts are compared, and there are a large number of ultra-short texts. Such short texts usually contain less contextual information, making the extraction of their semantic features inaccurate.

[0004] In the automatic correction of English-Chinese translation for vocabulary, if the student's answer is judged by directly comparing whether it matches the correct answer, it can only solve part of the problem. It cannot judge the situation where the character strings are different but the meaning is the same. If the judgment is made by comparing the similarity between the two, the similarity calculation will be inaccurate due to the inaccurate extraction of semantic features of ultra-short texts, which will affect the accuracy of correction. On the other hand, words also have part-of-speech information. Ignoring the part-of-speech information and only judging whether the meaning is the same will also cause correction errors.

[0005] Therefore, there is still a problem in the prior art that the accuracy of automatic correction of vocabulary translation from English to Chinese is general, which easily leads to correction errors. Summary of the invention

[0006] In view of the deficiencies in the prior art, the purpose of the present invention is to provide a method, system, storage medium and device for correcting answer data of English-Chinese test questions, aiming to solve the problem that the accuracy of automatic correction of vocabulary English-Chinese translation in the prior art is generally poor, which easily leads to correction errors.

[0007] The first aspect of the present invention is to provide a method for correcting answer data of English-Chinese translation test questions, the method comprising:

[0008] Acquire a batch of English words and English phrases, and establish a basic vocabulary library based on the batch of English words and English phrases;

[0009] Acquire user answer data of English-Chinese translation test questions, slice the user answer data, and obtain answer phrases to be corrected;

[0010] Obtaining Chinese definitions corresponding to the English words and English phrases required to be tested and recorded in the basic vocabulary database to obtain reference answer data, and slicing the reference answer data to obtain reference answer phrases;

[0011] The answer phrase to be corrected is compared with the reference answer phrase, and a preset hierarchical correction algorithm is used to evaluate the answer phrase to be corrected according to multiple evaluation indicators to output a correction result of the user's answer data.

[0012] According to one aspect of the above technical solution, a batch of English words and English phrases are obtained, and a step of establishing a basic vocabulary library based on the batch of English words and English phrases includes:

[0013] Acquire a batch of English words and English phrases, create a basic vocabulary library based on the English words and English phrases, and set the textbook interpretation and broad interpretation of each English word and English phrase in the basic vocabulary library;

[0014] According to the English words and English phrases in the basic vocabulary library, the synonyms and antonyms with different meanings corresponding to each English word or English phrase are traversed and queried within the preset reference data search range, and semantic classification and annotation are performed.

[0015] According to one aspect of the above technical solution, the data type of the user answer data includes picture type and text type.

[0016] According to one aspect of the above technical solution, the steps of obtaining user answer data of English-Chinese translation test questions, slicing the user answer data, and obtaining answer phrases to be corrected specifically include:

[0017] Acquire user answer data of an English-to-Chinese test question, and identify the data type of the user answer data;

[0018] When the data type of the user answer data is a picture type, the OCR image recognition technology is used to perform OCR recognition on the user answer data to obtain text data, and the text data is cleaned and sliced ​​to obtain the answer phrase to be corrected;

[0019] When the data type of the user answer data is text data of a text type, data cleaning and data slicing are performed on the text data to obtain answer phrases to be corrected.

[0020] According to one aspect of the above technical solution, the step of comparing the answer phrase to be corrected with the reference answer phrase, using a preset hierarchical correction algorithm to evaluate the answer phrase to be corrected according to multiple evaluation indicators to output the correction result of the user's answer data includes:

[0021] Performing character matching on the answer phrase to be corrected and the reference answer phrase, and removing the successfully matched answer items from the answer phrase to be corrected;

[0022] If the answer item in the answer phrase to be corrected is empty after polling and matching, it is determined that the answer to the answer phrase to be corrected is correct.

[0023] According to one aspect of the above technical solution, the steps further include:

[0024] If the answer items in the answer phrase to be corrected are not empty after the polling and matching, all the answer items in the answer phrase to be corrected are judged for the correctness of the Chinese phrase collocation, including the judgment of wrong characters;

[0025] If it is determined that the collocation of the Chinese phrase is wrong, then the answer to the answer item is directly determined to be wrong;

[0026] If it is determined that the Chinese phrase is correctly matched, all answer items in the answer phrase to be corrected are translated into English phrases and compared with the antonym phrases of the vocabulary to be corrected. If the comparison is successful, it is directly determined that the answer is wrong.

[0027] According to one aspect of the above technical solution, the steps further include:

[0028] If the comparison between the English phrase and the antonym phrase of the vocabulary to be corrected fails, then all the English phrases corresponding to the answer phrase to be corrected are compared with the synonym phrase of the vocabulary to be corrected;

[0029] When the semantics and part of speech of the English phrase are different from the semantics and part of speech of the corresponding synonym, similarities are calculated between all answer items in the answer phrase to be corrected and the reference answer phrase based on the Chinese semantic similarity model, and a similarity threshold is set;

[0030] When the similarity is higher than the threshold, it is determined that the answer item and the reference answer are semantically similar;

[0031] When the semantics of all answer items in the answer phrase to be corrected are similar to the reference answer, it is determined that the answer to the answer phrase to be corrected is correct.

[0032] The second aspect of the present invention is to provide a system for correcting answer data of English-Chinese test questions, which is applied to the method described in the above technical solution, and the system comprises:

[0033] A vocabulary library creation module is used to obtain a batch of English words and English phrases, and to establish a basic vocabulary library based on the batch of English words and English phrases;

[0034] The answer data processing module is used to obtain the user's answer data of the English-Chinese translation test questions, and slice the user's answer data to obtain the answer phrases to be corrected;

[0035] The answer data processing module is used to obtain the Chinese definitions of the English words and English phrases required to be tested recorded in the basic vocabulary library, obtain reference answer data, and slice the reference answer data to obtain reference answer phrases;

[0036] The comparison and evaluation module is used to compare the answer phrase to be corrected with the reference answer phrase, and use a preset hierarchical correction algorithm to evaluate the answer phrase to be corrected according to multiple evaluation indicators to output the correction result of the user's answer data.

[0037] A third aspect of the present invention is to provide a readable storage medium having a computer program stored thereon, which implements the method described in the above technical solution when executed by a processor.

[0038] The fourth aspect of the present invention is to provide an electronic device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the method described in the above technical solution when executing the computer program.

[0039] Compared with the prior art, the method, system, storage medium and device for correcting answer data of English-Chinese test questions shown in the present invention have the following beneficial effects:

[0040] The method for correcting the answer data of the English-to-Chinese test questions shown in the present invention is adopted, by obtaining English words and English phrases to create a basic vocabulary library, and then the user answer data and the reference answer data are sliced ​​and processed respectively to obtain the answer phrases to be corrected and the reference answer phrases, and finally the layered correction algorithm is used to compare the answer phrases to be corrected with the reference answer phrases to output the correction results. This embodiment provides an automatic correction idea for English-to-Chinese translation for English words and English phrases, which is suitable for teachers to detect students' daily words in English teaching in primary and secondary schools, can greatly reduce the burden of teachers' correction, improve work efficiency, and at the same time can more conveniently assist students in memorizing English vocabulary. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which:

[0042] Figure 1 It is a flowchart of a method for correcting answer data of English-Chinese translation test questions in one embodiment of the present invention;

[0043] Figure 2It is a structural block diagram of a system for correcting answer data of English-Chinese translation test questions in one embodiment of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings. Several embodiments of the present invention are shown in the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0045] It should be noted that when an element is referred to as being "fixed to" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0046] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0047] Embodiment 1

[0048] See also Figure 1 The first embodiment of the present invention provides a method for correcting answer data of an English-Chinese translation test question, the method comprising steps S10 to S40:

[0049] Step S10, obtaining a batch of English words and English phrases, and establishing a basic vocabulary library based on the batch of English words and English phrases.

[0050] First of all, it should be explained that the method shown in this embodiment is used to correct the user's answer data of English-to-Chinese test questions, that is, to correct the answers written or submitted by the user. In order to improve the applicability of the method shown in this embodiment, it is necessary to batch obtain the English words and English phrases corresponding to each grade or age group or academic level, for example, obtain the newly emerging English words and English phrases corresponding to each grade in the compulsory education stage, and based on this, create a basic vocabulary library that can completely cover the scope of the examination.

[0051] In this embodiment, the acquisition of English words and English phrases can be carried out through the network, by acquiring the corresponding teaching material data, and then acquiring the corresponding English words and English phrases in the teaching material data. In the basic vocabulary library, when the same vocabulary is acquired through the network, the content of the basic vocabulary library can be pre-detected and removed when the vocabulary is stored, so as to reduce the data volume of the basic vocabulary library.

[0052] In this embodiment, the steps of obtaining a batch of English words and English phrases and establishing a basic vocabulary library according to the batch of English words and English phrases include:

[0053] Acquire a batch of English words and English phrases, create a basic vocabulary library based on the English words and English phrases, and set the textbook interpretation and broad interpretation of each English word and English phrase in the basic vocabulary library;

[0054] According to the English words and English phrases in the basic vocabulary library, the synonyms and antonyms with different meanings corresponding to each English word or English phrase are traversed and queried within the preset reference data search range, and semantic classification and annotation are performed.

[0055] Among them, English words and English phrases corresponding to each academic level and their Chinese interpretations are collected and organized as the basic vocabulary database, including primary and secondary school English, university level 4, 6 and 8 English, TOEFL English and IELTS English, etc.; and the two types of interpretations, textbook interpretation and broad interpretation, are separately set for primary and secondary school English examination vocabulary. For vocabulary corresponding to other levels, only broad interpretations are set to meet users' stratified examination of textbook interpretations and broad interpretations.

[0056] Among them, textbook interpretation refers to the interpretation of English words and English phrases tested in primary and secondary school English textbooks, including all versions of textbooks nationwide; while broad interpretation is not limited to the meaning tested in the textbooks, but also includes all other possible meanings. Textbook interpretation can be obtained by collecting and arranging textbooks of various versions in primary and secondary schools nationwide, while broad interpretation can be obtained from major dictionaries. The English words and English phrases in the basic vocabulary database are processed into a pre-defined unified standard format to facilitate later searches.

[0057] Finally, based on the basic vocabulary database, all synonyms and antonyms with different meanings are collected for all English words and English phrases, and all synonyms and antonyms are clustered with different semantics, and semantic classification and labeling are performed at the same time.

[0058] Step S20, obtaining the user's answer data of the English-to-Chinese test question, and slicing the user's answer data to obtain answer phrases to be corrected.

[0059] Among them, the user answer data is the user's answer data for the English-Chinese translation test question, and its data type can be an image type or a text type. Among them, the image type user answer data can be obtained by scanning or photographing the answer area written by the user by a specific image acquisition device, for example, obtaining an answer sheet image, and then performing OCR detection and cutting on the answer sheet image to obtain multiple sub-images. The data corresponding to each sub-image is the user answer data required in this embodiment, so as to correct the answer situation of the English-Chinese translation test question based on the user answer data, such as outputting scores and score feedback.

[0060] In this embodiment, when it is detected that the data type of the user answer data is an image type, the user answer data will be sliced ​​to obtain multiple answer phrases to be corrected, that is, the user answer data written by the user is divided into multiple answer phrases to be corrected.

[0061] Specifically, the steps of obtaining user answer data of English-Chinese translation test questions, slicing the user answer data, and obtaining answer phrases to be corrected specifically include:

[0062] Acquire user answer data of an English-to-Chinese test question, and identify the data type of the user answer data;

[0063] The OCR image recognition technology is used to perform OCR recognition on the user's answer data to obtain text data, and the text data is cleaned and sliced ​​to obtain answer phrases to be corrected.

[0064] In addition, when the data type of the user answer data is text data of a text type, data cleaning and data slicing are performed on the text data to obtain answer phrases to be corrected.

[0065] Step S30, obtaining the Chinese definitions of the English words and English phrases required to be tested recorded in the basic vocabulary library, obtaining reference answer data, slicing the reference answer data, and obtaining reference answer phrases.

[0066] In this embodiment, when slicing the reference answer data, for English words or English phrases, after the format is standardized, the textbook interpretation or broad interpretation in the basic vocabulary library is traversed and queried according to its examination scope, and the textbook interpretation or broad interpretation is cleaned and segmented to obtain a reference answer group.

[0067] Step S40, comparing the answer phrase to be corrected with the reference answer phrase, using a preset hierarchical correction algorithm to evaluate the answer phrase to be corrected according to multiple evaluation indicators, so as to output a correction result of the user's answer data.

[0068] In this embodiment, after creating a basic vocabulary library and slicing to obtain answer phrases to be corrected and reference answer question groups, the answer phrases to be corrected are compared with the reference answer question groups, and then a hierarchical correction algorithm is used to evaluate the answer items in the answer phrases to be corrected according to multiple evaluation indicators to output the correction results of the user's answer data, for example, the score and score feedback can be output to realize automatic correction of answers to English-Chinese translation questions.

[0069] As a specific example, if the word to be tested is: "differ", the answer data to be corrected records: "different; different from".

[0070] In this embodiment, the textbook interpretation and / or broad interpretation will be found from the basic vocabulary library according to the examination scope. For example, the textbook interpretation of the word "differ" is "vi. different, not the same, different from, have distinction, have objection, disagreement".

[0071] Then, the answer data to be corrected and the reference answer data are cleaned and segmented, and the answer group to be corrected is obtained as ['different', 'distinguished from'], and the reference answer group is obtained as ['different', 'distinguished from', 'different from', 'not the same', 'different', 'dissenting', 'dissenting opinions']. Finally, each answer item in the answer group to be corrected and each answer item in the reference answer group are compared and judged using the idea of ​​layered correction, so as to output the correction result. For example, if the above-mentioned answer data to be corrected only records part of the reference answers, then the answer can be appropriately deducted. For example, the score of the above-mentioned test word is 1 point. Since the answer data to be corrected only records part of the reference answers, its completion degree is less than 1, then the answer score can be determined to be 0.5 points.

[0072] Compared with the prior art, the method for correcting the answer data of the English-Chinese translation test questions shown in this embodiment has the following beneficial effects:

[0073] The method for correcting the answer data of the English-to-Chinese test questions shown in this embodiment is adopted. By acquiring English words and English phrases to create a basic vocabulary library, the user's answer data and the reference answer data are sliced ​​and processed respectively to obtain the answer phrases to be corrected and the reference answer phrases. Finally, a hierarchical correction algorithm is used to compare the answer phrases to be corrected with the reference answer phrases to output the correction results. This embodiment provides an automatic correction idea for English-to-Chinese translation for English words and English phrases, which is suitable for teachers' daily word detection of students in English teaching in primary and secondary schools. It can greatly reduce the burden of teachers' correction and improve work efficiency. At the same time, it can more conveniently assist students in memorizing English vocabulary.

[0074] Embodiment 2

[0075] The second embodiment of the present invention also provides a method for correcting answer data of an English-to-Chinese test question. The method shown in this embodiment is basically similar to the method shown in the first embodiment, except that:

[0076] In this embodiment, the step of comparing the answer phrase to be corrected with the reference answer phrase, using a preset hierarchical correction algorithm to evaluate the answer phrase to be corrected according to multiple evaluation indicators to output the correction result of the user's answer data includes:

[0077] Performing character matching on the answer phrase to be corrected and the reference answer phrase, and removing the successfully matched answer items from the answer phrase to be corrected;

[0078] If the answer item in the answer phrase to be corrected is empty after polling and matching, it is determined that the answer to the answer phrase to be corrected is correct.

[0079] Wherein, if the answer items in the answer phrase to be corrected are not empty after polling matching, all the answer items in the answer phrase to be corrected are judged for the correctness of Chinese phrase collocation, including judgment of wrong characters;

[0080] If it is determined that the collocation of the Chinese phrase is wrong, then the answer to the answer item is directly determined to be wrong;

[0081] If it is determined that the Chinese phrase is correctly matched, all answer items in the answer phrase to be corrected are translated into English phrases and compared with the antonym phrases of the vocabulary to be corrected. If the comparison is successful, it is directly determined that the answer is wrong.

[0082] If the comparison between the English phrase and the antonym phrase of the vocabulary to be corrected fails, all the English phrases corresponding to the answer phrase to be corrected are compared with the synonym phrase of the vocabulary to be corrected;

[0083] When the semantics and part of speech of the English phrase are different from the semantics and part of speech of the corresponding synonym, similarities are calculated between all answer items in the answer phrase to be corrected and the reference answer phrase based on the Chinese semantic similarity model, and a similarity threshold is set;

[0084] When the similarity is higher than the threshold, it is determined that the answer item and the reference answer are semantically similar;

[0085] When the semantics of all answer items in the answer phrase to be corrected are similar to the reference answer, it is determined that the answer to the answer phrase to be corrected is correct.

[0086] Specifically, after slicing to obtain the answer phrase to be corrected and the reference answer phrase, the answer phrase to be corrected and the reference answer phrase will first be character matched. If the answer item in the answer phrase to be corrected is empty after matching, the answer is determined to be correct, otherwise it will proceed to the next step. Then all the answer items in the answer phrase to be corrected will be judged for Chinese phrase collocation errors, including typos. If it is determined that the Chinese phrase collocation is incorrect, the answer will be directly judged as incorrect, and the phrase with incorrect collocation will be returned for explanation. If the Chinese phrase collocation is correct, it will proceed to the next step. Then, all the answer items in the answer phrase to be corrected are translated into English phrases and compared with the antonym phrases of the vocabulary to be corrected. If the comparison is successful, the answer is directly judged to be wrong; if the comparison fails, all the English phrases corresponding to the answer phrase to be corrected are compared with the synonym phrases of the vocabulary to be corrected. In the process of synonym comparison, the semantics and part of speech of each group of synonyms of the vocabulary to be corrected need to be considered. If the English phrase matching the synonym is too different from the semantics and part of speech of the synonym in terms of semantics and part of speech, it cannot be directly judged as correct and enters the next step, otherwise the answer is judged to be correct. Finally, based on the Chinese semantic similarity model, the similarity between all the answer items in the answer phrase to be corrected and the reference answer phrase is calculated pairwise, and a similarity threshold is set. If it is higher than the similarity threshold, the answer item is considered to be semantically similar to the reference answer, and if it is lower than the similarity threshold, the answer item is considered to be semantically dissimilar to the reference answer, and the part of speech is considered together. Different part of speech categories are directly set as dissimilar. If the semantics of all answer items in the answer phrase to be corrected are similar to the semantics of the reference answer, the answer is judged to be correct, otherwise the answer is judged to be wrong and the wrong semantics are returned.

[0087] Among them, the application of the Chinese semantic similarity model is to apply the Bert pre-training model to the twin network for supervised training. When conducting supervised training, loss calculation is required, specifically cross-entropy loss calculation, that is, cross-entropy loss calculation is performed based on the cosine similarity between the encoded sentence vectors or word vectors, and the similarity between real sentence pairs or word pairs to obtain the Loss value; when performing reasoning, the cosine similarity between the sentence pairs or word pairs that need to be compared is calculated based on the trained Chinese semantic similarity model.

[0088] In the method shown in this embodiment, the correctness of English-Chinese translation answers at the vocabulary level is comprehensively judged by combining Chinese similarity with English similarity, and a layered correction approach is adopted, which integrates Chinese word error collocation detection, semantic features, Chinese and English parts of speech, synonyms and antonyms. This solves the problem of English-Chinese translation correction of different types of vocabulary, makes up for the shortcomings of direct character matching or judgment through semantic similarity, makes the correction effect more accurate, and promotes the further development of automatic correction technology in the field of English teaching.

[0089] Embodiment 3

[0090] See also Figure 2 The second embodiment of the present invention provides a system for correcting answer data of English-Chinese translation test questions, the system comprising: a vocabulary database creation module 10, an answer data processing module 20, an answer data processing module 30 and a comparison evaluation module 40.

[0091] The vocabulary database creation module 10 is used to obtain a batch of English words and English phrases, and to establish a basic vocabulary database based on the batch of English words and English phrases.

[0092] First of all, it should be explained that the method shown in this embodiment is used to correct the user's answer data of English-to-Chinese test questions, that is, to correct the answers written or submitted by the user. In order to improve the applicability of the method shown in this embodiment, it is necessary to batch obtain the English words and English phrases corresponding to each grade or age group or academic level, for example, obtain the newly emerging English words and English phrases corresponding to each grade in the compulsory education stage, and based on this, create a basic vocabulary library that can completely cover the scope of the examination.

[0093] In this embodiment, the acquisition of English words and English phrases can be carried out through the network, by acquiring the corresponding teaching material data, and then acquiring the corresponding English words and English phrases in the teaching material data. In the basic vocabulary library, when the same vocabulary is acquired through the network, the content of the basic vocabulary library can be pre-detected and removed when the vocabulary is stored, so as to reduce the data volume of the basic vocabulary library.

[0094] The answer data processing module 20 is used to obtain the user's answer data of the English-to-Chinese test questions, and perform slicing processing on the user's answer data to obtain answer phrases to be corrected.

[0095] Among them, the user answer data is the user's answer data for the English-Chinese translation test question, and its data type can be an image type or a text type. Among them, the image type user answer data can be obtained by scanning or photographing the answer area written by the user by a specific image acquisition device, for example, obtaining an answer sheet image, and then performing OCR cutting on the answer sheet image to obtain multiple sub-images. The data corresponding to each sub-image is the user answer data required in this embodiment, so as to correct the answer situation of the English-Chinese translation test question based on the user answer data, such as outputting scores and score feedback.

[0096] The answer data processing module 30 is used to obtain the Chinese definitions of the English words and English phrases required to be tested recorded in the basic vocabulary library, obtain reference answer data, and slice the reference answer data to obtain reference answer phrases.

[0097] In this embodiment, when slicing the reference answer data, for English words or English phrases, after the format is standardized, the textbook interpretation or broad interpretation in the basic vocabulary library is traversed and queried according to its examination scope, and the textbook interpretation or broad interpretation is cleaned and segmented to obtain a reference answer group.

[0098] The comparison and evaluation module 40 is used to compare the answer phrase to be corrected with the reference answer phrase, and use a preset hierarchical correction algorithm to evaluate the answer phrase to be corrected according to multiple evaluation indicators to output the correction result of the user's answer data.

[0099] In this embodiment, after creating a basic vocabulary library and slicing to obtain answer phrases to be corrected and reference answer question groups, the answer phrases to be corrected are compared with the reference answer question groups, and then a hierarchical correction algorithm is used to evaluate the answer items in the answer phrases to be corrected according to multiple evaluation indicators to output the correction results of the user's answer data, for example, the score and score feedback can be output to realize automatic correction of answers to English-Chinese translation questions.

[0100] Compared with the prior art, the system for correcting the answer data of the English-Chinese translation test questions shown in this embodiment has the following beneficial effects:

[0101] The answer data grading system for English-to-Chinese test questions shown in this embodiment is used to obtain English words and English phrases to create a basic vocabulary library, and then the user answer data and the reference answer data are sliced ​​and processed respectively to obtain answer phrases to be corrected and reference answer phrases. Finally, a layered grading algorithm is used to compare the answer phrases to be corrected with the reference answer phrases to output the grading results. This embodiment provides an automatic grading idea for English-to-Chinese translation of English words and English phrases, which is suitable for teachers' daily word detection of students in English teaching in primary and secondary schools. It can greatly reduce the burden of teachers' grading and improve work efficiency. At the same time, it can more conveniently assist students in memorizing English vocabulary.

[0102] Embodiment 4

[0103] A fourth embodiment of the present invention provides a readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method described in the above embodiment is implemented.

[0104] Embodiment 5

[0105] A fifth embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the method described in the above embodiments when executing the computer program.

[0106] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0107] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for those of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for correcting answer data of English-Chinese translation test questions, characterized in that: The method comprises: Acquire a batch of English words and English phrases, and establish a basic vocabulary library based on the batch of English words and English phrases; Acquire user answer data of English-Chinese translation test questions, slice the user answer data, and obtain answer phrases to be corrected; Obtaining Chinese definitions corresponding to the English words and English phrases required to be tested and recorded in the basic vocabulary database to obtain reference answer data, and slicing the reference answer data to obtain reference answer phrases; Comparing the answer phrase to be corrected with the reference answer phrase, using a preset hierarchical correction algorithm to evaluate the answer phrase to be corrected according to multiple evaluation indicators, so as to output a correction result of the user's answer data; The step of obtaining a batch of English words and English phrases and establishing a basic vocabulary library based on the batch of English words and English phrases includes: Acquire a batch of English words and English phrases, create a basic vocabulary library based on the English words and English phrases, and set the textbook interpretation and broad interpretation of each English word and English phrase in the basic vocabulary library; According to the English words and English phrases in the basic vocabulary library, the synonyms and antonyms of different meanings corresponding to each English word or English phrase are searched within the preset reference data search range, and semantic classification and annotation are performed; The step of comparing the answer phrase to be corrected with the reference answer phrase, and using a preset hierarchical correction algorithm to evaluate the answer phrase to be corrected according to multiple evaluation indicators to output the correction result of the user's answer data includes: Performing character matching on the answer phrase to be corrected and the reference answer phrase, and removing the successfully matched answer items from the answer phrase to be corrected; If the answer item in the answer phrase to be corrected is empty after polling and matching, then it is determined that the answer in the answer phrase to be corrected is correct; The steps also include: If the answer items in the answer phrase to be corrected are not empty after the polling and matching, all the answer items in the answer phrase to be corrected are judged for the correctness of the Chinese phrase collocation, including the judgment of wrong characters; If it is determined that the collocation of the Chinese phrase is wrong, then the answer to the answer item is directly determined to be wrong; If the Chinese phrase is determined to be correctly matched, all the answer items in the answer phrase to be corrected are translated into English phrases and compared with the antonym phrases of the vocabulary to be corrected. If the comparison is successful, the answer is directly judged to be wrong; The steps also include: If the comparison between the English phrase and the antonym phrase of the vocabulary to be corrected fails, then all the English phrases corresponding to the answer phrase to be corrected are compared with the synonym phrase of the vocabulary to be corrected; When the semantics and part of speech of the English phrase are different from the semantics and part of speech of the corresponding synonym, similarities are calculated between all answer items in the answer phrase to be corrected and the reference answer phrase based on the Chinese semantic similarity model, and a similarity threshold is set; When the similarity is higher than the threshold, the answer item is determined to be semantically similar to the reference answer; When the semantics of all answer items in the answer phrase to be corrected are similar to the reference answer, it is determined that the answer to the answer phrase to be corrected is correct.

2. The method for correcting answer data of English-Chinese translation test questions according to claim 1, characterized in that: The data types of the user answer data include image type and text type.

3. The method for correcting answer data of English-Chinese translation test questions according to claim 2, characterized in that: The steps of obtaining user answer data of an English-to-Chinese test question, slicing the user answer data, and obtaining answer phrases to be corrected specifically include: Acquire user answer data of an English-to-Chinese test question, and identify the data type of the user answer data; When the data type of the user answer data is a picture type, the OCR image recognition technology is used to perform OCR recognition on the user answer data to obtain text data, and the text data is cleaned and sliced ​​to obtain the answer phrase to be corrected; When the data type of the user answer data is text data of a text type, data cleaning and data slicing are performed on the text data to obtain answer phrases to be corrected.

4. A system for correcting answer data of English-Chinese test questions, characterized in that: The method applied to any one of claims 1 to 3, wherein the system comprises: A vocabulary library creation module is used to obtain a batch of English words and English phrases, and to establish a basic vocabulary library based on the batch of English words and English phrases; The answer data processing module is used to obtain the user's answer data of the English-Chinese translation test questions, slice the user's answer data, and obtain the answer phrases to be corrected; The answer data processing module is used to obtain the Chinese definitions of the English words and English phrases required to be tested recorded in the basic vocabulary library, obtain reference answer data, and slice the reference answer data to obtain reference answer phrases; The comparison and evaluation module is used to compare the answer phrase to be corrected with the reference answer phrase, and use a preset hierarchical correction algorithm to evaluate the answer phrase to be corrected according to multiple evaluation indicators to output the correction result of the user's answer data.

5. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and capable of being executed on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 3 is implemented.

Images