Information processing method and device, equipment, storage medium and computer program product

Abstract

The invention discloses an information processing method. The method comprises the following steps: acquiring a learning attribute parameter and a user performance parameter of a to-be-analyzed object; based on the learning attribute parameters and the user performance parameters, constructing a user portrait of the to-be-analyzed object; determining a course planning knowledge graph corresponding to the to-be-analyzed object; and determining a to-be-recommended learning path based on the user portrait, the learning target and the course planning knowledge graph. The invention further discloses an information processing device and equipment, a storage medium and a computer program product.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an information processing method, device, equipment, storage medium and computer program product. Background Art

[0002] Traditional education methods usually adopt a unified teaching plan and progress, which cannot meet the personalized learning needs of different learners. With the development of artificial intelligence and machine learning technology, adaptive learning systems have emerged, which can adjust teaching strategies according to learners' real-time feedback and performance. Currently, commonly used adaptive learning systems can dynamically generate or adjust teaching content according to students' learning levels. However, current adaptive learning systems lack the ability to perceive learning situations and cannot adjust teaching strategies according to students' emotions, environment and other factors, resulting in users often feeling limited in their experience during use and unable to obtain an immersive experience comparable to the real world.

[0003] Application Contents

[0004] In order to solve the above technical problems, the present application hopes to provide an information processing method, apparatus, equipment, storage medium and computer program product, which solves the problem that the current adaptive learning system lacks perception of learning context, and proposes an adaptive learning recommendation method, which comprehensively considers multiple factors such as personalized learning, automatically adjusts the teaching content and difficulty according to the user's learning progress and ability, effectively realizes the personalized learning needs of different users, and improves learning efficiency and quality.

[0005] The technical solution of this application is implemented as follows:

[0006] The present application provides an information processing method, the method comprising:

[0007] Obtaining learning attribute parameters and user performance parameters of the object to be analyzed;

[0008] Constructing a user profile of the object to be analyzed based on the learning attribute parameters and the user performance parameters;

[0009] Determine the course planning knowledge graph corresponding to the object to be analyzed;

[0010] Based on the user portrait, learning objectives and the course planning knowledge graph, a learning path to be recommended is determined.

[0011] In the above scheme, the learning attribute parameters include at least one or more of the following parameters: learning behavior parameters and performance parameters; the user performance parameters include at least one or more of the following parameters: interest parameters and social parameters.

[0012] In the above scheme, the method further comprises:

[0013] Obtaining learning progress and learning feedback information of the object to be analyzed;

[0014] The learning path is updated based on the learning progress and the learning feedback information to obtain an updated learning path.

[0015] In the above scheme, the method further comprises:

[0016] In the process of the object to be analyzed learning based on the learning path, determining a subject knowledge graph;

[0017] If the information to be queried is detected, the retrieval information of the information to be queried is determined based on the subject knowledge graph;

[0018] The search information is output.

[0019] In the above scheme, the method further comprises:

[0020] Obtain the knowledge points of subject corpus;

[0021] Extracting specific knowledge content from the knowledge point content and subject corpus through the trained knowledge extraction large language model;

[0022] The subject knowledge graph is generated based on the knowledge point content and the specific content of the knowledge.

[0023] In the above solution, if the information to be queried is detected, the retrieval information of the information to be queried is determined based on the subject knowledge graph, including:

[0024] If the information to be queried is detected, a trained content retrieval language model is obtained;

[0025] The content retrieval large language model is used to retrieve information corresponding to the information to be queried from the subject knowledge graph to obtain the retrieval information.

[0026] In the above solution, the content retrieval large language model has a retrieval enhancement generation function.

[0027] In the above solution, the outputting of the search information includes:

[0028] Performing sentiment analysis on the retrieved information to obtain sentiment information;

[0029] The search information is reported in voice form according to the emotional information through a virtual teaching assistant image.

[0030] The present application provides an information processing device, the device at least comprising: an acquisition unit, a construction unit, a first determination unit and a second determination unit; wherein:

[0031] The acquisition unit is used to acquire the learning attribute parameters and user performance parameters of the object to be analyzed;

[0032] The construction unit is used to construct a user profile of the object to be analyzed based on the learning attribute parameters and the user performance parameters;

[0033] The first determining unit is used to determine the course planning knowledge graph corresponding to the object to be analyzed;

[0034] The second determination unit is used to determine the learning path to be recommended based on the user portrait, the learning goal and the course planning knowledge graph.

[0035] The present application provides an information processing device, the device comprising at least: a communication interface, a memory, a processor and a communication bus; wherein:

[0036] The memory is used to store executable instructions;

[0037] The communication bus is used to realize the communication connection between the communication interface, the processor and the memory;

[0038] The processor is used to execute the information processing program stored in the memory to implement the steps of any of the above-mentioned information processing methods.

[0039] The present application provides a storage medium, on which an information processing program is stored. When the information processing program is executed, it is used to implement the steps of any of the above-mentioned information processing methods.

[0040] The present application provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the steps of any one of the above-mentioned information processing methods are implemented.

[0041] The embodiment of the present application provides an information processing method, device, equipment, storage medium and computer program product. After acquiring the learning attribute parameters and user performance parameters of the object to be analyzed, a user portrait of the object to be analyzed is constructed based on the learning attribute parameters and user performance parameters, and the course planning knowledge graph corresponding to the object to be analyzed is determined. Finally, based on the user portrait, learning goals and course planning knowledge graph, the learning path to be recommended is determined. In this way, after obtaining the user portrait according to the learning attribute parameters and user performance parameters of the object to be analyzed, the course planning knowledge graph of the object to be analyzed is used to determine the learning path to be recommended based on the user portrait and learning goals, and the course planning knowledge graph is used to determine the learning path, which solves the problem that the current adaptive learning system lacks the perception of the learning context, and proposes an adaptive learning recommendation method, which comprehensively considers various factors such as personalized learning, automatically adjusts the teaching content and difficulty according to the user's learning progress and ability, and effectively realizes the personalized learning needs of different users, and improves learning efficiency and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A flowchart of an information processing method provided in an embodiment of the present application;

[0043] Figure 2 A schematic diagram of the structure of an information processing system provided in an embodiment of the present application;

[0044] Figure 3 A schematic diagram of a flow chart of an application embodiment for implementing a portrait construction provided in an embodiment of the present application;

[0045] Figure 4 A schematic diagram of a learning path planning application implementation process provided in an embodiment of the present application;

[0046] Figure 5 A schematic diagram of a flow chart of an implementation of an application embodiment of a teaching assistant module provided in an embodiment of the present application;

[0047] Figure 6 A schematic diagram of the structure of a subject knowledge graph provided in an embodiment of the present application;

[0048] Figure 7 A schematic diagram of the structure of an information processing device provided in an embodiment of the present application;

[0049] Figure 8 A schematic diagram of the structure of an information processing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0051] The embodiment of the present application provides an information processing method, referring to Figure 1 As shown, the method is applied to an information processing device, and the method comprises the following steps:

[0052] Step 101: Obtain learning attribute parameters and user performance parameters of the object to be analyzed.

[0053] In the embodiment of the present application, the information processing device is an electronic device with computing capabilities, such as an intelligent terminal device, a computer device, a server device, etc. The object to be analyzed is a user object for which a learning path recommendation is required. The learning attribute parameter is used to represent the relevant characteristic parameters of the object to be analyzed about learning, and the user performance parameter is an attribute parameter used to represent the behavioral characteristics of the object to be analyzed.

[0054] Step 102: construct a user profile of the object to be analyzed based on the learning attribute parameters and the user performance parameters.

[0055] In the embodiment of the present application, a portrait analysis method is adopted to perform a comprehensive portrait analysis on the learning attribute parameters and user performance parameters of the object to be analyzed, so as to obtain a user portrait of the object to be analyzed.

[0056] Step 103: Determine the course planning knowledge graph corresponding to the object to be analyzed.

[0057] In the embodiment of the present application, the course planning knowledge map is the relationship between courses and knowledge points. The course planning knowledge map corresponding to the object to be analyzed can be determined by the course content that the object to be analyzed wants to learn.

[0058] Step 104: Determine the learning path to be recommended based on the user portrait, learning objectives, and course planning knowledge graph.

[0059] In an embodiment of the present application, according to the knowledge content in the course planning knowledge map, analysis is performed according to user portraits and learning goals to determine the learning path to be recommended. After determining the learning path to be recommended, the object to be analyzed can be guided to learn directly according to the learning path, or the learning path can be directly output to allow the object to be analyzed or the learning manager of the object to be analyzed, such as a parent or teacher, to view and adjust the learning arrangement reference of the object to be analyzed during the learning process, or the learning manager of the object to be analyzed can provide a reference for the learning arrangement of the object to be analyzed.

[0060] Based on the foregoing embodiments, in other embodiments of the present application, the learning attribute parameters include at least one or more of the following parameters: learning behavior parameters and performance parameters; the user performance parameters include at least one or more of the following parameters: interest parameters and social parameters.

[0061] In the embodiment of the present application, the learning behavior parameters may be, for example, learning time arrangements, learning preferences, learning efficiency, etc., the performance parameters may be learning performance, the interest parameters may be, for example, the interests and hobbies of the object to be analyzed in various aspects of life, and the social parameters may be, for example, the social preferences of the object to be analyzed.

[0062] Based on the foregoing embodiment, in other embodiments of the present application, the method further includes the following steps:

[0063] Obtain the learning progress and learning feedback information of the object to be analyzed;

[0064] The learning path is updated based on the learning progress and learning feedback information to obtain an updated learning path.

[0065] In an embodiment of the present application, after recommending a learning path for the object to be analyzed, the learning situation of the object to be analyzed is recorded in real time, including the learning progress and learning feedback information of the object to be analyzed, and then the learning path is dynamically updated in real time according to the learning progress and learning feedback information of the object to be analyzed, and the object to be analyzed is guided to learn according to the updated learning path. In this way, by adjusting the learning path in real time according to the learning status of the object to be analyzed, it is possible to better adapt to the learning status of the object to be analyzed, thereby ensuring the learning quality of the object to be analyzed learning according to the learning path.

[0066] Based on the foregoing embodiment, in other embodiments of the present application, the method further includes the following steps:

[0067] In the process of the object to be analyzed learning based on the learning path, determine the subject knowledge map;

[0068] If the information to be queried is detected, the retrieval information of the information to be queried is determined based on the subject knowledge graph;

[0069] Output the search information.

[0070] In an embodiment of the present application, in the process of learning by the object to be analyzed based on the learning path, the learning path may be a learning path initially formulated for the object to be analyzed. The learning path here may be determined according to the aforementioned path determination method, or may be obtained by other methods such as planning by the object to be analyzed itself, or planning by the corresponding learning manager for the object to be analyzed. It may also be a learning path obtained by dynamically adjusting and updating the aforementioned formulated learning path according to the learning progress and learning feedback information of the object to be analyzed. The specific learning path may be determined according to the actual application scenario and is not specifically limited here.

[0071] The information to be queried may be one or more types of information such as audio information, text information, and picture information. The information to be detected may be input to the information processing device through an audio acquisition device such as a microphone, a virtual keyboard or a physical keyboard, and the input device for the information to be detected may be an independent third-party device that has a communication connection with the information processing device, or may be a partial structural device that is integrated with the information processing device, which may be determined by actual conditions.

[0072] The subject knowledge graph is used to represent the graph information between the knowledge points of each subject.

[0073] In the process of learning the object to be analyzed based on the learning path, if the information to be queried is detected from the determined subject knowledge graph, the information to be queried is retrieved, the corresponding retrieval information is obtained, and the retrieval information is output and processed as a response to the information to be retrieved. In this way, the retrieval efficiency corresponding to the information to be queried is improved by searching through the subject knowledge graph.

[0074] Based on the foregoing embodiment, in other embodiments of the present application, the method further includes the following steps:

[0075] Obtain the knowledge points of subject corpus;

[0076] Through the trained knowledge extraction large language model, the specific knowledge content is extracted from the knowledge point content and subject corpus;

[0077] Generate subject knowledge graph based on knowledge point content and specific knowledge content.

[0078] In the embodiment of the present application, the implementation process of generating a subject knowledge graph can be implemented as an independent embodiment, which can be executed before the aforementioned step of using the subject knowledge graph, or after being implemented in other devices, the subject knowledge graph is sent to the information processing device, which can be determined by the actual situation. No specific limitation is made here.

[0079] The knowledge extraction big language model can be a model stored in the information processing device for extracting specific knowledge content from knowledge point content and subject corpus, or it can be obtained from a device storing the knowledge extraction big language model when the information processing device needs to use it. In the process of applying the knowledge extraction big language model to the information processing device, the knowledge extraction big language model can be continuously updated according to the actual application scenario to ensure that the specific content of the extracted knowledge is more in line with actual requirements.

[0080] Determine the corpus of each subject, and extract the knowledge point content of each subject corpus to obtain the knowledge point content of each subject corpus, and use the trained knowledge extraction large language model to perform knowledge content recognition processing on the knowledge point content and subject corpus to obtain the specific content of knowledge. Finally, analyze the knowledge point content and the specific content of knowledge, determine the specific relationship between each knowledge point, and generate a subject knowledge graph. Among them, when designing the subject knowledge graph, it can be designed to include three types of nodes and four types of relationships. Among them, the three types of nodes include: knowledge point nodes, sub-knowledge point nodes and multimodal content nodes, among which the multimodal content nodes include one or more of textbooks, network resources, exercises, PPTs and videos. The four types of relationships designed include at least: predecessor knowledge point relationship, successor knowledge point relationship, parent-child knowledge point relationship, and corresponding multimodal content relationship; among them, the predecessor knowledge point relationship is used to connect two knowledge points, the successor knowledge point relationship is used to correspond one-to-one with the predecessor knowledge point, the connected knowledge points remain unchanged but in the opposite direction, the parent-child knowledge point relationship is used to connect a knowledge point and a child knowledge point, and the corresponding multimodal content relationship is used to connect a child knowledge point and a multimodal content node.

[0081] Based on the foregoing embodiment, in other embodiments of the present application, if the information to be queried is detected, the search information of the information to be queried is determined based on the subject knowledge graph, which can be achieved by the following steps:

[0082] If the information to be queried is detected, the trained content retrieval language model is obtained;

[0083] Through the content retrieval large language model, the information corresponding to the information to be queried is retrieved from the subject knowledge graph to obtain the retrieval information.

[0084] In the embodiment of the present application, the information to be queried is input into the content retrieval big language model, so that the content retrieval big language model performs content query based on the subject knowledge graph, obtains information matching the information to be queried, and thus obtains retrieval information. The retrieval information is the reply content of the information to be queried.

[0085] Based on the foregoing embodiment, in other embodiments of the present application, the content retrieval large language model has a retrieval enhancement generation function.

[0086] Based on the above embodiment, in other embodiments of the present application, the step of outputting the search information can be implemented by the following steps:

[0087] Perform sentiment analysis on the retrieved information to obtain emotional information;

[0088] Through the image of a virtual teaching assistant, the search information is reported in voice form according to the emotional information.

[0089] In the embodiment of the present application, the method of performing sentiment analysis on the content in the search information can be implemented by artificial intelligence. The virtual teaching assistant image is a character image, and the virtual teaching assistant image is used to broadcast the search information to the object to be analyzed, so as to realize the interaction of the object to be analyzed in the learning process, thereby improving the learning efficiency.

[0090] Based on the foregoing embodiments, the embodiments of the present application provide an information processing method, the basic implementation process of which is: by collecting student behavior, grades and interest data to build a detailed student portrait, so as to gain insight into their learning preferences and needs. Then, use the portrait data to dynamically plan personalized learning paths to guide students to learn effectively. In combination with artificial intelligence (AI) digital human teaching assistants, large models and subject knowledge graphs are used to provide round-the-clock question-and-answer services, while recording questions to optimize teaching content. The system also supports teachers to participate in the creation and iteration of teaching content to ensure that teaching tools meet actual needs. Finally, through the management decision support module, data-based decision support is provided to teachers, so that targeted teaching decisions can be made and the quality and efficiency of education can be improved.

[0091] Correspondingly, a structure of an information processing system is provided, such as Figure 2 As shown, it includes a portrait building module, a path planning module, a teaching assistant module, a collaboration support module, and a management decision module. Among them:

[0092] Profile building module: used to collect and analyze students’ learning behaviors, grades, interests and other data to build detailed student profiles. In this way, the system can provide customized learning resources and activities based on each student’s learning situation and preferences.

[0093] Path planning module: used to dynamically plan personalized learning paths based on student profiles and learning goals.

[0094] Teaching assistant module: Using knowledge graphs, big models and digital human technology, it provides 24 / 7 Q&A services to understand and solve students' problems, provide instant and accurate answers, and record common problems to optimize teaching content.

[0095] Collaboration support module: allows teachers and others to participate in the creation, feedback and iteration process of teaching content. Teachers and others can participate in the customization and feedback of teaching content to ensure that teaching tools are consistent with actual teaching needs.

[0096] Management decision-making module: provides data-supported decision-making basis for teachers, educational administrators, etc. The learning data analysis provided by the system helps teachers, educational administrators, etc. make more targeted teaching decisions.

[0097] Based on the above description, one implementation of the portrait construction module can be as follows Figure 3 As shown, including:

[0098] Step a11: data collection.

[0099] The collected data generally includes students' learning behavior data, performance data, interest data and social data. Specifically, students' online learning behavior data, such as login time, learning time, course completion, etc., can be collected through learning management systems or educational applications. Students' test scores, homework scores and other data can be imported from the school's performance management system. Students' interest data can be collected through questionnaires, interest group activity records, etc. By analyzing students' activities on social platforms, their social preferences and behavior pattern data can be obtained.

[0100] Step a12: data storage.

[0101] Among them, when storing data, a relational database or a non-relational database can be used to store the basic information, learning behavior data, performance data, interest data, social data, etc. of the students collected above. In this way, the student data is centrally stored for easy subsequent search and call.

[0102] Step a13: data analysis.

[0103] The data analysis process can specifically include: data preprocessing, that is, cleaning and formatting the received student data, and then deleting or supplementing missing values ​​and outliers; feature processing, that is, extracting useful features from the original data, such as learning habits, learning efficiency and other characteristic parameters; finally, using machine learning algorithms such as clustering, classification, and recommendation algorithms to analyze students' behaviors and preferences and obtain analysis results in order to subsequently build student portraits.

[0104] Step a14: Generate student portraits.

[0105] According to the analysis results in step a13, a label is generated for each student, such as "math lover", "reading lover", etc., and a detailed description of each student is generated, including study habits, interests and hobbies, learning effects, etc.

[0106] In this way, a comprehensive student portrait can be constructed to help teachers better understand students and provide more targeted educational services.

[0107] Correspondingly, the aforementioned learning path planning module aims to design the most suitable learning path according to the student's personal situation and learning goals. One corresponding implementation method can be as follows: Figure 4 As shown, including:

[0108] Step b11: Data collection and integration.

[0109] Among them, student portrait data is collected, including learning habits, interest preferences, academic performance history, etc. At the same time, information such as course resources, teaching outlines, and learning materials are integrated.

[0110] Step b12: Set learning goals.

[0111] Among them, students, teachers or educational administrators are allowed to set learning goals for objects, wherein the learning goals can be short-term goals or long-term goals.

[0112] Step b13: Generate a learning path.

[0113] Among them, a course planning knowledge graph of the relationship between courses and knowledge points is constructed, and according to student portraits and learning objectives, the shortest path algorithm such as the Dijkstra algorithm or the A* search algorithm, or the simulated annealing algorithm is used to search for the optimal learning path from the course planning knowledge graph.

[0114] Step b14: Adjust the learning path.

[0115] Among them, students' learning progress and learning feedback are obtained, the generated learning path is dynamically adjusted, and students' learning status, such as understanding level, learning speed, etc., are monitored in real time.

[0116] In this way, the learning path planning module can provide each student with a customized learning plan, helping them achieve their learning goals more effectively.

[0117] Correspondingly, the aforementioned teaching assistant module aims to provide personalized and immediate learning support by using technologies such as knowledge graphs, large models, and digital humans. Figure 5 As shown, it can be as follows:

[0118] Step c11: Construct a subject knowledge graph.

[0119] Among them, a comprehensive knowledge graph including subject knowledge points, concepts and the relationships between them is constructed to ensure that students can get support in designated subject areas. In this way, knowledge graph technology that can achieve deep understanding and precise answers in specific fields can be adopted. That is, by constructing a knowledge point network covering a wide range of disciplines, a solid knowledge foundation and logical structure are provided for the large model. In this way, the constructed subject knowledge graph can not only help the large model understand and reason about complex problems more accurately, but also keep the knowledge up-to-date and accurate through continuous learning and updating.

[0120] Exemplarily, a subject knowledge graph can be constructed through the Neo4j platform, which includes three types of nodes: knowledge point nodes, child knowledge point nodes, and multimodal content nodes. The subject knowledge graph also includes four types of relationships: predecessor knowledge point relationship, successor knowledge point relationship, parent-child knowledge point relationship, and corresponding multimodal content relationship. In this way, the use of nodes and relationships to form a graph database can meet the preliminary preparation of knowledge reasoning functions.

[0121] The corpus for constructing the subject knowledge graph can include multimodal corpus including textbooks, online resources, exercises, PPT, and videos. When constructing the subject knowledge graph, firstly, the knowledge points in the corpus can be extracted manually or by a large model, and then the sub-knowledge points in the corpus can be extracted manually or by a large model. For example, the two-dimensional array knowledge point has multiple sub-knowledge points such as two-dimensional array initialization and two-dimensional array definition. Secondly, the extracted knowledge points and sub-knowledge points, and even the corpus content, are input into the large model, and the large model is used to pair the knowledge points and the specific knowledge content, and then the pairing results of the large model are paired and corrected manually. It should be noted that a multimodal content may correspond to multiple sub-knowledge points. In this way, the corresponding code is generated according to the extracted knowledge points and the corresponding content, so as to realize the rapid construction of an accurate subject knowledge graph.

[0122] For example, Figure 6 Shown is a structural schematic diagram of a subject knowledge graph provided in an embodiment of the present application.

[0123] Step c12: Knowledge quiz application.

[0124] Among them, a large language model is used to understand students' questions and generate natural and accurate answers. The model can understand the context of the questions and ensure that the answers provided are relevant to the students' current learning stage and knowledge points. Among them, the large language model is implemented through large model technology, so that an all-weather online knowledge question and answer application can be built to achieve functions similar to a virtual teacher. It can not only simulate the teaching behavior of a real teacher, but also provide personalized teaching content and interactive Q&A according to students' learning habits and needs. Furthermore, it can also be combined with retrieval-augmented generation (RAG) technology to further reduce the hallucination phenomenon of large language models when generating answers, improve the relevance and accuracy of answers, and provide students with more reliable and professional learning support. In this way, through the training of deep learning algorithms, the large model can continuously optimize its teaching strategy to adapt to the learning rhythm and style of different students, ensuring that each student can receive targeted guidance and help.

[0125] Exemplarily, langchain technology can be used to implement a large model question-and-answer application based on a local knowledge base, such as a vector database combined with a subject knowledge graph, which can be implemented using the RAG application architecture, thereby establishing a knowledge question-and-answer solution that is friendly to Chinese scenarios and open source models and can be run offline. In the process of technical implementation, it can rely on the open source large language model (LLM) and the encoding (Embedding) model, all of which can be deployed offline privately using the open source model, or it can be implemented by calling the Generative Pre-Trained Transformer (GPT) application programming interface (API) interface service. In this way, the specific implementation logic is: 1) It is necessary to load relevant text files from the storage medium, such as teaching materials, network resources, exercises, etc. 2) Text processing: read and parse the content of these text files, and segment the text to decompose it into smaller units, such as sentences or phrases. 3) Text vectorization: Each segmented text unit is quantized, for example, it can be converted into a vector form. This can be achieved in a variety of ways, such as using the BERT or WoBERT model to extract sentence vectors at the word and phrase level, or using a word vector based on Word2Vec. 4) Vector storage: Select a suitable vector database to store the vector representation of the text block, such as Milvus, Faiss, etc., and you can also build an index in the vector database to improve the efficiency of similarity search. 5) Understanding user queries: Analyze the intent of the user query, determine the type of information to be retrieved, and extract keywords from the user query to guide the retrieval process. 6) Retrieval enhancement: Perform a similarity search on the local knowledge base to find the text block most relevant to the query. At the same time, perform subgraph matching or path query in the subject knowledge graph constructed previously to find entities and relationships related to the query. 7) Information fusion: Integrate the text blocks retrieved from the local knowledge base and the entity relationships retrieved from the subject knowledge graph to build a comprehensive context containing text and graph information, and construct it into a prompt together with the user's query question. 8) Generate answers using a large model: Select a suitable pre-trained language model, such as GPT-3, BERT, etc., submit the prompt containing contextual information to the large model, and use the large model to generate answers or related content.

[0126] Step c13: Create a digital virtual person.

[0127] Among them, after the knowledge question and answer application in step c12 retrieves the corresponding answer content, it can use digital virtual human technology to create a virtual teaching assistant image, interact with students through technologies such as speech recognition and speech synthesis, and use the digital virtual human interactive interface to enable students to communicate with the AI ​​digital human teaching assistant through text, voice or video. Furthermore, the created digital virtual human can not only have a realistic appearance and movements, but also communicate naturally with students through voice, expression and emotional feedback. In this way, through a highly anthropomorphic interactive method, the immersion of learning can be greatly enhanced, allowing students to learn in a more intuitive and vivid environment.

[0128] For example, Unreal Engine is first used for high-quality three-dimensional (3D) modeling and rendering, which can create a realistic digital human image, and realize the voice interaction function of the digital human through voice recognition and voice synthesis technology. When a user asks a question using a voice input device, the voice recognition module is first triggered, which is responsible for converting the user's voice information into text information and passing it to the large language model selected in the previous step.

[0129] After the large language model generates an answer, the text information of the answer will be conveyed to two modules. The first module is the speech generation module, which can intelligently convert text into natural speech streams. The second module is the sentiment analysis module, which can determine the specific emotions and other tags contained based on the vocabulary analysis of the text. Finally, the emotional information is conveyed to the digital simulation human rendered by Unreal Engine, such as UE5, so that it can make logical actions and facial expressions. In this way, the realized digital virtual human can respond to the user's questions in a more realistic way, improve the user experience in terms of simulation, realize highly simulated voice interaction and emotional expression, and provide users with a more intelligent, natural and humanized interactive experience.

[0130] In summary, by comprehensively considering various factors such as personalized learning, efficient question answering, and interactive guidance, and automatically adjusting teaching content and difficulty according to learners' learning progress and abilities, we can more effectively meet the personalized learning needs of different learners and improve learning efficiency and quality. In this way, we can provide students with a personalized, efficient, and interactive adaptive learning environment, while providing teachers and education administrators with powerful data support and a collaborative platform.

[0131] It should be noted that, for the description of the same steps and the same contents in this embodiment as those in other embodiments, reference can be made to the description in other embodiments and will not be repeated here.

[0132] The information processing method provided in the embodiment of the present application obtains the learning attribute parameters and user performance parameters of the object to be analyzed, then constructs a user portrait of the object to be analyzed based on the learning attribute parameters and user performance parameters, and determines the course planning knowledge graph corresponding to the object to be analyzed, and finally determines the learning path to be recommended based on the user portrait, learning goals and course planning knowledge graph. In this way, after obtaining the user portrait according to the learning attribute parameters and user performance parameters of the object to be analyzed, the course planning knowledge graph of the object to be analyzed is used to determine the learning path to be recommended based on the user portrait and learning goals, and the course planning knowledge graph is used to determine the learning path, which solves the problem that the current adaptive learning system lacks the perception of the learning context, and proposes an adaptive learning recommendation method, which comprehensively considers various factors such as personalized learning, and automatically adjusts the teaching content and difficulty according to the user's learning progress and ability, effectively realizing the personalized learning needs of different users and improving learning efficiency and quality.

[0133] Based on the above embodiments, the embodiments of the present application provide an information processing device, which can be applied to Figure 1 In the information processing method provided in the corresponding embodiment, refer to Figure 7 As shown, the information processing device 2 may include: an acquisition unit 21, a construction unit 22, a first determination unit 23 and a second determination unit 24; wherein:

[0134] An acquisition unit 21, used to acquire learning attribute parameters and user performance parameters of the object to be analyzed;

[0135] A construction unit 22, used to construct a user profile of the object to be analyzed based on the learning attribute parameters and the user performance parameters;

[0136] A first determining unit 23 is used to determine a course planning knowledge graph corresponding to the object to be analyzed;

[0137] The second determination unit 24 is used to determine the learning path to be recommended based on the user portrait, learning objectives and course planning knowledge graph.

[0138] In other embodiments of the present application, the learning attribute parameters include at least one or more of the following parameters: learning behavior parameters and performance parameters; the user performance parameters include at least one or more of the following parameters: interest parameters and social parameters.

[0139] In other embodiments of the present application, the device further includes: an updating unit; wherein:

[0140] The acquisition unit is also used to acquire the learning progress and learning feedback information of the object to be analyzed;

[0141] The updating unit is used to update the learning path based on the learning progress and learning feedback information to obtain an updated learning path.

[0142] In other embodiments of the present application, the device further includes: an output unit; wherein:

[0143] The first determination unit is further used to determine the subject knowledge graph in the process of the object to be analyzed learning based on the learning path;

[0144] The second determination unit is further used to determine the retrieval information of the information to be queried based on the subject knowledge graph if the information to be queried is detected;

[0145] The output unit is used to output the search information.

[0146] In other embodiments of the present application, the device further includes: an extraction unit and a generation unit; wherein:

[0147] The acquisition unit is also used to obtain the knowledge points of the subject corpus;

[0148] The extraction unit is used to extract specific knowledge content from knowledge point content and subject corpus through the trained knowledge extraction large language model;

[0149] The generation unit is used to generate a subject knowledge graph based on the knowledge point content and the specific content of knowledge.

[0150] In other embodiments of the present application, if the second determination unit executes the step of detecting the information to be queried, determining the search information of the information to be queried based on the subject knowledge graph can be implemented by the following steps:

[0151] If the information to be queried is detected, the trained content retrieval language model is obtained;

[0152] Through the content retrieval large language model, the information corresponding to the information to be queried is retrieved from the subject knowledge graph to obtain the retrieval information.

[0153] In other embodiments of the present application, the content retrieval large language model has a retrieval enhancement generation function.

[0154] In other embodiments of the present application, the output unit is specifically used to implement the following steps:

[0155] Perform sentiment analysis on the retrieved information to obtain emotional information;

[0156] Through the image of a virtual teaching assistant, the search information is reported in voice form according to the emotional information.

[0157] It should be noted that the process of information interaction between the units and modules in this embodiment can refer to the description in other embodiments and will not be repeated here.

[0158] The information processing device provided by the embodiment of the present application obtains the learning attribute parameters and user performance parameters of the object to be analyzed, and then constructs a user portrait of the object to be analyzed based on the learning attribute parameters and user performance parameters, and determines the course planning knowledge graph corresponding to the object to be analyzed, and finally determines the learning path to be recommended based on the user portrait, learning goals and course planning knowledge graph. In this way, after obtaining the user portrait according to the learning attribute parameters and user performance parameters of the object to be analyzed, the course planning knowledge graph of the object to be analyzed is used to determine the learning path to be recommended based on the user portrait and learning goals, and the course planning knowledge graph is used to determine the learning path, which solves the problem that the current adaptive learning system lacks the perception of the learning context, and proposes an adaptive learning recommendation method, which comprehensively considers various factors such as personalized learning, and automatically adjusts the teaching content and difficulty according to the user's learning progress and ability, effectively realizing the personalized learning needs of different users and improving learning efficiency and quality.

[0159] Based on the above embodiments, the embodiments of the present application provide an information processing device, which can be applied to Figure 1 In the information processing method provided in the corresponding embodiment, refer to Figure 8 As shown, the information processing device 3 may include: a communication interface 31, a memory 32, a processor 33 and a communication bus 34; wherein:

[0160] A memory 32, for storing executable information;

[0161] A communication bus 34, used to realize communication connection between the communication interface 31, the processor 33 and the memory 32;

[0162] The processor 33 is used to execute the information processing program stored in the memory 32 to implement the following Figure 1 The implementation process of the information processing method provided in the corresponding embodiment will not be repeated here.

[0163] Based on the foregoing embodiments, the embodiments of the present application provide a computer-readable storage medium, referred to as a storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement reference Figure 1 The implementation process of the information processing method provided in the corresponding embodiment will not be repeated here.

[0164] Based on the foregoing embodiments, the embodiments of the present application further provide a computer program product, including a computer program, which can be executed by the processor 33 of the information processing device 3 to complete any of the foregoing method steps.

[0165] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of hardware embodiments, software embodiments, or embodiments in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) that contain computer-usable program code.

[0166] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0167] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0168] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0169] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application.

Claims

1. An information processing method, characterized in that: The method comprises: Obtaining learning attribute parameters and user performance parameters of the object to be analyzed; Constructing a user profile of the object to be analyzed based on the learning attribute parameters and the user performance parameters; Determine the course planning knowledge graph corresponding to the object to be analyzed; Based on the user portrait, learning objectives and the course planning knowledge graph, a learning path to be recommended is determined.

2. The method according to claim 1, characterized in that The learning attribute parameters include at least one or more of the following parameters: learning behavior parameters and performance parameters; the user performance parameters include at least one or more of the following parameters: interest parameters and social parameters.

3. The method according to claim 1, characterized in that The method further comprises: Obtaining learning progress and learning feedback information of the object to be analyzed; The learning path is updated based on the learning progress and the learning feedback information to obtain an updated learning path.

4. The method according to claim 1, characterized in that: The method further comprises: In the process of the object to be analyzed learning based on the learning path, determining a subject knowledge graph; If the information to be queried is detected, the retrieval information of the information to be queried is determined based on the subject knowledge graph; The search information is output.

5. The method according to claim 4, characterized in that The method further comprises: Obtain the knowledge points of subject corpus; Extracting specific knowledge content from the knowledge point content and subject corpus through the trained knowledge extraction large language model; The subject knowledge graph is generated based on the knowledge point content and the specific content of the knowledge.

6. The method according to claim 4, characterized in that If the information to be queried is detected, the retrieval information of the information to be queried is determined based on the subject knowledge graph, including: If the information to be queried is detected, a trained content retrieval language model is obtained; The content retrieval large language model is used to retrieve information corresponding to the information to be queried from the subject knowledge graph to obtain the retrieval information.

7. The method according to claim 6, characterized in that The content retrieval large language model has a retrieval enhancement generation function.

8. The method according to claim 4, characterized in that The outputting of the search information comprises: Performing sentiment analysis on the retrieved information to obtain sentiment information; The search information is reported in voice form according to the emotional information through a virtual teaching assistant image.

9. An information processing device, characterized in that: The device at least comprises: an acquisition unit, a construction unit, a first determination unit and a second determination unit; wherein: The acquisition unit is used to acquire the learning attribute parameters and user performance parameters of the object to be analyzed; The construction unit is used to construct a user profile of the object to be analyzed based on the learning attribute parameters and the user performance parameters; The first determining unit is used to determine the course planning knowledge graph corresponding to the object to be analyzed; The second determination unit is used to determine the learning path to be recommended based on the user portrait, the learning goal and the course planning knowledge graph.

10. An information processing device, characterized in that: The device at least includes: a communication interface, a memory, a processor and a communication bus; wherein: The memory is used to store executable instructions; The communication bus is used to realize the communication connection between the communication interface, the processor and the memory; The processor is used to execute the information processing program stored in the memory to implement the steps of the information processing method according to any one of claims 1 to 8.

11. A storage medium, characterized in that: The storage medium stores an information processing program, which is used to implement the steps of the information processing method according to any one of claims 1 to 8 when executed.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the steps of the information processing method according to any one of claims 1 to 8.

Images