International Chinese education vocabulary teaching knowledge personalized response method and device

Abstract

The embodiment of the invention provides a personalized response method and device for international Chinese education vocabulary teaching knowledge. The method comprises the steps that question information input by a user is received; vectorizing the problem information to obtain a problem vector; searching matched entry information from a pre-created vocabulary knowledge resource library according to the problem vector; taking the retrieved entry information as a context, and creating a response template; and obtaining user information, generating personalized statements according to the response template and the user information through a pre-trained large language model, outputting the personalized statements, and correcting wrong answers of the basic large language model to word usage knowledge. Through the scheme of the embodiment of the invention, after the question information input by the user is received, the matched entry information can be retrieved from the pre-created vocabulary knowledge resource library according to the question vector, and the personalized statement is generated and output according to the entry information and the user information by obtaining the user information. Therefore, the purpose of individually answering the questions proposed by the user according to the user information is achieved.

Description

Technical Field

[0001] The present application relates to the field of information technology, and particularly to a method and device for personalized response to knowledge of teaching international Chinese education vocabulary. Background Art

[0002] Teaching Chinese as a foreign language refers to an educational activity that targets non-native Chinese speakers and helps learners master Chinese language skills, understand Chinese culture, and cultivate cross-cultural communication skills through systematic language teaching and cultural dissemination. During the teaching process, through dictionary tools and the like, it is already possible to provide static and comprehensive supply of knowledge about word usage, but it is impossible to answer users' questions in a personalized and targeted manner. At the same time, when current AI (Artificial Intelligence) large model tools answer users' questions, they often just answer through fixed templates, which is not convenient for users to understand and learn. Summary of the Invention

[0003] The purpose of the embodiments of the present application is to provide a method and device for personalized response to knowledge of teaching international Chinese education vocabulary, so as to achieve the purpose of answering users' questions in a personalized manner according to user information. The specific technical solutions are as follows:

[0004] In the first aspect of the embodiments of the present application, first, a method for personalized response to knowledge of teaching international Chinese education vocabulary is provided. The method includes:

[0005] Receiving problem information input by the user for a Chinese vocabulary;

[0006] Vectorizing the problem information for the Chinese vocabulary to obtain a problem vector;

[0007] According to the problem vector, retrieving matching entry information from a pre-created Chinese vocabulary knowledge resource library;

[0008] Creating a response template with the retrieved entry information as the context;

[0009] Obtaining user information, and generating and outputting personalized sentences according to the response template and the user information through a pre-trained large language model.

[0010] In a possible implementation manner, the retrieving matching entry information from a pre-created Chinese vocabulary knowledge resource library according to the problem vector includes:

[0011] According to the problem vector, retrieving a matching entry from a pre-created Chinese knowledge base, as well as multiple types of interpretations and multiple types of example sentences of the entry;

[0012] Creating a response template with the retrieved entry information as the context, including:

[0013] Creating the response template with the retrieved entry, as well as various types of definitions and various types of examples of the entry, as the context.

[0014] In a possible implementation manner, the obtaining user information and generating and outputting a personalized statement according to the response template and the user information through a pre-trained large language model includes:

[0015] Obtaining the mode selection information of the user, where the mode selection indicates selecting an easy language mode and / or a professional language mode;

[0016] Selecting a target type of definition from the various types of definitions and a target type of example from the various types of examples according to the mode selection information of the user, where the target type of definition corresponding to the easy language mode is a popular type of explanation, the target type of example corresponding to the easy language mode is a life type of example, the target type of definition corresponding to the professional language mode is a form grammar type of explanation, and the target type of example corresponding to the professional language mode is an academic type of example;

[0017] Generating and outputting the personalized statement through a pre-trained large language model according to the target type of definition and the target type of example.

[0018] In a possible implementation manner, the method further includes:

[0019] Receiving a display request for a target vocabulary;

[0020] Obtaining and outputting the picture information and audio information corresponding to the target vocabulary, where the picture information and / or the audio information includes the meaning, form grammar, and examples of the target vocabulary.

[0021] In a possible implementation manner, the method further includes:

[0022] Receiving a practice request for a target vocabulary;

[0023] Generating and outputting practice questions according to the target vocabulary, where the practice questions are one of multiple-choice questions, fill-in-the-blank questions, and error-correction questions.

[0024] In the second aspect of the embodiments of the present application, an international Chinese education vocabulary teaching knowledge personalized response device is provided, and the device includes:

[0025] A question receiving module, configured to receive question information of the user input for Chinese vocabulary;

[0026] A question vectorization module, configured to vectorize the question information for the Chinese vocabulary to obtain a question vector;

[0027] A term matching module, configured to retrieve matching term information from a pre-created Chinese vocabulary knowledge resource library according to the question vector;

[0028] A template creation module, configured to create a response template with the retrieved term information as the context;

[0029] A statement generation module, configured to obtain user information, and generate and output personalized statements according to the response template and the user information through a pre-trained large language model.

[0030] In a possible implementation manner, the term matching module is specifically configured to retrieve matching terms, as well as various types of interpretations and various types of example sentences of the terms, from a pre-created Chinese knowledge base according to the question vector;

[0031] The template creation module is specifically configured to create the response template with the retrieved terms, as well as various types of interpretations and various types of example sentences of the terms, as the context.

[0032] In a possible implementation manner, the statement generation module is specifically configured to obtain the user's mode selection information, where the mode selection indicates selecting an easy language mode and / or a professional language mode; select a target type of interpretation from the various types of interpretations according to the user's mode selection information, and select a target type of example sentence from the various types of example sentences, where the target type of interpretation corresponding to the easy language mode is a popular type of explanation, the target type of example sentence corresponding to the easy language mode is a life type of example sentence, the target type of interpretation corresponding to the professional language mode is a form grammar type of explanation, and the target type of example sentence corresponding to the professional language mode is an academic type of example sentence; generate and output the personalized statement according to the target type of interpretation and the target type of example sentence through a pre-trained large language model.

[0033] In a possible implementation manner, the term matching module is specifically configured to retrieve matching terms, as well as the meaning, form grammar, and example sentences of the terms, from a pre-created knowledge base according to the question vector;

[0034] The template creation module is specifically configured to create the response template with the retrieved terms, as well as the meaning, left and right context collocations, and example sentences of the terms, as the context.

[0035] In a possible implementation manner, the statement generation module is specifically configured to obtain the user's mode selection information, where the mode selection indicates the selection of the plain language mode and / or the professional language mode; select a target type of paraphrase from the multiple types of paraphrases according to the user's mode selection information, and select a target type of example sentence from the multiple types of example sentences, where the target type of paraphrase corresponding to the plain language mode is a popular type of explanation, and the target type of example sentence corresponding to the plain language mode is a life type of example sentence, the target type of paraphrase corresponding to the professional language mode is a type grammar type of explanation, and the target type of example sentence corresponding to the professional language mode is an academic type of example sentence; generate and output the personalized statement according to the target type of paraphrase and the target type of example sentence through a pre-trained large language model.

[0036] In a possible implementation manner, the device further includes:

[0037] The vocabulary display module is configured to receive a display request for a target vocabulary; obtain and output the picture information and audio information corresponding to the target vocabulary, where the picture information and / or the audio information includes the meaning of the target vocabulary, the left and right context collocations, and example sentences.

[0038] In a possible implementation manner, the device further includes:

[0039] The exercise output module is configured to receive an exercise request for a target vocabulary; generate and output exercises according to the target vocabulary, where the exercises are one of multiple-choice questions, fill-in-the-blank questions, and error-correction questions.

[0040] On the other hand, an embodiment of the present application further provides an electronic device, including:

[0041] A memory for storing a computer program;

[0042] A processor, when executing the program stored in the memory, implements the personalized response method for international Chinese education vocabulary teaching knowledge described in any one of the above.

[0043] On the other hand, an embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, and when the computer program is executed by a processor, it implements the personalized response method for international Chinese education vocabulary teaching knowledge described in any one of the above.

[0044] On the other hand, an embodiment of the present application further provides a computer program product containing instructions, which, when running on a computer, causes the computer to execute the personalized response method for international Chinese education vocabulary teaching knowledge described in any one of the above.

[0045] Advantages of the embodiments of the present application:

[0046] The embodiments of the present application provide a method and device for personalized response to knowledge of international Chinese education vocabulary teaching. The method includes: receiving problem information about Chinese vocabulary input by a user; vectorizing the problem information about Chinese vocabulary to obtain a problem vector; retrieving matching entry information from a pre-created Chinese vocabulary knowledge resource library according to the problem vector; creating a response template with the retrieved entry information as the context; obtaining user information, and generating and outputting a personalized statement according to the response template and the user information through a pre-trained large language model. Through the solution of the embodiments of the present application, after receiving the problem information input by the user, matching entry information can be retrieved from the pre-created Chinese vocabulary knowledge resource library according to the problem vector, and by obtaining user information, a personalized statement can be generated and output according to the entry information and the user information, so as to achieve the purpose of answering the questions raised by the user in a personalized manner according to the user information, which can not only improve the user experience, but also increase the interest and learning motivation of learning.

[0047] Of course, it is not necessary for any product or method implementing the present application to achieve all the above advantages simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application, and those of ordinary skill in the art can also obtain other embodiments according to these drawings.

[0049] Figure 1 It is a schematic flowchart of a method for personalized response to knowledge of international Chinese education vocabulary teaching provided by the embodiments of the present application;

[0050] Figure 2 It is a schematic flowchart of generating a personalized statement provided by the embodiments of the present application;

[0051] Figure 3 It is a schematic flowchart of the learning function in the learning tool provided by the embodiments of the present application;

[0052] Figure 4 It is a schematic diagram of the learning function in the learning tool provided by the embodiments of the present application;

[0053] Figure 5 It is a schematic diagram of the intelligent question answering assistant provided by the embodiments of the present application;

[0054] Figure 6A schematic diagram of the scenario practice mode provided by the embodiments of the present application;

[0055] Figure 7 A schematic diagram of the exploration and query mode provided by the embodiments of the present application;

[0056] Figure 8 A schematic structural diagram of the personalized response device for international Chinese education vocabulary teaching knowledge provided by the embodiments of the present application;

[0057] Figure 9 A schematic structural diagram of an electronic device provided by the embodiments of the present application. Detailed implementation manners

[0058] Next, the technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art based on the present application belong to the scope of protection of the present application.

[0059] In the first aspect of the embodiments of the present application, first, a method for personalized response of international Chinese education vocabulary teaching knowledge is provided. Refer to Figure 1 , Figure 1 A schematic flowchart of the method for personalized response of international Chinese education vocabulary teaching knowledge provided by the embodiments of the present application. The method includes:

[0060] Step S11: Receive the problem information for Chinese vocabulary input by the user;

[0061] Step S12: Vectorize the problem information for the Chinese vocabulary to obtain a problem vector;

[0062] Step S13: Retrieve the matching entry information from the pre-created Chinese vocabulary knowledge resource library according to the problem vector;

[0063] Step S14: Create a response template with the retrieved entry information as the context;

[0064] Step S15: Obtain user information, and generate and output personalized sentences according to the response template and the user information through a pre-trained large language model.

[0065] First of all, it should be noted that the method of the embodiments of the present application is applied to Chinese learning tools. Specifically, the tool can be a learning tool for users learning Chinese as a foreign language. For example, it can be computer software or a mobile APP (Application).

[0066] Corresponding to the above step S11, receiving the problem information for Chinese vocabulary input by the user can receive the user's questions regarding Chinese language learning. The solution of this application is applied to vocabulary teaching in the field of international Chinese education. Specifically, the user can be a college student learning Chinese, and the user can ask Chinese questions through the solution of this application. For example: Why did the teacher tell me that the expression "tell a story" is incorrect?

[0067] Corresponding to the above step S12, vectorize the problem information for the Chinese vocabulary to obtain a problem vector, which can be vectorized by various methods. In one example, it can be vectorized by basic statistical methods, such as the bag-of-words model, term frequency and inverse document frequency weighting. In another example, it can be vectorized by semantic embedding methods. In one example, codes corresponding to each word or character can be preset in advance, and then the codes can be combined according to the order in the problem information to obtain the problem vector.

[0068] Corresponding to the above step S13, when retrieving the matching entry information from the pre-created Chinese vocabulary knowledge resource library according to the problem vector, the pre-created knowledge library can include vectors corresponding to multiple entries or questions, as well as the original text and metadata corresponding to the entry or question, such as ID (Identity document, identity card identification number), etc. Then, by calculating the similarity between the problem vector and each vector in the knowledge library, one or more with the highest similarity are selected as the matching entry information. For example, when matching the vector corresponding to "tell", multiple sets of matching results are obtained, such as: tell; to state or inform someone of something, making them know, and the general grammatical structure is "tell + [who] + [what]"; He told me about this matter; say; to utter or state, and the general grammatical structure is "say + [story]"; He told me a story.

[0069] Corresponding to the above step S14, when creating a response template with the retrieved entry information as the context, when only one entry information is retrieved, the response template can be directly created. When there are multiple entries, the retrieved entry information can be further screened through context analysis, and then one or more of the screened entry information are used as the context to create the response template, thereby obtaining one or more templates. For example, further screening the retrieved entry information can reveal that the corresponding problem answer lies in the issue of left and right context collocation. Then, the response template can be created based on this explanation, and this template can be a Prompt template. Through this Prompt template, it can be used for the preset framework of structured generation instructions, and improve the efficiency and accuracy of interacting with the large model through parameterized input and standardized processes.

[0070] Corresponding to the above step S15, when obtaining user information and generating personalized sentences according to the response template and the user information through a pre-trained large language model, a template can be selected from multiple generated templates as the final template for personalized sentence generation according to the user information. Specifically, the similarity between the user information and multiple targets can be calculated, and the one with the highest similarity can be selected as the final template. It is also possible to preset the type corresponding to the user information or the selected type, and the type corresponding to the response template, so as to select one belonging to the same type as the final template for personalized sentence generation. In this case, the user information can be the type selected by the user or the type corresponding to the preset user. Thus, through the finally selected template, a personalized sentence is generated through a pre-trained large language model, such as a RAG (Retrieval-Augmented Generation) large model, in combination with the response template Prompt template. Among them, the pre-trained large language model can be a pre-trained template for personalized sentence generation. Through this model, the response template can be used as the context input to guide the generation of more demand-compliant answers, solving problems such as the "hallucination" and knowledge limitations of traditional large models. In the actual use process, the pre-trained large language model can also be corrected according to the output result, so as to not only ensure that the large language model meets the actual needs, but also utilize the advantages of the pre-created Chinese vocabulary knowledge resource library to improve the applicability of the model.

[0071] It can be seen that through the solution of the embodiment of the present application, after receiving the question information input by the user, the corresponding entry information can be retrieved from the pre-created Chinese vocabulary knowledge resource library according to the question vector, and by obtaining the user information, a personalized sentence is generated and output according to the entry information and the user information, so as to achieve the purpose of answering the question raised by the user in a personalized manner according to the user information, which can not only improve the user experience, but also increase the fun and interest of learning.

[0072] In a possible implementation manner, retrieving the matching entry information from the pre-created Chinese vocabulary knowledge resource library according to the problem vector includes: retrieving the matching entry from the pre-created Chinese knowledge library according to the problem vector, as well as various types of definitions and various types of example sentences of the entry; creating a response template with the retrieved entry information as the context, including: creating the response template with the retrieved entry, as well as various types of definitions and various types of example sentences of the entry, as the context. To meet the needs of different users, the pre-created knowledge library in the embodiments of the present application may include various types of definitions and various types of example sentences. For example, popular type definitions and professional type definitions. Specifically, for example, the popular type definition of "tell" is that someone says to another person, so there must be an address of the listener following "tell", while the corresponding professional type definition is that someone informs another person to make the other person understand something; "tell" in Chinese should be followed by a double object. Another example is the cartoon type definition applicable to children and the ordinary type definition applicable to adults, etc.

[0073] In a possible implementation manner, referring to Figure 2 , obtaining the user information, and generating and outputting a personalized statement according to the response template and the user information through a pre-trained large language model, includes:

[0074] Step S21, obtaining the mode selection information of the user, where the mode selection indicates the selection of the plain language mode and / or the professional language mode;

[0075] Step S22, selecting a target type of definition from the various types of definitions and a target type of example sentence from the various types of example sentences according to the mode selection information of the user, where the target type of definition corresponding to the plain language mode is the popular type of explanation, the target type of example sentence corresponding to the plain language mode is the life type of example sentence, the target type of definition corresponding to the professional language mode is the form grammar type of explanation, and the target type of example sentence corresponding to the professional language mode is the academic type of example sentence;

[0076] Step S23, generating and outputting the personalized statement through the pre-trained large language model according to the target type of definition and the target type of example sentence.

[0077] Among them, obtaining the user's mode selection information can determine whether the mode selected by the user is the plain language mode or the professional language mode. Thus, when the user selects the plain language mode, the corresponding plain language mode's popular type of explanations and life type of example sentences are selected, and the personalized statement is generated and output through a pre-trained large language model. Among them, the paraphrase can include explanations and grammar collocations. For example, the explanation is: "tell" means someone says to another person; the grammar collocation is: tell + [who] + [what]; example sentence, "He told me it will rain tomorrow." Or when the user selects the professional language mode, the personalized statement is generated and output according to example sentences of different academic types with different grammatical structures. In the actual use process, when obtaining the user's mode selection information, it is also possible to choose to obtain all the information of word usage knowledge or personalized answers to questions. For example, see Figure 3 , when the solution of the embodiment of the present application is applied to the mobile phone APP, after the learner registers and logs in, personalized learning goal settings can be carried out, and then by selecting learning functions, different modes can be selected. For example, vocabulary learning mode, scenario practice mode, retrieval query mode, and intelligent Q&A assistant. The solution of the above embodiment can be implemented through the intelligent Q&A assistant; then different modes are used in combination according to the learning situation, and shared in the community and learning progress is managed. See Figure 4 , when the solution of the embodiment of the present application is implemented through the intelligent Q&A assistant, word usage knowledge and word usage corpus can be obtained through knowledge resource data cleaning, and then combined with the RAG knowledge base and Prompt engineering, using the local deployment of the large language model, through the QA (a retrieval system) retrieval system, similarity analysis and Prompt injection problem - knowledge module matching are carried out to obtain the personalized statement and output. When the retrieval system retrieves, the user input question will also be considered, and this input question can be input through the input end UI (User Interface) interface. In one example, the obtained personalized statement is: The sentence "He told a story" has a problem, mainly because the usage of the verb "tell" does not conform to the Chinese habit. According to the grammar rules of modern Chinese, "tell" is usually used in the structure of "tell + [who] + [what thing]", that is, it is necessary to clearly point out the object and content of telling. And "story" as a specific content usually needs to be expressed with the verb "tell", that is, "tell a story", and the correct expression should be "He tells a story". The following are 3 related example sentences: 1. Correct usage: tell + [story]; example sentence: He tells stories to his children every night. 2. Correct usage: tell + [who] + [what thing]; example sentence: He told me it will rain tomorrow. 3. Correct usage: tell + [story]; example sentence: The teacher is telling an interesting story to the students in the classroom.

[0078] In a possible implementation, the method further includes: receiving a display request for a target vocabulary; obtaining and outputting the picture information and audio information corresponding to the target vocabulary, where the picture information and / or the audio information include the meaning, left and right context collocations, and example sentences of the target vocabulary. Specifically, it is possible to combine pictures, audio, and card-based interactions to display the left and right context grammar patterns of words grouped by HSK (Hanyu Shuiping Kaoshi, Chinese Proficiency Test) level or theme. By extracting entries from a database with reliable vocabulary meaning and usage knowledge and grouping them by HSK level or theme. On the basis of displaying the meaning, example sentences, and pattern grammar collocations of each word, combining multimodal elements such as pictures and audio, and using card-based learning and progress tracking as the interaction design to improve the learning effect. When organizing knowledge content, it is possible to use requirements and services as the guiding principle for dictionary annotation methods, focusing on "local grammar" and "pattern grammar", and describing the usage of some commonly used modern Chinese vocabulary in different contexts at the language level and communicative level, assisting students in understanding in the context and discourse, and effectively improving the pertinence and validity of the results generated by AI (Artificial Intelligence). For the common meaning and usage of words, the method of combining standard example sentences and daily life examples can be adopted, selecting numerous example sentences from conversations, literary works, film and television works, or generated by AI specifically for second language learners to study and practice. In one example, the target vocabulary can be a vocabulary in a personalized sentence. See Figure 5 , the corresponding embodiment of this paragraph can be implemented through a vocabulary learning mode. After selecting the learning theme and HSK level information, learning cards are displayed; then, it is implemented through basic meaning explanations, left and right context usage displays, example sentences, misusage examples, and relevant pictures and audio, and the learning status and progress can also be recorded.

[0079] In a possible implementation, the method further includes: receiving a practice request for a target vocabulary; generating and outputting practice questions according to the target vocabulary, where the practice questions are one of multiple-choice questions, fill-in-the-blank questions, and error-correction questions. Specifically, it is possible to dynamically generate multiple-choice questions, fill-in-the-blank questions, and error-correction questions, and recommend review content based on the user's performance. Multiple question types can be generated in the forms of selection, error correction, filling in the blanks, etc., including static questions adapted from the example sentences in the word book and stored in the database and dynamic questions generated using large language models, and with the help of the correct context collocations or example sentences prompted after answering wrong for analysis, recording the user's correct rate, recommending weak words for review and other feedback mechanisms to help users quickly master the usage and consolidate knowledge in the context-based Chinese learning scenario. In one example, the target vocabulary can be a vocabulary in a personalized sentence. See Figure 6The solution corresponding to the embodiment of this paragraph can be implemented through the situational practice mode. By identifying the user's learning progress and learning records, entering the practice interface, and then selecting filling in the blanks and correcting errors, listening to words and filling in words, looking at pictures and selecting words to fill in words, and reading and following sentences, the above functions can be realized. Different modes can be used according to the learning situation, the accuracy rate can be recorded, and a review plan can be recommended.

[0080] In actual use, the tool corresponding to the scheme of the embodiment of the present application can also be compatible with traditional word book queries, supporting fuzzy queries and grammatical hierarchical indexing. After the user enters a word or condition, the matching entry is retrieved from the vector database or traditional database, the system analyzes the meaning of the question and performs similarity calculation and classification decision in the knowledge base, interacts the knowledge module of the knowledge to which it belongs with the answer to the question in the large language model, and displays the meaning, usage, example sentences and left and right context collocation of the word in the result. In order to improve the query efficiency, the system can adopt a hierarchical storage strategy for the language knowledge involved, and subdivide the knowledge resources according to the standards of "basic pragmatic knowledge", "advanced communication knowledge" and "common misuse examples and misuse cause analysis". Basic pragmatic knowledge covers the common collocation and grammatical rules of the target vocabulary, and provides a knowledge basis for the question-answering system by describing in detail the basic usage of the target vocabulary in different contexts. On the basis of basic pragmatic knowledge, context-level advanced communication knowledge further explores the usage of the target vocabulary in specific communication scenarios. In the section on common misuse and misuse examples and misuse reasons analysis, we can collect and analyze cases in the dynamic composition corpus, list in detail the common misuse and misuse in the process of learning Chinese and provide guidance. Figure 7 The solution corresponding to the embodiment of this paragraph can be realized by exploring the query mode, by loading the dictionary resources and the built-in query system, searching the query system, and combining the input of the input UI interface with the word meaning that is expected to be matched, to obtain the query result, such as, corresponding to "put", we get: 1. Basic meaning; 1. Disposal meaning: means to dispose of or influence something; 2. Causative meaning: means to make something produce a certain result or state change; 2. Core usage; (a) Basic structure; 1. [agent] + "put" + [patient] + [verb] + [complement / ".

[0081] The second aspect of the embodiment of the present application provides a personalized response device for international Chinese education vocabulary teaching knowledge, see Figure 8 , the device comprises:

[0082] The question receiving module 801 is used to receive question information about Chinese vocabulary input by the user;

[0083] A question vectorization module 802 is used to vectorize the question information for the Chinese vocabulary to obtain a question vector;

[0084] The entry matching module 803 is used to retrieve matching entry information from a pre-created Chinese vocabulary knowledge resource library according to the problem vector;

[0085] The template creation module 804 is used to create a response template with the retrieved entry information as the context;

[0086] The statement generation module 805 is used to obtain user information and generate and output personalized statements according to the response template and the user information through a pre-trained large language model.

[0087] In a possible implementation manner, the entry matching module is specifically configured to retrieve matching entries from a pre-created Chinese knowledge base according to the problem vector, as well as various types of interpretations and various types of example sentences of the entry;

[0088] The template creation module is specifically configured to create the response template with the retrieved entry, as well as various types of interpretations and various types of example sentences of the entry, as the context.

[0089] In a possible implementation manner, the statement generation module is specifically configured to obtain the mode selection information of the user, where the mode selection indicates selecting an easy language mode and / or a professional language mode; select a target type of interpretation from the various types of interpretations according to the mode selection information of the user, and select a target type of example sentence from the various types of example sentences, where the target type of interpretation corresponding to the easy language mode is a popular type of explanation, the target type of example sentence corresponding to the easy language mode is a life type of example sentence, the target type of interpretation corresponding to the professional language mode is a formal grammar type of explanation, and the target type of example sentence corresponding to the professional language mode is an academic type of example sentence; generate and output the personalized statement according to the target type of interpretation and the target type of example sentence through a pre-trained large language model.

[0090] In a possible implementation manner, the entry matching module is specifically configured to retrieve matching entries from a pre-created knowledge base according to the problem vector, as well as the meaning, grammatical form, and example sentences of the entry;

[0091] The template creation module is specifically configured to create the response template with the retrieved entry, as well as the meaning, formal grammar, and example sentences of the entry, as the context.

[0092] In a possible implementation, the statement generation module is specifically configured to obtain the user's mode selection information, where the mode selection indicates the selection of the plain language mode and / or the professional language mode; select a target type of paraphrase from the multiple types of paraphrases according to the user's mode selection information, and select a target type of example sentence from the multiple types of example sentences, where the target type of paraphrase corresponding to the plain language mode is a popular type of explanation, the target type of example sentence corresponding to the plain language mode is a life type of example sentence, the target type of paraphrase corresponding to the professional language mode is a type grammar type of explanation, and the target type of example sentence corresponding to the professional language mode is an academic type of example sentence; generate and output the personalized statement according to the target type of paraphrase and the target type of example sentence through a pre-trained large language model.

[0093] In a possible implementation, the device further includes:

[0094] The vocabulary display module is configured to receive a display request for a target vocabulary; obtain and output the picture information and audio information corresponding to the target vocabulary, where the picture information and / or the audio information includes the meaning, type grammar and example sentences of the target vocabulary.

[0095] In a possible implementation, the device further includes:

[0096] The exercise output module is configured to receive an exercise request for a target vocabulary; generate and output exercises according to the target vocabulary, where the exercises are one of multiple-choice questions, fill-in-the-blank questions and error-correction questions.

[0097] It can be seen that through the solution of the embodiments of the present application, after receiving the question information input by the user, according to the question vector, the matching entry information can be retrieved from the pre-created Chinese vocabulary knowledge resource library, and by obtaining the user information, the personalized statement can be generated and output according to the entry information and the user information, so as to achieve the purpose of answering the questions raised by the user according to the user information in a personalized manner, which can not only improve the user experience, but also increase the interest and learning motivation of learning.

[0098] The embodiments of the present application also provide an electronic device, as Figure 9 shown, including:

[0099] A memory 901 for storing a computer program;

[0100] A processor 902, when executing the program stored in the memory 901, implements the following steps:

[0101] Receive the question information input by the user for a Chinese vocabulary;

[0102] Vectorize the problem information for the Chinese vocabulary to obtain a problem vector;

[0103] Retrieve matching entry information from a pre-created Chinese vocabulary knowledge repository according to the problem vector;

[0104] Create a response template with the retrieved entry information as the context;

[0105] Obtain user information, and generate and output personalized statements according to the response template and the user information through a pre-trained large language model.

[0106] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity, only a thick line is used to represent it in the figure, but it does not mean that there is only one bus or one type of bus.

[0107] The communication interface is used for communication between the above electronic device and other devices.

[0108] The memory can include a Random Access Memory (RAM), and can also include a Non-Volatile Memory (NVM), such as at least one disk memory. Optionally, the memory can also be at least one storage device located far from the aforementioned processor.

[0109] The above-mentioned processor can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0110] In another embodiment provided by the present application, a computer-readable storage medium is further provided. A computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the steps of any of the above-mentioned personalized response methods for international Chinese education vocabulary teaching knowledge are implemented.

[0111] In another embodiment provided by the present application, a computer program product containing instructions is further provided. When it runs on a computer, the computer is caused to execute any of the personalized response methods for international Chinese education vocabulary teaching knowledge in the above embodiments.

[0112] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from a website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more integrated available media. The available medium may be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a solid-state disk (SSD), etc.

[0113] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variation thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the element.

[0114] Each embodiment in this specification is described in a related manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the device, electronic device, and storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiments.

[0115] The above are only the preferred embodiments of the present application and are not intended to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application are all included in the protection scope of the present application.

Claims

1. A personalized response method for teaching knowledge of international Chinese education vocabulary, characterized in that The method includes: Receiving problem information for Chinese words input by the user; Vectorizing the problem information for the Chinese words to obtain a problem vector; Retrieving matching entry information from a pre-created Chinese word knowledge resource library according to the problem vector; Creating a response template with the retrieved entry information as the context; Obtaining user information, and generating and outputting a personalized statement according to the response template and the user information through a pre-trained large language model.

2. The method according to claim 1, wherein The retrieving matching entry information from a pre-created Chinese word knowledge resource library according to the problem vector includes: Retrieving a matching entry from a pre-created Chinese knowledge library according to the problem vector, as well as various types of interpretations and various types of example sentences of the entry; The creating a response template with the retrieved entry information as the context includes: Creating the response template with the retrieved entry, as well as various types of interpretations and various types of example sentences of the entry, as the context.

3. The method according to claim 2, characterized in that, The obtaining user information, and generating and outputting a personalized statement according to the response template and the user information through a pre-trained large language model includes: Obtaining the user's mode selection information, where the mode selection indicates the selection of an easy language mode and / or a professional language mode; Selecting a target type of interpretation from the various types of interpretations and a target type of example sentence from the various types of example sentences according to the user's mode selection information, where the target type of interpretation corresponding to the easy language mode is a popular type of explanation, the target type of example sentence corresponding to the easy language mode is a life type of example sentence, the target type of interpretation corresponding to the professional language mode is a form grammar type of explanation, and the target type of example sentence corresponding to the professional language mode is an academic type of example sentence; Generating and outputting the personalized statement through a pre-trained large language model according to the target type of interpretation and the target type of example sentence.

4. The method according to claim 1, wherein The method further includes: Receiving a display request for a target word; Obtaining and outputting the picture information and audio information corresponding to the target word, where the picture information and / or the audio information includes the meaning of the target word, the left and right context collocations, and example sentences.

5. The method according to claim 4, wherein The method further includes: Receiving a practice request for a target word; Generating and outputting a practice question according to the target word, where the practice question is one of a multiple-choice question, a fill-in-the-blank question, and a correction question.

6. An individualized response device for teaching knowledge of Chinese language education vocabulary in the international context, characterized in that, The device includes: A problem receiving module for receiving problem information for Chinese words input by the user; A problem vectorizing module for vectorizing the problem information for the Chinese words to obtain a problem vector; An entry matching module for retrieving matching entry information from a pre-created Chinese word knowledge resource library according to the problem vector; A template creating module for creating a response template with the retrieved entry information as the context; A statement generation module, configured to obtain user information, and generate and output personalized statements according to the response template and the user information through a pre-trained large language model.

7. The apparatus according to claim 6, wherein The entry matching module is specifically configured to retrieve matching entries, as well as various types of paraphrases and various types of example sentences of the entry, from a pre-created Chinese knowledge base according to the question vector; The template creation module is specifically configured to create the response template with the retrieved entry, as well as various types of paraphrases and various types of example sentences of the entry, as the context.

8. The apparatus according to claim 7, wherein The statement generation module is specifically configured to obtain the mode selection information of the user, wherein the mode selection indicates the selection of the plain language mode, and / or, the professional language mode; select a target type of paraphrase from the various types of paraphrases according to the mode selection information of the user, and select a target type of example sentence from the various types of example sentences, wherein the target type of paraphrase corresponding to the plain language mode is a popular type of explanation, the target type of example sentence corresponding to the plain language mode is a life type of example sentence, the target type of paraphrase corresponding to the professional language mode is a formal grammar type of explanation, and the target type of example sentence corresponding to the professional language mode is an academic type of example sentence; generate and output the personalized statement according to the target type of paraphrase and the target type of example sentence through a pre-trained large language model.

9. An electronic device, characterized in that, Comprising: A memory for storing a computer program; A processor, when executing the program stored on the memory, implements the personalized response method for international Chinese education vocabulary teaching knowledge according to any one of claims 1-5.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, it implements the personalized response method for international Chinese education vocabulary teaching knowledge according to any one of claims 1-5.

Images