A story interactive active questioning method and system based on cognitive characteristics of children

Abstract

The application discloses a story interactive active questioning method and system based on cognitive characteristics of children, and relates to the technical field of artificial intelligence. The method comprises the following steps: obtaining story content; performing multi-modal analysis on the story content to obtain an analysis result; constructing a four-dimensional cognitive model of a child according to the analysis result and historical interaction data of the child; selecting a target problem level in preset multi-level problems according to the four-dimensional cognitive model; generating an adaptive question corresponding to the target problem level; and updating the four-dimensional cognitive model and the multi-level problems according to the answer and emotional feedback of the child to the adaptive question. The application first analyzes the story content and then generates a corresponding adaptive question in combination with the cognition of the child, so that the generated story question matches the cognition of the child, the problems beyond the cognition of the child are reduced, the adaptability is improved, and the learning and understanding effect of the child is improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a story-based interactive active questioning method and system based on children's cognitive characteristics. Background Technology

[0002] With the widespread application of artificial intelligence technology in education, storytelling robots have become an important tool for early childhood education. However, existing storytelling robots on the market have many shortcomings in their interactive question-and-answer sessions, making it difficult to meet children's personalized and diverse learning needs. Most products only provide a fixed question bank or simple question-and-answer functions, lacking a dynamic adjustment mechanism for individual differences among children. This results in a mismatch between the difficulty of the questions and children's actual cognitive abilities, affecting learning outcomes. Summary of the Invention

[0003] The main objective of this application is to propose a story-based interactive active questioning method and system based on children's cognitive characteristics, so as to improve the matching effect between questions and children's cognition.

[0004] To achieve the above objectives, one aspect of this application proposes a story-based interactive active questioning method based on children's cognitive characteristics, the method comprising the following steps: Get the story content; Multimodal analysis was performed on the story content to obtain the analysis results; Based on the analysis results and the child's historical interaction data, a four-dimensional cognitive model of the child is constructed; Based on the four-dimensional cognitive model, the target question level is selected from the preset multi-level questions; Generate the adaptation problem corresponding to the target problem level; The four-dimensional cognitive model and the multi-level questions are updated based on the child's answers to the adaptation questions and emotional feedback.

[0005] In some embodiments, constructing the child's four-dimensional cognitive model based on the analysis results and the child's historical interaction data includes the following steps: The basic profile of the child, the child's semantic understanding threshold of the story content, the expressive ability score, and the emotional engagement are obtained as the historical interaction data; The four-dimensional cognitive model is constructed based on the analysis results, the child's basic profile, and the historical interaction data.

[0006] In some embodiments, obtaining the child's basic profile, the child's semantic understanding threshold of the story content, expressive ability score, and emotional engagement as the historical interaction data includes the following steps: A basic profile of the child is constructed based on the child's age, gender, and geographic location information; The BERT model is used to calculate the semantic similarity between the child's historical answer text and the standard answer, thereby generating the child's semantic understanding threshold of the story content; An LSTM network is used to extract speech features from the child's historical answers to questions, and then a score of the child's expressive ability to the story content is generated. The emotional engagement of the child with the story content is generated using a sentiment analysis model based on the historical response text and the historical response speech.

[0007] In some embodiments, selecting the target question level from a preset multi-level question based on the four-dimensional cognitive model includes the following steps: The historical interaction data is processed using an emotion analysis model to obtain the child's emotion memory mapping table; The target question hierarchy is generated using a dynamic routing algorithm based on the semantic understanding threshold, the expressiveness score, the emotional engagement, and the emotion memory mapping table.

[0008] In some embodiments, the method further includes the following steps: Match the question corresponding to the story content in the question bank; Based on the matched questions, three levels of questions are generated in sequence: memory-related questions, reasoning-related questions, and creative questions, which serve as the preset multi-level questions.

[0009] In some embodiments, generating the adaptation problem corresponding to the target problem level includes the following steps: Cultural elements are generated based on the regional information in the child's basic portrait. The corresponding adaptation question is generated based on the cultural elements and the target question hierarchy.

[0010] In some embodiments, the method further includes the following steps: The keyword matching algorithm is used to dynamically detect the relevance of the children's answers to the theme of the story. If the relevance exceeds a preset threshold, the child is guided back to the theme of the story.

[0011] To achieve the above objectives, another aspect of this application proposes a story-based interactive active questioning system based on children's cognitive characteristics, the system comprising: The story acquisition unit is used to acquire story content; The story analysis unit is used to perform multimodal analysis on the story content and obtain analysis results; A cognitive building unit is used to construct a four-dimensional cognitive model of the child based on the analysis results and the child's historical interaction data. The level selection unit is used to select the target question level from a preset multi-level question based on the four-dimensional cognitive model. The problem generation unit is used to generate adaptation problems corresponding to the target problem level; An adaptive feedback unit is used to update the four-dimensional cognitive model and the multi-level questions based on the child's answers to the adaptation questions and emotional feedback.

[0012] To achieve the above objectives, another aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-described method.

[0013] To achieve the above objectives, another aspect of the embodiments of this application proposes a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0014] To achieve the above objectives, another aspect of this application provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0015] The embodiments of this application include at least the following beneficial effects: This application provides a story-based interactive active questioning method and system based on children's cognitive characteristics. The solution involves: acquiring story content; performing multimodal analysis on the story content to obtain analysis results; constructing a four-dimensional cognitive model of the child based on the analysis results and the child's historical interaction data; selecting a target question level from a pre-set multi-level question set based on the four-dimensional cognitive model; generating appropriate questions corresponding to the target question level; and updating the four-dimensional cognitive model and multi-level questions based on the child's answers to the appropriate questions and emotional feedback. This application first analyzes the story content and then combines it with the child's cognition to generate corresponding appropriate questions. This ensures that the generated story questions match the child's cognition, reduces the generation of questions beyond the child's understanding, improves adaptability, and thus enhances the child's learning comprehension. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1A flowchart illustrating a story-based interactive active questioning method based on children's cognitive characteristics, provided as an embodiment of this application; Figure 2 A flowchart of the system analysis steps provided in the embodiments of this application; Figure 3 A schematic diagram of the structure of a story-interactive active questioning system based on children's cognitive characteristics provided in this application embodiment; Figure 4 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit it. In the following description, when referring to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those of this application; they are merely examples of systems and methods consistent with some aspects of the embodiments of this application as detailed in the appended claims.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0020] Before providing a detailed description of the embodiments of this application, some related technologies involved in the embodiments of this application will be described first, as follows: Natural Language Processing (NLP) is an interdisciplinary field of computer science and artificial intelligence that aims to enable computers to understand, interpret, and generate human language, achieving natural human-computer interaction. Core technologies include word segmentation, part-of-speech tagging, named entity recognition, semantic understanding, and text generation, and it is widely used in scenarios such as intelligent customer service, machine translation, and information extraction.

[0021] BERT (Bidirectional Encoder Representations from Transfrmers) is a pre-trained language model developed by Google, based on the Transfrmer architecture of bidirectional encoders. It learns contextual semantic representations through masked language modeling (MLM) and next-sentence prediction (NSP) tasks, significantly improving the performance of NLP tasks such as question answering systems and text classification, and pioneering the pre-training-fine-tuning paradigm.

[0022] LSTM (Ling Shrt-Term Memry) network: A special type of recurrent neural network (RNN) that effectively solves the vanishing gradient problem of traditional RNNs through gating mechanisms (input gate, forget gate, output gate), enabling it to learn long-term dependencies in long sequences of data. It is commonly used for time-series data processing tasks such as speech recognition, machine translation, and time series prediction.

[0023] Bloom's Taxonomy: A cognitive objective classification system proposed by educational psychologist Benjamin Bloom, which divides learning objectives into six levels: remembering, understanding, applying, analyzing, evaluating, and creating. This theory provides a framework for instructional design and curriculum assessment, and is commonly used in the field of education for goal setting and competency assessment.

[0024] The embodiments of this application originate from in-depth research and analysis of the current children's story interactive device market during the development of AI desktop robots. With the widespread application of artificial intelligence technology in the field of education, story robots have become an important tool for early childhood education. However, existing story robots on the market have many shortcomings in the interactive questioning process, making it difficult to meet children's personalized and diverse learning needs. Most products only provide a fixed question bank or simple question-and-answer functions, lacking a dynamic adjustment mechanism for individual differences among children, resulting in a mismatch between the difficulty of the questions and children's actual cognitive abilities, thus affecting learning outcomes. At the same time, existing products fail to fully consider regional cultural differences and gender cognitive development characteristics, limiting children's personalized development.

[0025] This application, in line with the educational goals of the "Guidelines for Learning and Development of Children Aged 3-6," aims to develop a method and system for intelligent, dynamic, and personalized interactive story questioning based on children's cognitive characteristics. By constructing a multi-dimensional cognitive model and designing a dynamic question engine, the system can adjust the content and difficulty of questions in real time according to children's comprehension, expression, and emotional engagement, thereby enhancing children's participation and cognitive development during story learning and filling a gap in existing technologies in this field.

[0026] Currently, children's story interactive devices on the market have the following core problems: 1. Rigid Questioning Patterns: Existing story robots use fixed question templates, with monotonous question content and methods. They cannot dynamically adjust according to children's cognitive abilities and individual differences, resulting in a mismatch between the difficulty of the questions and children's actual abilities, which affects children's learning enthusiasm and effectiveness.

[0027] 2. Insufficient cross-cultural and gender adaptation: Lack of consideration for regional cultural differences and the characteristics of gender cognitive development. Children from different regions have different understandings and interests in story content due to different cultural backgrounds; children of different genders have their own advantages in cognitive development, such as girls having an advantage in language ability and boys excelling in spatial thinking. However, existing robots cannot provide adaptive interactive content to address these differences, thus limiting children's personalized development.

[0028] 3. Lack of emotional feedback: Most existing systems lack real-time monitoring and feedback mechanisms for children's emotional states, making it impossible to adjust questioning strategies based on children's emotional reactions, and thus failing to stimulate children's learning interest and participation.

[0029] 4. Limited cognitive models: Existing technologies typically construct child profiles based solely on age or simple learning records, failing to integrate multi-dimensional assessment indicators such as comprehension, expressiveness, and emotional engagement, resulting in inaccurate cognitive characteristic analysis.

[0030] 5. Insufficient cultural sensitivity: The existing system is prone to deviating from the story theme or cultural background during questioning, and lacks an effective mechanism for detecting and correcting theme deviations, which affects the achievement of teaching objectives.

[0031] This application discloses a story-based interactive active questioning method and system based on children's cognitive characteristics. The solution accurately assesses children's cognitive state by constructing a three-dimensional cognitive model (covering comprehension, expression, and emotional engagement) that includes age, gender, and geographic factors. It employs a three-level progressive question engine, designing memory, reasoning, and creative questions based on Bloom's Taxonomy, and dynamically selecting the question level according to children's historical cognitive characteristics. The system can be widely applied in early childhood education at home, kindergarten education, and other scenarios, effectively solving problems such as rigid questioning patterns, insufficient cross-cultural and gender adaptation, and mismatch between question difficulty and children's cognitive abilities in existing story robots, thereby improving the effectiveness of children's story learning and cognitive development.

[0032] This application provides a story-based interactive active questioning method and system based on children's cognitive characteristics, relating to the field of artificial intelligence technology. The story-based interactive active questioning method and system provided in this application can be applied to a terminal, a server, or software running on a terminal or server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, or in-vehicle terminal, but is not limited to these. The server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network. The software can be an application implementing the story-based interactive active questioning method, but is not limited to the above forms.

[0033] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0034] Reference Figure 1 This application provides a story-based interactive active questioning method based on children's cognitive characteristics. This method may include, but is not limited to, steps S100 to S150, as follows: S100: Obtain story content.

[0035] For example, multiple story input methods are supported: parents can manually select or the story can be automatically pushed based on the child's preferences.

[0036] S110: Perform multimodal analysis on the story content to obtain the analysis results.

[0037] For example: Text analysis: Using NLP techniques to analyze the semantics, difficulty, and emotional tone of a story.

[0038] Cultural analysis: Extracting regional cultural elements from the story.

[0039] Gender analysis: Identifying gender-related elements in a story.

[0040] S120: Construct a four-dimensional cognitive model of the child based on the analysis results and the child's historical interaction data.

[0041] For example, the four-dimensional cognitive model includes a basic profile of the child, the child's semantic comprehension threshold for the story content, expressive ability score, and emotional engagement.

[0042] S130: Select the target problem level from the preset multi-level problems according to the four-dimensional cognitive model.

[0043] For example, design three levels of questions—memory, reasoning, and creation—based on the story content and Bloom's Taxonomy, and select an appropriate level from the three levels of questions.

[0044] S140: Generate the adaptation problem corresponding to the target problem level.

[0045] For example, relevant elements are extracted from a cultural knowledge base based on children's regional information and injected into the question generation process to generate questions at the target question level.

[0046] S150: Update the four-dimensional cognitive model and the multi-level questions based on the child's answers to the adaptation questions and emotional feedback.

[0047] For example, the following steps are included: Cognitive model update: The four-dimensional cognitive model is updated based on the latest answers.

[0048] Question database optimization: Record the correct answer rate and adjust the question weights.

[0049] Culture adaptation optimization: Analyze the effects of culture injection issues and optimize culture adaptation strategies.

[0050] Optionally, constructing the child's four-dimensional cognitive model based on the analysis results and the child's historical interaction data includes the following steps: The basic profile of the child, the child's semantic understanding threshold of the story content, the expressive ability score, and the emotional engagement are obtained as the historical interaction data; The four-dimensional cognitive model is constructed based on the analysis results, the child's basic profile, and the historical interaction data.

[0051] Specifically, this embodiment is used to construct children's cognitive characteristics based on children's information and children's understanding of the story.

[0052] Optionally, obtaining the child's basic profile, the child's semantic understanding threshold of the story content, expressive ability score, and emotional engagement as the historical interaction data includes the following steps: A basic profile of the child is constructed based on the child's age, gender, and geographic location information; The BERT model is used to calculate the semantic similarity between the child's historical answer text and the standard answer, thereby generating the child's semantic understanding threshold of the story content; An LSTM network is used to extract speech features from the child's historical answers to questions, and then a score of the child's expressive ability to the story content is generated. The emotional engagement of the child with the story content is generated using a sentiment analysis model based on the historical response text and the historical response speech.

[0053] Specifically, the system receives information about the child's age, gender, and location from parents, processes it using a rule engine, and generates a basic profile of the child.

[0054] The semantic similarity between the child's answer and the standard answer is calculated using the BERT model (bert-base-chinese) as the semantic understanding threshold. The input text is: story content + child's answer (maximum sequence length 512 tokens). The output is the semantic understanding threshold α (α∈[0,1]), and the closer α is to 1, the stronger the understanding.

[0055] The LSTM network is used to process speech data, extract speech features, and then generate a child's expressive ability score for the story content. Among them, speech features include: number of pauses, speech rate, vocabulary diversity, etc. The expressive ability score β is calculated by the formula: β = (vocabulary diversity × 0.6) + (1 / number of pauses × 0.4); β∈[0,100], reflecting the child's language expression ability.

[0056] This study analyzes the tone and speed of children's responses using a speech emotion recognition model to generate a score of their expressiveness in the story content. A multimodal emotion analysis algorithm is used, combining textual and speech emotions. The emotion engagement level U is calculated using the formula: U = ∑W(Ui) (i=1,2,3,⋯6), where Ui represents six emotion dimensions: happiness, boredom, anger, sadness, anxiety, and surprise, with weights W=(0.3934,0.1520,0.0997,0.1194,0.1296,0.1059). Emotional state is represented by the average emotional state score m1 and the number of positive and negative emotions m2.

[0057] Optionally, selecting the target question level from a preset multi-level question based on the four-dimensional cognitive model includes the following steps: The historical interaction data is processed using an emotion analysis model to obtain the child's emotion memory mapping table; The target question hierarchy is generated using a dynamic routing algorithm based on the semantic understanding threshold, the expressiveness score, the emotional engagement, and the emotion memory mapping table.

[0058] Specifically, the system uses neural network methods to perform sentiment analysis on historical interaction records; records children's emotional responses (positive / negative / neutral) to memory, reasoning, and creative problems; and establishes an emotional memory mapping table to record children's emotional tendencies towards specific topics / problems.

[0059] Optionally, the method further includes the following steps: Match the question corresponding to the story content in the question bank; Based on the matched questions, three levels of questions are generated in sequence: memory-related questions, reasoning-related questions, and creative questions, which serve as the preset multi-level questions.

[0060] For example, design three levels of different types of questions based on Bloom's Taxonomy: Level 1 (Memory-based questions): Tests children's memory of basic facts in a story, such as "Where does Grandma live?".

[0061] Level 2 (Reasoning Questions): These involve the level of understanding and application, such as "Why did the Big Bad Wolf pretend to be Grandma?"

[0062] Level 3 (Creative Questions): These belong to the analysis and evaluation level, such as "If you were Little Red Riding Hood, what would you do?"

[0063] Optionally, generating the adaptation problem corresponding to the target problem level includes the following steps: Cultural elements are generated based on the regional information in the child's basic portrait. The corresponding adaptation question is generated based on the cultural elements and the target question hierarchy.

[0064] Specifically, the system calls the regional cultural knowledge base API to obtain cultural elements, and then generates adaptation questions. These cultural elements include dialect vocabulary, regional allusions, and local customs; the cultural adaptation rules include automatically replacing common elements in the story with regionally specific elements based on regional tags.

[0065] Optionally, the method further includes the following steps: The keyword matching algorithm is used to dynamically detect the relevance of the children's answers to the theme of the story. If the relevance exceeds a preset threshold, the child is guided back to the theme of the story.

[0066] Specifically, relevant elements are extracted from a cultural knowledge base based on children's regional information, and a keyword matching algorithm is used to detect the relevance of the answers to the story theme in real time. When a deviation is detected, the system automatically guides the children back to the story theme.

[0067] The following sections will provide a detailed description and explanation of some optional embodiments of this application, using specific application examples.

[0068] The embodiments of this application include the following usage scenarios: 1. Early Childhood Education at Home: Parents apply the system described in this application to a family story robot. During daily story learning, the system provides personalized story questions and interactions based on the child's age, gender, region, and real-time cognitive state, helping the child better understand the story content and improve their language expression, logical thinking, and other abilities. For example, when telling the story of "The Little Tadpoles Looking for Their Mother," the system incorporates elements of southern water town culture into the questions, dynamically adjusting the difficulty and type of questions based on her responses.

[0069] 2. Kindergarten Education Scenario: Kindergartens can integrate the system described in this application into their teaching equipment. In group story-based teaching activities, teachers can use the system to analyze the cognitive characteristics of different children, enabling differentiated instruction. The system can pose questions tailored to each child's characteristics, guiding children's thinking and expression, thereby improving classroom participation and teaching effectiveness.

[0070] 3. Online Education Platform: Applied to online education scenarios such as online story teaching and children's language training, teachers can use the system to provide personalized teaching for different children, thereby improving the quality and effectiveness of online education.

[0071] Main modules and detailed descriptions: (a) Children's cognitive characteristics modeling module.

[0072] 1. Basic profile construction submodule.

[0073] Function: Receives information on the child's age, gender, and location entered by parents and processes it using a rule engine.

[0074] Technical Implementation: Children are divided into different stages based on their age, corresponding to different initial difficulty levels L (L0, L5). For example, the initial difficulty level for a 3-year-old child is L0, and for a 6-year-old child it is L2.

[0075] Based on regional information, relevant elements are extracted from the regional cultural knowledge base to inject regional characteristics into subsequent story content. For example, cultural elements such as "winter wind protection" are automatically added to stories for children in northern regions.

[0076] Adjust the question design bias by incorporating gender information, such as adding spatial reasoning questions for boys and verbal expression questions for girls.

[0077] Data flow: Parents input data → Rule engine processes it → Generates basic profile → Passes it to other sub-modules.

[0078] 2. Comprehension Assessment Submodule.

[0079] Function: To assess children's comprehension of the story content.

[0080] Technical Implementation: The BERT model (bertbasechinese) was used to calculate the semantic similarity between children's answers and standard answers.

[0081] Input text: Story content + child's answer (maximum sequence length 512 tokens).

[0082] Output: Semantic understanding threshold α (α∈[0,1]), the closer α is to 1, the stronger the understanding.

[0083] Parameter configuration: learning rate 2e5, batch size 32, Drput 0.1, using Adam optimizer.

[0084] Data flow: Children's answer data → BERT model processing → generating α value → passing to dynamic selection logic submodule.

[0085] 3. Expressiveness Assessment Submodule.

[0086] Function: To assess children's language expression abilities.

[0087] Technical Implementation: LSTM networks are used to process speech data and extract speech features.

[0088] Speech features include: number of pauses, speech rate, and lexical diversity.

[0089] The formula for calculating the expressiveness score β is: β = (vocabulary diversity × 0.6) + (1 / number of pauses × 0.4).

[0090] β∈[0,100] reflects a child's language expression ability.

[0091] Data flow: Children's voice data → Feature extraction → LSTM network processing → Generate β value → Pass to dynamic selection logic submodule.

[0092] 4. Emotional Engagement Assessment Submodule.

[0093] Function: To assess a child's emotional state during the interaction process.

[0094] Technical Implementation: The speech emotion recognition model is used to analyze the tone, speed and other characteristics of children's answers.

[0095] A multimodal sentiment analysis algorithm is used, combining text sentiment and voice sentiment.

[0096] The formula for calculating emotional engagement U is: U=∑W(Ui) (i=1,2,3,⋯6), where Ui represents the six emotional dimensions of happiness, boredom, anger, sadness, anxiety, and surprise, and the weights are W=(0.3934,0.1520,0.0997,0.1194,0.1296,0.1059).

[0097] Emotional state is represented by the average emotional state score m1 and the number of positive and negative emotions m2.

[0098] Data flow: Children's voice / text data → sentiment analysis model processing → U-value generation → passed to dynamic selection logic submodule.

[0099] (ii) Three-level progressive problem engine module.

[0100] 1. Design sub-modules based on problem hierarchy.

[0101] Function: Design three levels of different types of questions based on Bloom's Taxonomy.

[0102] Technical Implementation: Level 1 (Memory-based questions): Tests children's memory of basic facts in a story, such as "Where does Grandma live?".

[0103] Level 2 (Reasoning Questions): These involve the level of understanding and application, such as "Why did the Big Bad Wolf pretend to be Grandma?"

[0104] Level 3 (Creative Questions): These belong to the analysis and evaluation level, such as "If you were Little Red Riding Hood, what would you do?"

[0105] Each story is designed with a question bank based on these three levels of question types, and the question bank contains 2000+ questions.

[0106] Data flow: Story content → Question database matching → Level 3 question generation → Passed to dynamic selection logic submodule.

[0107] 2. Emotional Memory Submodule.

[0108] Function: Record children's emotional responses to various issues.

[0109] Technical Implementation: Sentiment analysis is performed on historical interaction records using neural network methods.

[0110] Record children's emotional responses (positive / negative / neutral) to memory, reasoning, and creative problems.

[0111] Establish an emotional memory mapping table to record children's emotional tendencies toward specific topics / issues.

[0112] Data flow: Historical interaction data → Sentiment analysis model processing → Generation of emotion memory mapping table → Passed to dynamic selection logic submodule.

[0113] 3. Dynamically select logical sub-modules.

[0114] Function: Dynamically selects the question level based on children's cognitive characteristics.

[0115] Technical Implementation: The problem-level selection probability P is calculated using the Dynamic Routing algorithm: P = α × w1 + β × w2 + U × w3, where w1 = 0.4 (comprehension weight), w2 = 0.3 (expression weight), and w3 = 0.3 (emotion weight).

[0116] Dynamic redirection based on P-value: If P ≥ 0.8 → Creative problem.

[0117] If 0.5 ≤ P < 0.8 → Reasoning problem.

[0118] If P < 0.5 → Memory-related problem.

[0119] By using an emotional memory mapping chart, avoid repeatedly asking questions that children find offensive.

[0120] Data flow: α, β, U values ​​+ emotion memory mapping table → dynamic routing algorithm processing → generation of question level selection → transmission to question generator.

[0121] (iii) Culturally sensitive anti-deviation protocol module.

[0122] 1. Topic Deviation Detection Submodule.

[0123] Function: Real-time detection of the similarity between children's answers and the story theme.

[0124] Technical Implementation: Keyword matching algorithms were used to detect the relevance of the answers to the story's theme.

[0125] The quality of responses is monitored in real time using the Heffding inequality algorithm.

[0126] Topic similarity threshold setting: When the similarity is below 0.3, a deviation warning is triggered.

[0127] Data flow: Children's answers + story theme → keyword matching → similarity calculation → determining whether it deviates from the topic.

[0128] 2. Cultural elements are injected into the submodule.

[0129] Function: Incorporate local cultural elements based on children's regional information.

[0130] Technical Implementation: Use the regional culture knowledge base API to obtain cultural elements.

[0131] Cultural elements include dialect vocabulary, regional allusions, and local customs.

[0132] Cultural adaptation rules: Automatically replace common elements in the story with regionally specific elements based on regional tags.

[0133] Data flow: Regional information → Cultural knowledge base API call → Obtain cultural elements → Inject question generation.

[0134] (iv) Adaptive feedback module.

[0135] 1. Cognitive Model Update Submodule.

[0136] Function: Updates the cognitive model based on children's answers.

[0137] Technical Implementation: Real-time updated alpha value: based on the semantic similarity between the latest answer and the standard answer.

[0138] Real-time updated β value: an expressiveness score based on the latest speech features.

[0139] U-value is updated in real time: based on the latest sentiment analysis results.

[0140] Use a weighted average method to update model parameters to avoid abrupt changes.

[0141] Data flow: Child's answer + emotional feedback → Model parameter update → Passed to cognitive feature modeling module.

[0142] 2. Problem database optimization submodule.

[0143] Function: Optimize the question bank based on children's performance.

[0144] Technical Implementation: Record the accuracy rate of children's answers to various questions.

[0145] Adjust the weight of questions based on their accuracy, and increase the priority of high-quality questions.

[0146] The question bank should be evaluated and updated regularly to ensure that the content is timely.

[0147] Data flow: Answer records → Accuracy analysis → Question weight adjustment → Update question database.

[0148] 3. Cultural adaptation optimization submodule.

[0149] Function: Optimize cultural adaptation strategies based on children's feedback.

[0150] Technical Implementation: Record children’s level of participation in cultural input issues.

[0151] The Analytic Hierarchy Process (AHP) was used to evaluate the effectiveness of cultural adaptation.

[0152] Adjust the strategy for injecting cultural elements based on participation levels.

[0153] Data flow: Cultural question answer records → Participation analysis → Adjustment of cultural adaptation strategies → Update of cultural knowledge base.

[0154] Reference Figure 2 System analysis steps and detailed explanation.

[0155] 1. Story input.

[0156] Function: Receive story content selected by the user or pushed by the system.

[0157] Detailed explanation: It supports multiple story input methods: parents can manually select, or the system can automatically push stories based on children's preferences.

[0158] Story content preprocessing: Extracting story keywords, themes, plot information, etc.

[0159] Establish a story-question mapping relationship to ensure that the questions are closely related to the story content.

[0160] Data flow: Story content → Preprocessing → Extraction of key information → Passing to the multimodal analysis engine.

[0161] 2. Multimodal analysis engine.

[0162] Function: Perform multi-dimensional analysis of story content.

[0163] Detailed explanation: Text analysis: Using NLP techniques to analyze the semantics, difficulty, and emotional tone of a story.

[0164] Cultural analysis: Extracting regional cultural elements from the story.

[0165] Gender analysis: Identifying gender-related elements in a story.

[0166] Data flow: Preprocessed story content → Multimodal analysis → Generation of analysis results → Transfer to the child cognitive feature modeling module.

[0167] 3. Modeling children's cognitive characteristics.

[0168] Function: To build a four-dimensional cognitive model for children.

[0169] Detailed explanation: Basic profile construction: Generate initial difficulty level and regional cultural elements based on age, gender, and regional information.

[0170] Comprehension assessment: The semantic similarity between children's answers and standard answers was calculated using the BERT model to obtain the α value.

[0171] Expressiveness assessment: Speech features were analyzed using an LSTM network, and the β value was calculated.

[0172] Emotional engagement assessment: The U-value is calculated using a multimodal sentiment analysis algorithm.

[0173] Integrating a four-dimensional cognitive model: combining α, β, U values ​​and basic profiles to form complete cognitive features.

[0174] Data flow: Story analysis results + historical interaction data → four-dimensional model construction → passed to the three-level progressive question engine module.

[0175] 4. Three-level progressive problem engine.

[0176] Function: Generates appropriate questions based on children's cognitive characteristics.

[0177] Detailed explanation: Question hierarchy design: Based on the story content and Bloom's Taxonomy, three types of questions are designed: memory, reasoning, and creation.

[0178] Emotional memory analysis: Recording children's emotional responses to various issues.

[0179] Dynamic selection logic: The P-value is calculated using a dynamic routing algorithm, and the problem level is selected based on the P-value.

[0180] Data flow: Four-dimensional cognitive model → dynamic routing algorithm → selection of problem level → transmission to culturally sensitive anti-deviation protocol module.

[0181] 5. Culturally sensitive anti-deviation protocols.

[0182] Function: Ensures the cultural relevance and topic relevance of the questions asked.

[0183] Detailed explanation: Cultural elements are incorporated: relevant elements are extracted from the cultural knowledge base based on the children's regional information.

[0184] Topic deviation detection: Using keyword matching algorithms to detect the relevance of answers to the story topic in real time.

[0185] Anti-deviation strategy: When a deviation is detected, the system automatically guides the child back to the story theme.

[0186] Data flow: Selected question level + regional information → Injection of cultural elements → Generation of adaptive questions → Passing to the adaptive feedback module.

[0187] 6. Adaptive feedback.

[0188] Function: Updates the cognitive model and optimizes the question bank based on children's answers.

[0189] Detailed explanation: Cognitive model update: Update α, β, and U values ​​based on the latest responses.

[0190] Question database optimization: Record the correct answer rate and adjust the question weights.

[0191] Culture adaptation optimization: Analyze the effects of culture injection issues and optimize culture adaptation strategies.

[0192] Data flow: Child's answer + emotional feedback → model update + question optimization → passed to child cognitive feature modeling module.

[0193] Examples and data.

[0194] (I) Example 1: Age-based cognitive feature modeling.

[0195] 1. Implementation scenarios.

[0196] Target participants: 3 children aged 3, 3 children aged 5, and 3 children aged 6.

[0197] Equipment: Intelligent story robot (including speech recognition and emotion analysis modules).

[0198] The stories include: "The Tadpoles Looking for Their Mother," "The Little Bear Crossing the River," and "The Three Little Pigs."

[0199] 2. Implementation steps.

[0200] Step 1: Parents enter the child's age, gender, and location information.

[0201] Step 2: The system divides the initial difficulty level according to age (3 years old → L0, 5 years old → L1, 6 years old → L2).

[0202] Step 3: Children listen to the story and answer questions.

[0203] Step 4: The system evaluates comprehension α, expressiveness β, and emotional engagement U in real time.

[0204] Step 5: The dynamic routing algorithm calculates the P value and selects the problem level.

[0205] Step 6: Culturally sensitive deviation prevention protocols ensure the cultural appropriateness of the questions asked.

[0206] Step 7: The adaptive feedback module updates the cognitive model and optimizes the question library.

[0207] 3. Experimental data are shown in Table 1: Table 1

[0208] Experimental conclusion: The system can dynamically adjust the difficulty of questions based on the child's age to achieve accurate matching.

[0209] The accuracy rate of 3-year-old children on memory-related questions has improved significantly.

[0210] The performance of 5-year-old children on reasoning questions has been improved.

[0211] Six-year-old children demonstrate stronger thinking abilities when faced with creative problems.

[0212] Example 2: Question generation based on regional culture adaptation.

[0213] 1. Implementation scenarios.

[0214] Target participants: 3 children from the south and 3 children from the north.

[0215] Equipment: Intelligent storytelling robot (including regional cultural knowledge base API).

[0216] Story content: "The Little Tadpole Looking for Its Mother" and "The Snow Child".

[0217] 2. Implementation steps.

[0218] Step 1: Parents enter the child's geographic information.

[0219] Step 2: The system retrieves relevant cultural elements from the regional cultural knowledge base.

[0220] Step 3: Children listen to the story and answer questions.

[0221] Step 4: The cultural element injection submodule incorporates regional characteristic elements into the problem.

[0222] Step 5: The topic deviation detection submodule ensures that the question content is relevant to the story topic.

[0223] Step 6: The adaptive feedback module records the cultural adaptation effect and optimizes the strategy.

[0224] 3. Experimental data are shown in Table 2: Table 2

[0225] Experimental conclusion: The system can effectively incorporate cultural elements based on children's regional information, thereby increasing their participation in asking questions.

[0226] Southern children showed a significantly higher level of participation (+20%) in questions containing elements of southern culture.

[0227] Northern children also showed a significantly higher level of participation in questions containing elements of northern culture.

[0228] Even for non-regional stories, culturally relevant questions can still increase engagement (Southern children's engagement with "The Snow Child" increased by 20%).

[0229] Compared with the prior art, the embodiments of this application include the following inventive differences: 1. Comparison with CN110427477A.

[0230] CN110427477A: Fixed questions are generated based on keyword extraction and weight values. The questions are monotonous and cannot be dynamically adjusted according to children's cognitive abilities.

[0231] This application's embodiment constructs a four-dimensional cognitive model (comprehension α, expression β, emotional engagement U, and historical interaction), and uses a dynamic routing algorithm to calculate the P-value in real time, enabling intelligent switching between memory, reasoning, and creation questions.

[0232] Novelty: For the first time, multimodal assessment (voice + text + emotion) is combined with dynamic question adjustment to achieve a truly personalized questioning experience.

[0233] 2. Comparison with CN117877052A.

[0234] CN117877052A: Story content is generated by constructing anchor information through user input, but it does not involve a personalized dynamic mechanism for asking questions.

[0235] This application's embodiment not only focuses on content generation but also on the personalized and dynamic adjustment of questions, designing a three-level question engine using Bloom's Taxonomy.

[0236] Novelty: This is the first time that Bloom's Taxonomy has been applied to the hierarchical design of questions in children's stories, forming a complete cognitive development support system.

[0237] 3. Comparison with CN120182061A.

[0238] CN120182061A: Predicts learning efficiency and dynamically adjusts course difficulty based on student performance, but does not involve the dynamic generation of questions.

[0239] This application's embodiment: Design a three-level progressive problem engine, which realizes intelligent jumping between problem levels through a dynamic routing algorithm.

[0240] Novelty: This is the first time that a dynamic routing algorithm has been applied to question selection in an educational setting, achieving a precise match between the difficulty of the questions and the children's cognitive abilities.

[0241] 4. Comparison with CN119938849A.

[0242] CN119938849A: Adjusts the difficulty of animated dialogue based on the user's cognitive level, but only for language comprehension.

[0243] This application's embodiment constructs a four-dimensional cognitive model and achieves comprehensive adjustment of question content and difficulty through multimodal assessment (voice + text + emotion).

[0244] Novelty: This is the first time that multimodal assessment has been combined with question generation to achieve more comprehensive cognitive support.

[0245] 5. Comparison with CN116595181A.

[0246] CN116595181A: The dialogue content is optimized through sentiment analysis, but questions are not dynamically generated in conjunction with a cognitive model.

[0247] This application's embodiment: Design an adaptive feedback module that combines sentiment analysis results with cognitive model updates to form a closed-loop system.

[0248] Novelty: For the first time, sentiment analysis is combined with cognitive model construction and dynamic question selection to achieve more effective learning guidance.

[0249] The beneficial effects of the embodiments of this application include: 1. Precise and personalized interaction.

[0250] A comprehensive assessment of children's cognitive status is conducted using a four-dimensional cognitive model (age, gender, region + comprehension α, expressive ability β, emotional engagement U, and historical interaction).

[0251] The dynamic routing algorithm (P=α×0.4+β×0.3+U×0.3) achieves a precise match between the difficulty of the questions and the child's ability.

[0252] Compared with existing technologies, the question matching accuracy of the embodiments of this application is improved by more than 20%, which significantly improves children's participation and learning effect in story learning.

[0253] 2. Cross-cultural and gender universality.

[0254] Consider regional cultural differences (via regional cultural knowledge base API calls) and gender awareness development characteristics (adjust question bias through gender labels).

[0255] To provide suitable interactive content for children of different regions and genders, and to expand the scope of application.

[0256] In tests conducted on children in the south, participation in questions containing regional cultural elements increased by 30%, while in tests conducted on boys, the accuracy rate of answers to spatial reasoning questions increased by 15%.

[0257] 3. Effective dialogue guidance.

[0258] The culturally sensitive anti-deviation protocol works in conjunction with the three-level progressive question engine to effectively prevent the conversation from getting out of control.

[0259] The dynamic jump mechanism adjusts in real time according to the child's learning progress, improving learning efficiency.

[0260] In topic deviation detection tests, this patent achieved an accuracy rate of 93%, far exceeding that of traditional methods.

[0261] The inventiveness of the embodiments in this application includes: 1. Multi-technology integration and innovation.

[0262] For the first time, Natural Language Processing (BERT), Machine Learning (LSTM), cognitive model building, Bloom's Taxonomy, cultural adaptation, and sentiment analysis technologies are organically integrated to form a complete interactive and proactive questioning system for children's stories.

[0263] By fusing multimodal data (voice fluency, text semantics, and emotional state), a four-dimensional cognitive model is constructed to achieve a comprehensive assessment of children's cognitive characteristics.

[0264] This multi-technology integration approach is unprecedented in existing patents and represents an innovative breakthrough in story robot interaction technology.

[0265] 2. Innovative mechanism design.

[0266] A three-level progressive question engine dynamic jump mechanism is proposed, and a progressive questioning path of memory → reasoning → creation is designed based on Bloom's taxonomy.

[0267] An innovative dynamic routing algorithm is introduced, which realizes intelligent selection at the problem level through a weighted formula (P=α×0.4+β×0.3+U×0.3).

[0268] We designed a culturally sensitive anti-deviation protocol, which uses a regional cultural knowledge base injection and topic deviation detection algorithms to ensure the cultural suitability and topic relevance of the questions.

[0269] An adaptive feedback closed-loop system is constructed to use children's answers and emotional feedback in real time to update the cognitive model and optimize the question bank.

[0270] These innovative designs address core issues that have long existed in existing technologies, providing new technological ideas and solutions for the field of interactive children's stories, and have outstanding substantive features and significant progress.

[0271] Practical application description of the embodiments in this application: (a) Application scenarios: 1. Smart Educational Hardware Products: These products can be integrated into various children's story robots, smart learning tablets, and other hardware devices, giving them more powerful intelligent interactive functions, enhancing their competitiveness, and meeting the needs of parents and children for high-quality educational products.

[0272] 2. Online Education Platform: Applied to online education scenarios such as online story teaching and children's language training, teachers can use the system to provide personalized teaching for different children, thereby improving the quality and effectiveness of online education.

[0273] 3. In the field of children's cultural dissemination: In scenarios such as children's cultural dissemination activities and children's guided tours in museums, this patented method, combined with different regional and cultural backgrounds, disseminates cultural knowledge to children in the form of interactive storytelling, enhancing children's awareness and understanding of multiculturalism.

[0274] 4. Special Education Institutions: Provide appropriate cognitive development support for children with special educational needs (such as children with autism and learning disabilities) to help them improve their abilities in a relaxed story atmosphere.

[0275] (ii) Value: 1. Educational value.

[0276] Through precise and personalized questioning, children's cognitive abilities in language expression, logical thinking, creativity, and other aspects can be effectively improved, thus promoting their all-round development.

[0277] To provide more scientific and effective tools and methods for early childhood education and promote the improvement of education quality.

[0278] 2. Commercial value.

[0279] For educational technology companies, this patented technology can enhance the market competitiveness of their products, attract more users, and bring significant economic benefits.

[0280] This technology can be used to expand related value-added services, such as personalized educational content subscriptions and cultural course development, creating more business opportunities.

[0281] 3. Social value.

[0282] To promote educational equity so that children from different regions and backgrounds can enjoy high-quality, personalized educational resources.

[0283] Integrating cultural elements helps to inherit and promote diverse cultures, and enhances children's cultural identity and national pride.

[0284] In special education settings, this system can help children with autism improve their social skills, increasing the success rate and having significant social implications.

[0285] Reference Figure 3 This application also provides a story-based interactive active questioning system based on children's cognitive characteristics, which can implement the above-mentioned story-based interactive active questioning method based on children's cognitive characteristics. The system includes: The story acquisition unit is used to acquire story content; The story analysis unit is used to perform multimodal analysis on the story content and obtain analysis results; A cognitive building unit is used to construct a four-dimensional cognitive model of the child based on the analysis results and the child's historical interaction data. The level selection unit is used to select the target question level from a preset multi-level question based on the four-dimensional cognitive model. The problem generation unit is used to generate adaptation problems corresponding to the target problem level; An adaptive feedback unit is used to update the four-dimensional cognitive model and the multi-level questions based on the child's answers to the adaptation questions and emotional feedback.

[0286] It is understood that the content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0287] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the method of this application. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0288] It is understood that the content of the above method embodiments is applicable to the device embodiments. The specific functions implemented by the device embodiments are the same as those of the methods of this application, and the beneficial effects achieved are the same as those achieved by the methods of this application.

[0289] Please see Figure 4 , Figure 4 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 401 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 402 can be implemented as a read-only memory (RM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 402 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 402 and is called and executed by the processor 401. Input / output interface 403 is used to implement information input and output; The communication interface 404 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 405 transmits information between various components of the device (e.g., processor 401, memory 402, input / output interface 403, and communication interface 404); The processor 401, memory 402, input / output interface 403 and communication interface 404 are connected to each other within the device via bus 405.

[0290] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method of this application.

[0291] It is understood that the content of the above method embodiments is applicable to this storage medium embodiment. The specific functions implemented in this storage medium embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0292] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0293] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0294] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0295] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0296] The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0297] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0298] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0299] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0300] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between systems or units may be electrical, mechanical, or other forms.

[0301] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0302] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0303] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (RM), random access memory (RAM), magnetic disks, or optical disks.

[0304] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A story-based interactive active questioning method based on children's cognitive characteristics, characterized in that, The method includes the following steps: Get the story content; Multimodal analysis was performed on the story content to obtain the analysis results; Based on the analysis results and the child's historical interaction data, a four-dimensional cognitive model of the child is constructed; Based on the four-dimensional cognitive model, the target question level is selected from the preset multi-level questions; Generate the adaptation problem corresponding to the target problem level; The four-dimensional cognitive model and the multi-level questions are updated based on the child's answers to the adaptation questions and emotional feedback. The construction of the child's four-dimensional cognitive model based on the analysis results and the child's historical interaction data includes the following steps: The basic profile of the child, the child's semantic understanding threshold of the story content, the expressive ability score, and the emotional engagement are obtained as the historical interaction data; The four-dimensional cognitive model is constructed based on the analysis results, the basic profile of the child, and the historical interaction data. The step of selecting the target question level from a preset multi-level question based on the four-dimensional cognitive model includes the following steps: The historical interaction data is processed using an emotion analysis model to obtain the child's emotion memory mapping table; The target question hierarchy is generated using a dynamic routing algorithm based on the semantic understanding threshold, the expressiveness score, the emotional engagement, and the emotion memory mapping table.

2. The story-based interactive active questioning method based on children's cognitive characteristics according to claim 1, characterized in that, The process of obtaining the child's basic profile, the child's semantic understanding threshold of the story content, expressive ability score, and emotional engagement as the historical interaction data includes the following steps: A basic profile of the child is constructed based on the child's age, gender, and geographic location information; The BERT model is used to calculate the semantic similarity between the child's historical answer text and the standard answer, thereby generating the child's semantic understanding threshold of the story content; An LSTM network is used to extract speech features from the child's historical answers to questions, and then a score of the child's expressive ability to the story content is generated. The emotional engagement of the child with the story content is generated using a sentiment analysis model based on the historical response text and the historical response speech.

3. The story-based interactive active questioning method based on children's cognitive characteristics according to claim 1, characterized in that, The method further includes the following steps: Match the question corresponding to the story content in the question bank; Based on the matched questions, three levels of questions are generated in sequence: memory-related questions, reasoning-related questions, and creative questions, which serve as the preset multi-level questions.

4. The story-based interactive active questioning method based on children's cognitive characteristics according to claim 1, characterized in that, The process of generating the adaptation problem corresponding to the target problem level includes the following steps: Cultural elements are generated based on the regional information in the child's basic portrait. The corresponding adaptation question is generated based on the cultural elements and the target question hierarchy.

5. The story-based interactive active questioning method based on children's cognitive characteristics according to claim 4, characterized in that, The method further includes the following steps: The keyword matching algorithm is used to dynamically detect the relevance of the children's answers to the theme of the story. If the relevance exceeds a preset threshold, the child is guided back to the theme of the story.

6. A story-based interactive active questioning system based on children's cognitive characteristics, characterized in that, The system is used to implement the story-based interactive active questioning method based on children's cognitive characteristics as described in claim 1, the system comprising: The story acquisition unit is used to acquire story content; The story analysis unit is used to perform multimodal analysis on the story content and obtain analysis results; A cognitive building unit is used to construct a four-dimensional cognitive model of the child based on the analysis results and the child's historical interaction data. The level selection unit is used to select the target question level from a preset multi-level question based on the four-dimensional cognitive model. The problem generation unit is used to generate adaptation problems corresponding to the target problem level; An adaptive feedback unit is used to update the four-dimensional cognitive model and the multi-level questions based on the child's answers to the adaptation questions and emotional feedback.

7. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 5.

Images