A method, device, medium and product for intelligent intervention of children's abilities.

Abstract

This invention relates to an intelligent method, device, medium, and product for collaborative intervention of children's abilities, and pertains to the field of data processing technology. Implementing this method, a computer device collects children's complex interactive behavioral information and transforms it into multimodal data packets, achieving a comprehensive capture of children's ability status. This involves both precise location of ability intensity values ​​for different ability dimensions through semantic analysis and in-depth mining of sensitive period intensity values ​​for different ability dimensions through acoustic analysis. Based on the ability intensity values ​​and sensitive period intensity values, the computer device determines the target ability dimension for which collaborative intervention is currently needed, ensuring the targeted and timely nature of the intervention and avoiding indiscriminate intervention. Subsequently, the computer device loads a target ability development ladder model corresponding to the target ability dimension, generating a personalized target ability development ladder strategy for the child, and integrating the child's personalized interest characteristics to push a set of intervention content.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to an intelligent method, device, medium and product for collaborative intervention of children's abilities. Background Technology

[0002] With the continuous advancement of educational philosophies, the cultivation and development of children's abilities are receiving increasing attention. Research shows that children exhibit different developmental characteristics at different age stages. Scientifically understanding the developmental patterns and individual differences of children, and providing targeted ability cultivation, is of great significance to children's healthy growth. Therefore, how to better identify children's developmental status and provide personalized intervention plans has become an important issue in the current education field.

[0003] Currently, children's ability intervention systems typically employ fixed assessment scales and pre-set intervention plans. This involves collecting basic data on children through methods such as questionnaires and behavioral observations, then scoring children's performance on different ability dimensions based on pre-set assessment criteria, and finally selecting appropriate intervention content from an intervention content library based on the assessment results to develop an intervention plan.

[0004] However, the above methods have significant limitations in practical application. Because they only focus on children's weaknesses and ignore their learning motivation, emotional state, and developmental sensitive periods, the intervention results in less effective intervention. Summary of the Invention

[0005] This application provides an intelligent method, device, medium, and product for collaborative intervention of children's abilities, which addresses the problem of insufficient intervention accuracy caused by the lack of multimodal data collaborative analysis and personalized decision-making mechanisms in traditional methods for children's ability intervention.

[0006] In a first aspect, this application provides an intelligent method for collaborative intervention of children's abilities, applied to a computer device. The method includes: collecting complex interactive behavior information generated by children responding to preset tasks in preset situations; performing speech recognition and acoustic analysis on the complex interactive behavior information to obtain a multimodal data packet, which includes speech text input and acoustic signal data; parsing the speech text input through a semantic analysis module to obtain the speech content structure, thereby determining the ability intensity value corresponding to each of the child's different ability dimensions; parsing the acoustic signal data through an acoustic analysis module to obtain the acoustic expression pattern, thereby determining the sensitive period intensity value corresponding to each of the child's different ability dimensions; calculating the intervention priority score corresponding to each of the child's different ability dimensions based on the ability intensity value and the sensitive period intensity value; and then... The ability dimension with the highest pre-priority score is determined as the target ability dimension that currently requires collaborative intervention. The target ability development ladder model corresponding to the target ability dimension is loaded, and combined with the target ability intensity value corresponding to the target ability dimension, the child's current target ability level is determined. A target ability development ladder strategy is generated. The target ability development ladder model includes different ability intensity value ranges corresponding to different ability ladders of the target ability dimension, different state performance descriptions, and different next ability development indicators. The target ability development ladder strategy refers to the improvement plan from the target ability level to the next ability level. The target ability development ladder strategy is integrated with the child's personalized interest characteristics to obtain a content query vector. The content query vector is used to search in a preset intervention content library to obtain an intervention content set, which is then pushed to the relevant authorities.

[0007] By employing the aforementioned technical solution, computer devices collect children's complex interactive behavioral information and transform it into multimodal data packets, achieving a comprehensive capture of children's ability status. This involves both precise location of ability strength values ​​across different ability dimensions through semantic analysis and in-depth mining of sensitive period strength values ​​for different ability dimensions through acoustic analysis, providing dual-dimensional data support for intervention decisions. Based on ability strength values ​​and sensitive period strength values, the computer devices determine the target ability dimension for which the child currently requires coordinated intervention, ensuring the targetedness and timeliness of the intervention and avoiding indiscriminate intervention. Subsequently, the computer devices load the target ability development ladder model corresponding to the target ability dimension, generating personalized target ability development ladder strategies for the child. Furthermore, they integrate the child's personalized interest characteristics to push a set of intervention content, ensuring a high degree of alignment between the intervention content and ability development goals while fully considering the child's willingness to accept it, significantly improving the attractiveness and participation of the intervention content, thereby greatly optimizing the intervention effect.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, the speech text input is parsed by a semantic analysis module to obtain the speech content structure, so as to determine the ability strength value corresponding to each child in different ability dimensions. Specifically, this includes: inputting the speech text into a pre-trained semantic understanding model and outputting the speech content structure; identifying multiple key feature words based on the speech content structure; classifying and mapping the multiple key feature words based on a preset ability dimension feature word list to obtain a set of key feature words under different ability dimensions; calculating the keyword frequency and word importance score under each ability dimension based on the set of key feature words, and determining the ability strength value corresponding to each ability dimension by weighting the keyword frequency and word importance score.

[0009] By employing the aforementioned technical solutions, the computer device, based on a pre-trained semantic understanding model, deeply analyzes the speech-text input to obtain the speech content structure, accurately extracting key feature words and avoiding the subjectivity and bias of manual analysis. The computer device uses a pre-set feature word list for ability dimensions to perform classification mapping, establishing a clear association between key feature words and ability dimensions, ensuring the rationality of feature classification. The computer device determines the ability strength value by weighting keyword frequency and word importance scores, considering both the frequency of key feature words and their importance in the corresponding ability dimensions, making the calculation results more accurately reflect the child's ability level. This refined semantic analysis method lays a solid data foundation for the subsequent calculation of intervention priority scores and the determination of target ability dimensions, ensuring the scientific nature and effectiveness of the entire intervention process.

[0010] In conjunction with some embodiments of the first aspect, in some embodiments, acoustic signal data is parsed by an acoustic analysis module to obtain acoustic expression patterns, thereby determining the intensity values ​​of sensitive periods corresponding to different ability dimensions for children. Specifically, this includes: extracting acoustic features from the acoustic signal data to obtain an acoustic feature parameter set, which includes pitch, intensity, speech rate, and pauses; inputting the acoustic feature parameter set into a pre-trained acoustic expression pattern classifier to obtain acoustic expression patterns, which include emotional activation, engagement, and interest; and converting the acoustic expression patterns into the intensity values ​​of sensitive periods corresponding to different ability dimensions through a preset ability dimension sensitive period feature model.

[0011] By employing the aforementioned technical solution, firstly, the computer device extracts key acoustic feature parameters such as pitch, intensity, speech rate, and pauses to comprehensively capture the details of children's acoustic expressions during their responses to pre-defined tasks in pre-defined scenarios. These acoustic feature parameters can indirectly reflect children's emotional and participation states. Then, the computer device inputs these acoustic feature parameters into a pre-trained acoustic expression pattern classifier to obtain acoustic expression patterns including emotional activation, participation, and interest levels, thus transforming raw data into meaningful information. Next, the computer device maps the acoustic expression patterns to the sensitivity period intensity values ​​of each ability dimension through a pre-defined sensitive period feature model for each ability dimension, establishing a correlation between acoustic features and children's developmental sensitive periods. This allows intervention decisions to fully grasp the key moments in children's ability development. This method allows interventions to not only focus on "ability gaps" but also on "timing windows," ensuring that intervention measures are implemented during the golden period of children's ability development. This significantly improves the efficiency and effectiveness of interventions, providing strong support for achieving precise and timely interventions for children's abilities.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, based on the ability intensity value and the sensitive period intensity value, the intervention priority score corresponding to each child in different ability dimensions is calculated. Specifically, this includes: substituting the ability intensity value into the formula for calculating the urgency of ability shortcomings to obtain the urgency of each child's ability shortcomings in different ability dimensions; mapping the sensitive period intensity value to the maturity of the timing window; and calculating the intervention priority score corresponding to each child in different ability dimensions based on the urgency of ability shortcomings and the maturity of the timing window. The formula for calculating the urgency of ability shortcomings is: Where Ud represents the urgency of a child's skill gaps under skill dimension d, Ed (age) represents the expected skill strength of a child at their current age under skill dimension d, and Sd represents the skill strength of a child under skill dimension d.

[0013] By employing the aforementioned technical solutions, the computer device substitutes the ability intensity value into the formula for calculating the urgency of ability shortcomings. This formula precisely quantifies the degree of shortcomings in each ability dimension, avoiding subjective judgments about "shortcomings." The computer device maps the sensitivity period intensity value to the maturity of the timing window, capturing the critical developmental periods for children in each ability dimension, ensuring that interventions seize the "golden opportunity." Finally, based on the urgency of ability shortcomings and the maturity of the timing window, the computer device calculates an intervention priority score, achieving synergistic consideration—prioritizing interventions in the most prominent ability dimensions while also considering whether those dimensions are in critical periods. This avoids forcibly intervening in shortcomings during non-critical periods, leading to poor results, or missing opportunities to efficiently improve abilities within critical periods. This quantitative priority calculation method makes the determination of target ability dimensions more scientific and objective, laying a crucial foundation for the precise implementation of subsequent intervention measures and significantly improving the rationality and effectiveness of intervention decisions.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, the target ability development ladder strategy is fused with the child's personalized interest characteristics to obtain a content query vector. The content query vector is then used to search a preset intervention content library to obtain an intervention content set. Specifically, this includes: encoding the target ability development ladder strategy into a strategy feature vector; extracting the child's personalized interest characteristics from the child's historical interaction data to construct an interest profile vector; fusing the strategy feature vector and the interest profile vector to obtain a fused feature vector; performing dimensionality reduction on the fused feature vector to obtain a content query vector; calculating the cosine similarity between the content query vector and each content feature vector in the preset intervention content library; and selecting the top N content feature vectors with the highest cosine similarity to obtain the intervention content set.

[0015] By adopting the above technical solution, the computer device encodes the target ability development ladder strategy into a strategy feature vector, clarifying the ability improvement direction and requirements that the intervention content must meet. The computer device extracts children's personalized interest features from historical interaction data to construct an interest profile vector, accurately capturing children's preferences (such as favorite animation styles, game types, etc.). Through feature fusion and dimensionality reduction, the computer device obtains a content query vector, achieving a deep integration of "ability improvement goals" and "children's interests," avoiding the problem of intervention content focusing solely on ability improvement while neglecting children's receptiveness. Subsequently, the computer device uses cosine similarity to select the optimal set of intervention content, ensuring that the pushed intervention content not only meets the target ability development ladder strategy but also highly matches the children's personalized interest characteristics. This interest-integrated content retrieval method significantly enhances children's willingness to actively participate in intervention content—children exhibit higher concentration and interactive enthusiasm when exposed to content they are interested in, thereby improving the absorption effect of intervention content and making the ability improvement process more efficient and sustainable.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, after calculating the intervention priority scores corresponding to children in different ability dimensions based on ability intensity values ​​and sensitive period intensity values; and determining the ability dimension with the highest intervention priority score as the target ability dimension that currently needs collaborative intervention, the method further includes: obtaining the child's historical intervention data; calculating the historical intervention frequency and historical intervention effect score of the target ability dimension based on the historical intervention data; and updating the ability dimension with the second highest intervention priority score as the target ability dimension that currently needs collaborative intervention when the historical intervention frequency is greater than a preset frequency threshold and the historical intervention effect score is lower than a preset first score threshold.

[0017] By adopting the above technical solution, after determining the target ability dimension with the highest intervention priority score, the computer device does not directly execute the intervention. Instead, it first obtains the child's historical intervention data and calculates the historical intervention frequency and historical intervention effect score for the target ability dimension. If a certain ability dimension has been frequently intervened in (historical intervention frequency is greater than a preset frequency threshold), but the effect is consistently poor (historical intervention effect score is lower than a preset first score threshold), it indicates that there may be problems such as mismatched intervention methods or difficulty in the child's current stage of development. In this case, continuing to use it as the target ability dimension would not only consume a lot of intervention resources but may also damage the child's self-confidence. Therefore, the computer device updates the ability dimension with the second highest intervention priority score as the target ability dimension, achieving "dynamic adjustment": it avoids blindly persisting in ineffective intervention directions while promptly shifting to other ability dimensions with greater intervention value. This ensures that each intervention focuses on ability dimensions that are "more likely to be effective and more worthwhile to invest in," optimizing the allocation efficiency of intervention resources and improving the overall intervention plan's fault tolerance and effectiveness.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, after the steps of integrating the target ability development ladder strategy with the child's personalized interest characteristics to obtain a content query vector, searching in a preset intervention content library using the content query vector to obtain an intervention content set, and pushing the intervention content set, the method further includes: monitoring the child's interactive behavior data on the intervention content set, calculating an interaction effect score based on the interactive behavior data; when the interaction effect score is lower than a preset second score threshold, updating the content query vector, re-searching in the preset intervention content library, and pushing the re-matched intervention content set to the child.

[0019] By adopting the above technical solution, the computer device does not end the intervention process after pushing the set of intervention content. Instead, it continuously monitors the child's interactive behavior data (such as interaction duration, operation frequency, completion rate, etc.) and calculates an interaction effect score based on this data. If the interaction effect score is lower than a preset second score threshold, it indicates that the current intervention content may have problems such as "interest mismatch" (e.g., the child has no interest in the content, resulting in less interaction) or "difficulty inappropriate" (e.g., the content is too difficult, making it impossible to complete). At this time, the computer device re-retrieves intervention content by updating the content query vector, which is equivalent to optimizing the "strategy feature vector" or "interest profile vector," thereby matching intervention content that is more suitable for the child's current state. This closed-loop mechanism of "push-monitoring-evaluation-adjustment" transforms the intervention content from "fixed output" to "dynamic adaptation," enabling real-time response to the child's feedback during the intervention process. It avoids intervention interruptions or poor results due to inappropriate content, ensuring that the intervention process is always in a state of "content adaptation, child participation, and controllable effect," significantly enhancing the flexibility and targeting of the intervention plan.

[0020] In a second aspect, embodiments of this application provide a computer device comprising: one or more processors and a memory; the memory is coupled to the one or more processors and is used to store computer program code, the computer program code including computer instructions, wherein the one or more processors invoke the computer instructions to cause the computer device to perform the method described in the first aspect and any possible implementation thereof.

[0021] Thirdly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a computer device, cause the computer device to perform the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a computer device, cause the computer device to perform the method described in the first aspect and any possible implementation thereof.

[0023] It is understood that the computer device provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the methods provided in the embodiments of this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. By adopting the above technical solution, the computer device collects children's complex interactive behavioral information and transforms it into multimodal data packets, achieving a comprehensive capture of children's ability status. It accurately locates the ability intensity values ​​of different ability dimensions through semantic analysis and deeply mines the sensitive period intensity values ​​of different ability dimensions through acoustic analysis, providing dual-dimensional data support for intervention decisions. Based on the ability intensity values ​​and sensitive period intensity values, the computer device determines the target ability dimension for which the child currently needs collaborative intervention, ensuring the targetedness and timeliness of the intervention and avoiding blind intervention. Subsequently, the computer device loads the target ability development ladder model corresponding to the target ability dimension, generating a personalized target ability development ladder strategy for the child, and integrating the child's personalized interest characteristics to push a set of intervention content. This not only ensures a high degree of alignment between the intervention content and the ability development goals but also fully considers the child's willingness to accept it, significantly improving the attractiveness and participation of the intervention content, thereby greatly optimizing the intervention effect.

[0025] 2. By adopting the above technical solution, the computer device substitutes the ability intensity value into the ability deficiency urgency calculation formula to calculate the urgency of ability deficiencies. This formula accurately quantifies the degree of deficiency in each ability dimension, avoiding subjective judgments about "deficiencies." The computer device maps the sensitive period intensity value to the timing window maturity, capturing the critical developmental periods for children in each ability dimension, ensuring that interventions seize the "golden opportunity." Finally, based on the ability deficiency urgency and timing window maturity, the computer device calculates an intervention priority score, achieving synergistic consideration—prioritizing intervention in the most prominent ability dimensions while also considering whether the ability dimension is in a critical period. This avoids forcibly intervening in deficient abilities during non-critical periods, leading to poor results, or missing opportunities to efficiently improve abilities during critical periods. This quantitative priority calculation method makes the determination of target ability dimensions more scientific and objective, laying a crucial foundation for the precise implementation of subsequent intervention measures and significantly improving the rationality and effectiveness of intervention decisions.

[0026] 3. By adopting the above technical solution, after determining the target ability dimension with the highest intervention priority score, the computer device does not directly execute the intervention. Instead, it first obtains the child's historical intervention data and calculates the historical intervention frequency and historical intervention effect score for the target ability dimension. If a certain ability dimension has been frequently intervened in (historical intervention frequency is greater than a preset frequency threshold), but the effect is consistently poor (historical intervention effect score is lower than a preset first score threshold), it indicates that there may be problems such as mismatched intervention methods or difficulty in the child's current stage of development. In this case, continuing to use it as the target ability dimension would not only consume a lot of intervention resources but may also damage the child's self-confidence. Therefore, the computer device updates the ability dimension with the second highest intervention priority score as the target ability dimension, achieving "dynamic adjustment": it avoids blindly persisting in ineffective intervention directions while promptly shifting to other ability dimensions with greater intervention value, ensuring that each intervention focuses on ability dimensions that are "more effective and more worthwhile to invest in," optimizing the allocation efficiency of intervention resources, and improving the fault tolerance and effectiveness of the overall intervention plan. Attached Figure Description

[0027] Figure 1 This is a flowchart illustrating an intelligent collaborative intervention method for children's abilities in an embodiment of this application; Figure 2 This is another flowchart illustrating the intelligent children's ability collaborative intervention method in the embodiments of this application; Figure 3 This is a schematic diagram of the physical device structure of a computer device in an embodiment of this application. Detailed Implementation

[0028] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification of this application, the singular expressions “a,” “an,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0029] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0030] The following describes the process of the method provided in this implementation. Please refer to [link / reference]. Figure 1This is a flowchart illustrating an intelligent children's ability collaborative intervention method in an embodiment of this application.

[0031] S101. Collect complex interactive behavior information generated by children when responding to preset tasks in preset situations, perform speech recognition and acoustic analysis on complex interactive behavior information to obtain multimodal data packets, which include speech text input and acoustic signal data. Among them, the preset scenario refers to a specific setting for children's ability assessment or daily activities, such as game challenges, question-and-answer interactions, and storytelling. The preset task refers to the specific activity that children are required to complete within the preset scenario, such as answering questions, completing game levels, or telling stories. Composite interactive behavior information represents multi-dimensional behavioral data such as voice, facial expressions, and actions generated by children in responding to preset tasks within the preset scenario. The multimodal data package refers to a structured data set that has undergone preliminary processing, including voice and text input and acoustic signal data that can be used for subsequent analysis.

[0032] Specifically, when a child enters a preset scenario, the computer device uses sensors to collect information on the child's complex interactive behaviors, such as voice and facial expressions, as they respond to preset tasks. Next, the computer device preprocesses this collected information: it recognizes the speech to obtain text content while preserving the acoustic features of the speech, such as pitch and speech rate. The computer device then packages this preprocessed data according to a predetermined format, forming a multimodal data packet that includes both voice and text input and acoustic signal data, preparing the data for subsequent in-depth analysis.

[0033] The scenario is set as follows: User: Lele, 4 years old; Interactive device: A smart story machine that incorporates the intelligent children's ability collaborative intervention method as described in the embodiments of this application; Interactive scenario: The story machine shows Lele a picture of a kitten chasing butterflies in the garden and asks, "Lele, what do you see?" Action execution: Lele, facing the story machine, said in a hesitant but excited tone, "Cat. Butterfly. Run." Internal processing: Automatic Speech Recognition (ASR): Converted Lele's speech into text: "Cat. Butterfly. Run." Acoustic analysis: not only listening to the "content", but also analyzing the "sound" itself; Pitch: Steady, indicating that Lele is very focused; Energy: High, indicating that Lele is emotionally excited and highly engaged; Speech Rate: Slow, with noticeable pauses between words, indicating that he is trying to think about how to express himself.

[0034] S102. The semantic analysis module parses the speech text input to obtain the speech content structure, so as to determine the ability intensity value corresponding to each child in different ability dimensions; the acoustic analysis module parses the acoustic signal data to obtain the acoustic expression pattern, so as to determine the sensitive period intensity value corresponding to each child in different ability dimensions. The semantic analysis module is used to parse speech-text input. Speech content structure refers to the semantic hierarchy and key information of the speech-text. The ability strength value represents the child's developmental level score in a specific ability dimension. The acoustic analysis module is used to analyze acoustic signal data. Acoustic expression patterns refer to the emotional, attitudinal, and other expressive features contained in the sound. The sensitive period intensity value represents the child's developmental sensitivity in a specific ability dimension.

[0035] Specifically, the computer device analyzes the multimodal data packets in two directions: on the one hand, the semantic analysis module parses the speech-text input, identifies keywords and semantic structures in the speech-text input, and calculates the child's ability strength value in each ability dimension by combining preset ability dimension assessment standards; on the other hand, the acoustic analysis module analyzes the sound signal data, extracts acoustic expression features such as emotions and attitudes, and converts these acoustic expression features into sensitive period strength values ​​for each ability dimension to reflect the child's current developmental sensitive period status.

[0036] The capability dimensions mainly include five areas, as shown below: Health sector (60 indicators not listed here) Motor development: gross motor skills, fine motor skills, body coordination; Lifestyle habits: self-care ability, hygiene habits, regular routine; Health awareness: body awareness, health knowledge, and safety awareness; Safety awareness: hazard identification, emergency response, and self-protection; Language domain (70 indicators not listed here) Listening comprehension: instruction execution, content comprehension, and speech recognition; Oral expression: vocabulary usage, sentence construction, and narrative ability; Pre-reading skills: literacy, story comprehension, and reading interest; Pre-writing preparation: writing awareness, graphic perception, fine motor control; Social sector (60 indicators not listed here) Self-awareness: identity awareness, emotional awareness, and ability awareness; Interpersonal skills: communication skills, teamwork ability, conflict resolution; Emotional management: emotion recognition, emotion expression, and emotion regulation; Rule awareness: understanding rules, abiding by rules, and creating rules; Scientific fields (70 indicators not listed here) Scientific inquiry: observation skills, questioning skills, and experimental thinking; Mathematical cognition: number concepts, spatial concepts, logical thinking; Natural cognition: natural phenomena, biological cognition, environmental protection; Logical thinking: reasoning ability, problem-solving, critical thinking; The arts field (50 indicators not listed here) Experience and Appreciation: Music appreciation, art appreciation, and natural aesthetics; Creative expression: musical expression, artistic creation, physical expression; Musical perception: sense of rhythm, sense of melody, and timbre discrimination; Artistic expression: use of color, composition ability, and creative expression.

[0037] Optionally, in general, parsing the speech text input through a semantic analysis module to obtain the speech content structure and determine the corresponding ability strength values ​​of children in different ability dimensions can be achieved in the following ways, without limitation: inputting the speech text into a pre-trained semantic understanding model and outputting the speech content structure; identifying multiple key feature words based on the speech content structure; classifying and mapping the multiple key feature words based on a preset ability dimension feature word list to obtain key feature word sets under different ability dimensions; calculating the keyword frequency and word importance score under each ability dimension based on the key feature word set, and determining the ability strength value corresponding to each ability dimension by weighting the keyword frequency and word importance score.

[0038] The following section will use a specific scenario to explain in detail how to use the semantic analysis module to parse the voice text input, obtain the voice content structure, and determine the corresponding ability strength values ​​of children in different ability dimensions.

[0039] Scene setting: Child: Xiaoming (5 years old); Interactive task: Describe the picture. The screen displays a picture of "a white rabbit eating a red carrot in the grass". Voice text input (obtained after speech recognition): "A little white rabbit is on the grass, eating a big red carrot."

[0040] Step 1: Input the speech text into the pre-trained semantic understanding model and output the speech content structure.

[0041] This step utilizes mature Natural Language Processing (NLP) techniques for structured analysis. The specific tasks performed by this "semantic understanding model" include: (1) Word segmentation: dividing a continuous sentence into independent words; Input: "A little white rabbit is on the grass, eating a big red carrot." Output: ["a", "little white rabbit", ",", "on", "grass", ",", "it", "on", "eating", "one", "big", "of", "carrot", "."]; Note: For clarity, "little white rabbit" and "red carrot" are treated as a whole here, but the actual word segmentation may be more detailed; (2) Part-of-speech tagging: Tagging the grammatical attributes of each word; Output: ["one" (quantifier), "little white rabbit" (noun), "on" (preposition), "on the grass" (noun), "it" (pronoun), "on" (adverb), "eat" (verb), "one" (quantifier), "big" (adjective), "of" (particle), "carrot" (noun)]; (3) Named entity recognition: Identify entities with specific meanings, such as animals, plants, colors, etc.; Output: {"Animal": ["White Rabbit"], "Plant": ["Grass", "Carrot"], "Color": ["White", "Red"], "Quantity": ["One", "One"]}; Note: The "little white rabbit" here is identified as an "animal" entity, and its color attribute "white" is also extracted; The structured information obtained through word segmentation, part-of-speech tagging, and named entity recognition collectively constitutes the "speech content structure".

[0042] Step 2: Based on the structure of the speech content, identify multiple key feature words.

[0043] This step extracts key feature words that carry core information from the speech content structure using preset rules. These preset rules may include: Extract all nouns; Extract all verbs; Extract all adjectives that represent attributes (such as color and size); Extract all words indicating quantity; Extract all identified named entities; In this example, the key feature words identified are: {"one", "little white rabbit", "on the grass", "eat", "one", "big", "carrot", "white", "red"}.

[0044] Step 3: Based on the preset capability dimension feature word list, classify and map multiple key feature words to obtain key feature word sets under different capability dimensions.

[0045] The core of this step is a publicly available "preset capability dimension feature word list," either manually constructed or statistically generated based on a corpus. This preset capability dimension feature word list maps key feature words to different capability dimensions:

[0046] Table 1. Preset Capability Dimension Feature Word List Important note: A word can belong to multiple dimensions. For example, "carrot" is both a food in the field of health and a plant in the field of science. This cross-disciplinary nature is a true reflection of children's comprehensive development. Matching the key feature words with the table above yields key feature word sets for different capability dimensions: Health: {"Eating", "Carrots"}; Language: {"one", "one", "big", "white", "red"}; Society: {“it”}; Science: {"Little rabbit", "grass", "carrot"}; Art: {"Big", "White", "Red"}.

[0047] Step 4: Based on the set of key feature words, calculate the keyword frequency and importance score for each capability dimension, and determine the capability strength value corresponding to each capability dimension by weighting the keyword frequency and importance score.

[0048] This step requires defining a "word importance score" (which needs to be adjusted and confirmed according to the five major domains, and is not listed here). The formula for calculating the capability strength value is: Capability Strength Value = Σ (Keyword Frequency × Word Importance Score)). Health Capacity Strength Value = (1 × Score ("Eating")) + (1 × Score ("Carrots")) = (1 × 1.0) + (1 × 1.8) = 2.8; Language proficiency strength score = (1 × score (“one”)) + (1 × score (“one”)) + (1 × score (“big”)) + (1 × score (“white”)) + (1 × score (“red”)) = (1 × 1.0) + (1 × 1.0) + (1 × 1.5) + (1 × 1.2) + (1 × 1.2) = 5.9; Social competence strength score = (1 × score ("it")) = 1 × 0.5 = 0.5; Science Ability Strength Value = (1 × Score ("White Rabbit")) + (1 × Score ("On the Grass")) + (1 × Score ("Carrot")) = (1 × 2.0) + (1 × 1.5) + (1 × 1.8) = 5.3; Artistic Ability Strength Value = (1 × Score ("Big")) + (1 × Score ("White")) + (1 × Score ("Red")) = (1 × 1.5) + (1 × 1.2) + (1 × 1.2) = 3.9.

[0049] Optionally, under normal circumstances, the acoustic signal data can be parsed by the acoustic analysis module to obtain acoustic expression patterns, and the intensity values ​​of the sensitive periods corresponding to different ability dimensions of children can be determined in the following ways, without limitation: Acoustic feature extraction is performed on the acoustic signal data to obtain an acoustic feature parameter set, which includes pitch, intensity, speech rate, and pauses; the acoustic feature parameter set is input into a pre-trained acoustic expression pattern classifier to obtain acoustic expression patterns, which include emotional activation, participation, and interest; and the acoustic expression patterns are converted into the corresponding sensitive period intensity values ​​for different ability dimensions using a preset sensitive period feature model for each ability dimension.

[0050] The following section will continue using the previous "Xiaoming's Picture Description" scenario to explain in detail how to analyze the acoustic signal data generated by Xiaoming and ultimately determine the corresponding sensitivity period intensity values ​​for each of the different capability dimensions.

[0051] Step 1: Extract acoustic features from the acoustic signal data to obtain a set of acoustic feature parameters, which includes pitch, intensity, speech rate, and pauses.

[0052] This step uses digital signal processing techniques to extract a series of quantifiable acoustic feature parameters from the acoustic signal data. The algorithms for extracting these acoustic feature parameters are all well-known techniques in the field of signal processing. 1. Pitch: Reflects the fundamental frequency of a sound; Extract parameters: Average fundamental frequency: the average pitch of the entire sentence; Fundamental frequency standard deviation: the degree of pitch variation, reflecting the fluctuation of intonation; Example values: Assuming the average fundamental frequency of a 5-year-old boy is 250Hz, the analysis shows that the average fundamental frequency of Xiaoming's speech is 280Hz (above average level), and the standard deviation is 40Hz (large fluctuation). 2. Sound intensity: Reflects the energy level of a sound; Extract parameters: Average energy: the average loudness of the entire sentence; Energy standard deviation: the magnitude of loudness variation; Example values: Assuming the normal speaking energy is 60dB, the analysis shows that Xiaoming's average speaking energy is 68dB (loud voice), and the energy standard deviation is relatively large; 3. Speech rate: Reflects how fast or slow one speaks; Extracted parameter: syllables per second; Example values: Assuming a child's normal speaking speed is 3.5 syllables / second, the analysis shows that Xiaoming's speaking speed is 4.5 syllables / second (speaking speed is too fast); 4. Pauses: Reflect the fluency of speech; Extract parameters: Total duration of silent segments: the sum of all silent parts in a sentence; Number of pauses: The number of times a silent segment appears in a sentence; Example values: Xiaoming's sentence lasted 5 seconds in total, with the silent segment lasting only 0.3 seconds and pausing once (at the two commas), indicating that his expression was very fluent; Acoustic characteristic parameter set (example output): {Average fundamental frequency: 280Hz, fundamental frequency standard deviation: 40Hz, average energy: 68dB, speech: 4.5syl / s, total duration of silence segment: 0.3s}.

[0053] Step 2: Input the acoustic feature parameter set into the pre-trained acoustic expression pattern classifier to obtain the acoustic expression pattern, which includes emotional activation, engagement and interest.

[0054] The "pre-trained acoustic expression pattern classifier" in this step is a well-defined, rule-based scoring system or weighted calculation model based on psychoacoustics and child development theory. Definition of acoustic expression mode: Emotional activation level: refers to the degree of excitement or calmness of a child's emotions. The score range is as follows: high score indicates excitement; low score indicates calmness or dullness. Engagement: This refers to the child's level of engagement and focus on the current task. The score range indicates that the child is focused and engaged, while the score indicates that the child is distracted and perfunctory. Interest level: A comprehensive measure of activation and participation, reflecting children's intrinsic liking and desire to explore the current content; score range. Calculation rules (example): First, the acoustic feature parameters extracted in step one are normalized to a range by comparing them with the preset baseline values ​​for children of the same age: Normalized average fundamental frequency = (280Hz - 250Hz) / (300Hz_Max - 250Hz_Min) = 0.6; Normalized fundamental frequency standard deviation = (40Hz - 20Hz) / (50Hz_Max - 20Hz_Min) = 0.67; Normalized average energy = (68dB - 60dB) / (75dB_Max - 60dB_Min) = 0.53; Normalized speech rate = (4.5 - 3.5) / (5.0_Max - 3.5_Min) = 0.67; Normalized fluency (calculated in reverse from pauses) = 1 - (0.3s / 5s) = 0.94; Then, the scores for the three expression modes are calculated using a linear weighting formula: 1. Emotional activation level (primarily determined by changes in pitch and intensity): Emotional activation level = 0.6 × 0.67 + 0.4 × 0.53 = 0.614; 2. Engagement (primarily determined by speaking speed and fluency): Participation rate = 0.5 × 0.67 + 0.5 × 0.94 = 0.805; 3. Interest level (a comprehensive reflection of activation and participation, with participation having a higher weight because sustained investment is a better indicator of genuine interest than momentary excitement): Interest level = 0.3 × Emotional activation level + 0.7 × Participation level = 0.3 × 0.614 + 0.7 × 0.805 = 0.7477; Acoustic expression pattern (example output): {Emotional activation: 0.61, Engagement: 0.81, Interest: 0.75}.

[0055] Step 3: By using a preset capability dimension sensitive period feature model, the acoustic expression pattern is converted into the corresponding sensitive period intensity value for each capability dimension.

[0056] The "Preset Ability Dimension Sensitive Period Feature Model" in this step is a context-based attention allocation mechanism. Its core logic is that the interest children show when expressing a certain topic should be mainly attributed to the ability dimension corresponding to that topic. Allocation rules: 1. Obtaining context weights: From the aforementioned "ability strength value" calculation steps, the content relevance of each ability dimension has been obtained. These ability strength values ​​can be used directly or normalized as context weights. Health Capacity Strength Value = 2.8; Language proficiency strength score = 5.9; Social competence strength value = 0.5; Scientific ability strength value = 5.3; Artistic Ability Strength Value = 3.9; Total strength value = 2.8 + 5.9 + 0.5 + 5.3 + 3.9 = 18.4; Context weight (capability dimension d) = Capability strength value of that capability dimension / Total strength value; W (healthy) = 2.8 / 18.4 ≈ 0.152; W (language) = 5.9 / 18.4 ≈ 0.321; W (social)=0.5 / 18.4≈0.027; W (science) = 5.3 / 18.4 ≈ 0.288; W (Art) = 3.9 / 18.4 ≈ 0.212; 2. Calculate the sensitivity period intensity value: Using a clear weighted allocation formula, the overall "interest level" calculated in step two is allocated to each ability dimension. The calculation formula is: Sensitive period intensity value (ability dimension d) = interest level × context weight (d) + basic sensitivity period value; Interest level: Here, 0.7477 is simplified to 0.75; Context weights (d): The weights of each capability dimension calculated in the previous step; Baseline sensitive period value: A pre-set small constant (e.g., 0.05) that represents a child's basic level of acceptance or potential sensitivity to any area, which is unrelated to the current topic. Detailed calculation process: The intensity value of the health sensitivity period = (0.75 × 0.152) + 0.05 = 0.114 + 0.05 = 0.164; The intensity value of the language sensitive period = (0.75 × 0.321) + 0.05 = 0.241 + 0.05 = 0.291; The intensity value of the socially sensitive period = (0.75 × 0.027) + 0.05 = 0.020 + 0.05 = 0.070; The intensity value of the scientific sensitivity period = (0.75 × 0.288) + 0.05 = 0.216 + 0.05 = 0.266; Art sensitivity period intensity value = (0.75 × 0.212) + 0.05 = 0.159 + 0.05 = 0.209.

[0057] The following is a specific example to illustrate the ability strength values ​​and sensitive period strength values ​​for a child (Xiaoya) in different ability dimensions:

[0058] Table 2 Competency Dimensions Health (ability strength value 45): This reflects Xiaoya's condition of "being physically healthy but needing to improve coordination". She is not weak or sickly, but large muscle control (such as balance, precise running and jumping) is not her strength. Language (ability strength value 35): This reflects Xiaoya's deficiency in "language expression" because she is introverted and does not like to speak proactively. Social (ability strength value 25): Xiaoya's weakest ability dimension, which directly reflects her shyness and social avoidance behavior; Science (ability strength value 60): Xiaoya is full of curiosity about nature and has accumulated some scattered knowledge through observation. Her ability is at a medium-to-high level. Art (ability strength value 85): This is Xiaoya's "strength" or talent. She often draws, and her skills have far exceeded the average level of her peers. Health (Sensitive period intensity value 60): Xiaoya likes running and jumping and has good intrinsic motivation, but it has not reached the point of obsession. Language (sensitive period intensity value 75): Xiaoya loves listening to stories and has a strong desire to absorb language; Social (Sensitive period intensity value 20): Xiaoya exhibits avoidance behavior, has very low intrinsic motivation, and is in a state of "lack of interest" or even "slight resistance"; Science (Sensitive period intensity value 95): This is Xiaoya's current greatest passion, showing extremely strong focus and desire to explore, almost reaching its peak; Art (Sensitive Period Intensity Value 90): Xiaoya's enthusiasm for drawing is second only to science. She actively engages in drawing every day and is in a state of "high immersion".

[0059] S103. Based on the ability intensity value and the sensitive period intensity value, calculate the intervention priority score corresponding to each child in different ability dimensions; determine the ability dimension with the highest intervention priority score as the target ability dimension that needs to be coordinated for intervention. The intervention priority score indicates the urgency of needing intervention for a particular capability dimension. The target capability dimension refers to the capability dimensions that are ultimately determined to require priority intervention.

[0060] Specifically, the computer device uses capability strength value and sensitive period strength value as two key indicators to comprehensively evaluate the intervention value of each capability dimension through a specific calculation model, i.e., the intervention priority score. This calculation model considers both capability development (the lower the capability strength value, the more urgent the intervention) and development timing (the higher the sensitive period strength value, the better the intervention effect). Based on the calculated intervention priority score, the computer device selects the capability dimension with the highest intervention priority score as the target capability dimension that needs to be focused on intervention at present. This ensures both the necessity of intervention and the grasp of the optimal timing for intervention.

[0061] Optionally, in general, based on the ability intensity value and the sensitive period intensity value, the intervention priority score corresponding to each child in different ability dimensions can be calculated in the following way, without limitation: Substitute the ability intensity value into the formula for calculating the urgency of the ability deficit to obtain the urgency of the child's ability deficit in different ability dimensions; map the sensitive period intensity value to the maturity of the timing window; calculate the intervention priority score corresponding to each child in different ability dimensions based on the urgency of the ability deficit and the maturity of the timing window; the formula for calculating the urgency of the ability deficit is: Where Ud represents the urgency of a child's skill gaps under skill dimension d, Ed (age) represents the expected skill strength of a child at their current age under skill dimension d, and Sd represents the skill strength of a child under skill dimension d.

[0062] The urgency of a skill gap reflects the degree to which it "needs to be addressed." The lower the skill strength value (compared to the age standard), the higher the urgency of the skill gap.

[0063] The maturity of the timing window reflects the degree of "suitability" for learning. The higher the sensitivity period intensity value, the higher the child's current acceptance, interest, and efficiency in this type of learning, and therefore the higher the maturity of the timing window.

[0064] The intervention priority score takes into account both of the above two factors, with the highest priority given to interventions for those ability dimensions that "both need supplementation and are suitable for learning".

[0065] The formula for calculating the urgency of capability shortcomings is: Where Ud represents the urgency of a child's skill gap under skill dimension d, Ed(age) represents the expected skill strength value of a child at the current age under skill dimension d, Sd represents the skill strength value of a child under skill dimension d, and max(0, ...) ensures that when a child's skill strength value exceeds the expected skill strength value, the urgency of the skill gap is 0, rather than a negative number.

[0066] Mapping the sensitivity period intensity value to the timing window maturity, we can make a very simple assumption here: the sensitivity period intensity value itself can be directly used as the timing window maturity.

[0067] It should be noted that the units of the capability strength value and the sensitive period strength value should be consistent, as should the units of the urgency of capability shortcomings and the maturity of the timing window. If they are inconsistent, normalization is required.

[0068] By combining the urgency of capability gaps with the maturity of the timing window, a weighted summation method can be used to obtain an intervention priority score, or other methods can be used, which are not limited here.

[0069] Following the example from step S102, assume that Xiaoya is 5 years old and the expected ability strength value Ed (age) for each ability dimension is 70 (simplified processing).

[0070] (1) Calculate the urgency of the capability gaps in each capability dimension: Health: Ud_health = max(0, (70-45) / 70) = 0.36; Language: Ud_language = max(0, (70-35) / 70) = 0.50; Society: Ud_society = max(0, (70-25) / 70) = 0.64; Science: Ud_science = max(0, (70-60) / 70) = 0.14; Art: Ud_Art = max(0, (70-85) / 70) = 0 (because 85>70); (2) Determine the maturity of the timing window for each capability dimension (directly use the intensity value of the sensitive period, divide by 100 for normalization): Health: 0.60; Language: 0.75; Social: 0.20; Science: 0.95; Art: 0.90; (3) Calculate the intervention priority score (using a weighted summation method, with each weight being 0.5): Intervention priority score = 0.5 × urgency of capability gaps + 0.5 × maturity of timing window; Health: 0.5 × 0.36 + 0.5 × 0.60 = 0.48; Language: 0.5 × 0.50 + 0.5 × 0.75 = 0.625; Social: 0.5 × 0.64 + 0.5 × 0.20 = 0.42; Science: 0.5 × 0.14 + 0.5 × 0.95 = 0.545; Art: 0.5×0 + 0.5×0.90 = 0.45; (4) Intervention priority scores after ranking (from high to low): Language (0.625) > Science (0.545) > Health (0.48) > Art (0.45) > Society (0.42); (5) Determine the dimensions of the target capabilities: Based on the calculation results, the language ability dimension had the highest intervention priority score (0.625), and was therefore identified as the target ability dimension that requires current collaborative intervention.

[0071] S104. Load the target ability development ladder model corresponding to the target ability dimension, combine the target ability intensity value corresponding to the target ability dimension, determine the target ability level that the child is currently at, and generate the target ability development ladder strategy. The target ability development ladder model includes different ability intensity value ranges corresponding to different ability ladders of the target ability dimension, different state performance descriptions, and different next ability development indicators. The target ability development ladder strategy refers to the improvement plan from the target ability level to the next ability level. The Target Ability Development Ladder Model is a phased model describing the development process of target ability dimensions, encompassing multiple developmental levels from basic to advanced. The target ability level refers to the child's current developmental stage in that target ability dimension. The ability intensity range refers to the threshold level of ability required to reach at the corresponding ability level. The state performance description is a written description of the typical behavioral characteristics and performance that the child should possess at the corresponding ability level. The next ability development indicator represents the specific goal requirements for advancing from the current ability level to the next level. The Target Ability Development Ladder Strategy refers to a specific improvement plan developed based on the current ability level, pointing towards the next ability level.

[0072] Specifically, after determining the target ability dimension, the computer device loads the corresponding target ability development ladder model from a pre-set model library. The target ability development ladder model defines in detail the development path of the target ability dimension: dividing the development process into multiple progressive levels, each level having a corresponding ability strength value range, a specific state description, and the development indicators (next ability development indicators) required to advance from the current level to the next level. The computer device compares the target ability strength value corresponding to the previously obtained target ability dimension with the ability strength value range of each ability ladder in the target ability development ladder model to determine the child's current target ability level. Then, based on the state of the target ability level and the requirements of the next ability level, the computer device generates a targeted target ability development ladder strategy, including key areas for ability improvement, specific goals, and expected achievement timelines.

[0073] Assuming the target competency dimension is language ability, the following table lists an example of a language ability development ladder model:

[0074] Table 3. Ladder Model for Target Capability Development S105. Integrate the target ability development ladder strategy with the child's personalized interest characteristics to obtain a content query vector. Search the preset intervention content library using the content query vector to obtain the intervention content set and push the intervention content set.

[0075] Personalized interest characteristics represent the unique preferences and interests exhibited by children during the learning process. Content query vectors refer to the feature representation after mathematically vectorizing the target ability development ladder strategy and personalized interest characteristics. The pre-set intervention content library represents a database storing various intervention content, which is labeled and categorized according to different dimensions and characteristics. The intervention content set refers to the combination of content that meets intervention needs, selected based on query conditions.

[0076] Specifically, to ensure that intervention content aligns with both developmental needs and children's individual interests, the computer device transforms the target ability development ladder strategy into a strategy feature vector. Simultaneously, it extracts personalized interest features (such as preferred game types and story themes) from children's historical interaction data and constructs an interest profile vector. The computer device then fuses these two vectors to obtain a content query vector that comprehensively considers both developmental needs and children's individual interests. Using this content query vector, the computer device performs similarity matching within a pre-defined intervention content library to select the most suitable set of intervention content. This set of intervention content effectively supports ability improvement while stimulating learning interest, ensuring the intervention process is both engaging and effective. After selection, the computer device delivers the intervention content set to children in an appropriate manner, potentially including games, stories, interactive exercises, and other formats.

[0077] Optionally, in general, the strategy for developing target abilities is integrated with the child's personalized interest characteristics to obtain a content query vector. The content query vector is then used to search a pre-defined intervention content library to obtain the intervention content set. This can be achieved in the following ways, without limitation: Encode the target ability development strategy into a strategy feature vector; extract the child's personalized interest characteristics from the child's historical interaction data to construct an interest profile vector; fuse the strategy feature vector and the interest profile vector to obtain a fused feature vector; perform dimensionality reduction on the fused feature vector to obtain the content query vector; calculate the cosine similarity between the content query vector and each content feature vector in the pre-defined intervention content library; and select the top N content feature vectors with the highest cosine similarity to obtain the intervention content set.

[0078] By employing the aforementioned technical solution, computer devices collect children's complex interactive behavioral information and transform it into multimodal data packets, achieving a comprehensive capture of children's ability status. This involves both precise location of ability strength values ​​across different ability dimensions through semantic analysis and in-depth mining of sensitive period strength values ​​for different ability dimensions through acoustic analysis, providing dual-dimensional data support for intervention decisions. Based on ability strength values ​​and sensitive period strength values, the computer devices determine the target ability dimension for which the child currently requires coordinated intervention, ensuring the targetedness and timeliness of the intervention and avoiding indiscriminate intervention. Subsequently, the computer devices load the target ability development ladder model corresponding to the target ability dimension, generating personalized target ability development ladder strategies for the child. Furthermore, they integrate the child's personalized interest characteristics to push a set of intervention content, ensuring a high degree of alignment between the intervention content and ability development goals while fully considering the child's willingness to accept it, significantly improving the attractiveness and participation of the intervention content, thereby greatly optimizing the intervention effect.

[0079] The following provides a more detailed description of the process of the method provided in this implementation. Please refer to [link / reference]. Figure 2 This is another flowchart illustrating the intelligent children's ability collaborative intervention method in the embodiments of this application.

[0080] S201. Collect complex interactive behavior information generated by children when responding to preset tasks in preset situations, perform speech recognition and acoustic analysis on the complex interactive behavior information to obtain multimodal data packets, which include speech text input and acoustic signal data.

[0081] For details, please refer to step S101, which will not be repeated here.

[0082] S202. The semantic analysis module parses the speech text input to obtain the speech content structure, so as to determine the corresponding ability intensity values ​​of children in different ability dimensions; the acoustic analysis module parses the acoustic signal data to obtain the acoustic expression pattern, so as to determine the corresponding sensitive period intensity values ​​of children in different ability dimensions.

[0083] For details, please refer to step S102, which will not be repeated here.

[0084] S203. Based on the ability intensity value and the sensitive period intensity value, calculate the intervention priority score corresponding to each child in different ability dimensions; determine the ability dimension with the highest intervention priority score as the target ability dimension that needs to be coordinated for intervention.

[0085] For details, please refer to step S103, which will not be repeated here.

[0086] S204. Obtain the child's historical intervention data, and based on the historical intervention data, calculate the historical intervention frequency and historical intervention effect score for the target ability dimension.

[0087] Historical intervention data refers to detailed records of various ability interventions received by children over a past period, including information such as the time, content, process, and effects of the interventions. Historical intervention frequency indicates the number and time density of interventions received for a specific ability dimension within a historical period. Historical intervention effectiveness score represents a quantitative assessment of the effectiveness of historical interventions, including a comprehensive score across multiple assessment dimensions such as the degree of ability improvement, participation, and completion quality.

[0088] Specifically, first, the computer device retrieves the child's historical intervention data from the database, extracting all historical intervention activity records related to the target ability dimension. Then, the computer device calculates the historical intervention frequency for the target ability dimension, specifically by statistically analyzing the number of historical interventions per unit of time, such as "average number of interventions per month." Simultaneously, the computer device scores the effectiveness of the historical intervention activities. The scoring process comprehensively considers multiple indicators: such as the difference in ability assessments before and after the intervention, participation data during the intervention, task completion rate, etc., and calculates the final historical intervention effectiveness score through weighted averages.

[0089] S205. When the historical intervention frequency is greater than the preset frequency threshold and the historical intervention effect score is lower than the preset first score threshold, the capability dimension with the second highest intervention priority score is updated to the target capability dimension that currently requires collaborative intervention.

[0090] The preset frequency threshold refers to the standard value for judging whether the intervention frequency of a certain ability dimension is too high, such as "4 times per month". The preset first scoring threshold refers to the standard value for judging whether the intervention effect of a certain ability dimension is ideal, such as "70 points". The ability dimension with the second highest intervention priority score indicates the ability dimension that ranked second in the previous intervention priority score calculation.

[0091] Specifically, after the computer device obtains historical intervention frequency and historical intervention effect scores, it performs a dual threshold judgment: comparing the historical intervention frequency with a preset frequency threshold, and simultaneously comparing the historical intervention effect scores with a preset first score threshold. If a situation of "high frequency, low effect" occurs (i.e., the historical intervention frequency is greater than the preset frequency threshold, and the historical intervention effect score is lower than the preset first score threshold), it indicates that continuing to invest intervention resources in that ability dimension may not be effective. In this case, the computer device will dynamically adjust the intervention strategy, determining the ability dimension with the second-highest intervention priority score as the new target ability dimension. This dynamic adjustment mechanism can avoid resource waste and ensure that intervention investment more effectively promotes children's all-round development.

[0092] S206. Load the target ability development ladder model corresponding to the target ability dimension, combine the target ability intensity value corresponding to the target ability dimension, determine the target ability level that the child is currently at, and generate the target ability development ladder strategy. The target ability development ladder model includes different ability intensity value ranges corresponding to different ability ladders of the target ability dimension, different state performance descriptions, and different next ability development indicators. The target ability development ladder strategy refers to the improvement plan from the target ability level to the next ability level.

[0093] For details, please refer to step S104, which will not be repeated here.

[0094] S207. Integrate the target ability development ladder strategy with the child's personalized interest characteristics to obtain a content query vector. Search the preset intervention content library using the content query vector to obtain the intervention content set and push the intervention content set.

[0095] For details, please refer to step S105, which will not be repeated here.

[0096] S208. Monitor children's interactive behavior data on the set of intervention content, and calculate the interaction effect score based on the interactive behavior data.

[0097] Interactive behavior data represents all interactive behavior records generated by children during the use of intervention content, including quantitative indicators such as usage duration, operation frequency, completion progress, and response speed. Interactive effectiveness score is a comprehensive score calculated based on interactive behavior data to assess the suitability of the intervention content, reflecting children's acceptance of the current intervention content and their enthusiasm for participation.

[0098] Specifically, the computer device collects interactive behavior data from multiple dimensions: recording the time children spend on each intervention activity, counting the number and frequency of interactive actions, tracking task completion progress, and analyzing the timeliness of action responses. Based on this raw interactive behavior data, the computer device uses a pre-set scoring model to calculate an interaction effectiveness score. The pre-set scoring model assigns different weights to different dimensions of behavioral data; for example, it may place greater emphasis on sustained focus time and task completion quality, while assigning lower weight to simple click counts, ultimately resulting in a comprehensive score (interaction effectiveness score) that objectively reflects the suitability of the intervention content.

[0099] S209. When the interaction effect score is lower than the preset second score threshold, update the content query vector, search again in the preset intervention content library, and push the re-matched intervention content set to the child.

[0100] The preset second scoring threshold represents the standard value for determining whether intervention content needs adjustment, used to promptly identify content suitability issues. Updating the content query vector refers to optimizing and adjusting the original content query vector based on current interaction feedback, which may include adjusting the weight of ability objectives or modifying interest feature parameters. The re-matched intervention content set represents a combination of content more suitable for the child's current state, retrieved and filtered after the content query vector adjustment.

[0101] Specifically, when the computer device detects that the interaction effect score is lower than the preset second scoring threshold, it indicates that the currently pushed intervention content set may be unsuitable and needs to be adjusted promptly. The computer device analyzes the specific reasons for the low interaction effect score, such as the content being too difficult, lacking interest, or not matching the child's interests. Based on the analysis results, the computer device adjusts the parameter configuration of the content query vector accordingly: if it is a difficulty issue, the ability target requirement is lowered; if it is an interest issue, the weight of interest features is strengthened. The computer device uses the adjusted content query vector to re-search in the preset intervention content library and select a more suitable content combination. The new intervention content set will fully incorporate previous interaction feedback, ensuring the ability improvement goals are met while better aligning with the child's current state and needs. The computer device pushes the new intervention content set to the child and continues to monitor, ensuring dynamic optimization and continuous improvement of the intervention process.

[0102] The computer device in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 3 This is a schematic diagram of the physical device structure of a computer device in an embodiment of this application.

[0103] It should be noted that, Figure 3 The structure of the computer device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0104] like Figure 3 As shown, the computer device includes a CPU 301, which can perform various appropriate actions and processes according to a program stored in the read-only memory ROM 302 or a program loaded from the storage section 308 into the random access memory RAM 303, such as performing the methods described in the above embodiments. The RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An I / O interface 305 is also connected to the bus 304.

[0105] The following components are connected to I / O interface 305: input section 306 including audio input devices, push-button switches, etc.; output section 307 including a liquid crystal display (LCD) and audio output devices, indicator lights, etc.; storage section 308 including a hard disk, etc.; and communication section 309 including a network interface card such as a LAN (Local Area Network) card, modem, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0106] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by CPU 301, it performs the various functions defined in the present invention.

[0107] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0108] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, program segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those shown in the drawings.

[0109] Specifically, the computer device in this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, it implements the intelligent children's ability collaborative intervention method provided in the above embodiment.

[0110] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the computer device described in the above embodiments; or it may exist independently and not assembled into the computer device. The storage medium carries one or more computer programs that, when executed by a processor of the computer device, cause the computer device to implement the intelligent child ability collaborative intervention method provided in the above embodiments.

[0111] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0112] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0113] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. An intelligent method for collaborative intervention of children's abilities, characterized in that, Applied to computer devices, the method includes: collecting complex interactive behavior information generated by a child responding to a preset task in a preset context; performing speech recognition and acoustic analysis on the complex interactive behavior information to obtain a multimodal data packet, wherein the multimodal data packet includes speech text input and acoustic signal data; parsing the speech text input through a semantic analysis module to obtain the speech content structure, thereby determining the ability intensity value corresponding to each of the child's different ability dimensions; parsing the acoustic signal data through an acoustic analysis module to obtain the acoustic expression pattern, thereby determining the sensitive period intensity value corresponding to each of the child's different ability dimensions; calculating the intervention priority score corresponding to each of the child's different ability dimensions based on the ability intensity value and the sensitive period intensity value; and determining the ability dimension with the highest intervention priority score. The system identifies the target ability dimension requiring collaborative intervention. It loads the target ability development ladder model corresponding to the target ability dimension, combines it with the target ability intensity value, determines the child's current target ability level, and generates a target ability development ladder strategy. The target ability development ladder model includes different ability intensity value ranges, different state performance descriptions, and different next-stage ability development indicators for different ability steps within the target ability dimension. The target ability development ladder strategy refers to an improvement plan from the target ability level to the next ability level. The system integrates the target ability development ladder strategy with the child's personalized interest characteristics to obtain a content query vector. This content query vector is then used to search a preset intervention content library to obtain an intervention content set, which is then pushed to the system. The step of parsing the acoustic signal data through an acoustic analysis module to obtain acoustic expression patterns, and determining the sensitive period intensity values ​​corresponding to different ability dimensions for the child, specifically includes: extracting acoustic features from the acoustic signal data to obtain an acoustic feature parameter set, which includes pitch, intensity, speech rate, and pauses; inputting the acoustic feature parameter set into a pre-trained acoustic expression pattern classifier to obtain the acoustic expression pattern, which includes emotional activation, engagement, and interest; and converting the acoustic expression pattern into sensitive period intensity values ​​corresponding to different ability dimensions through a preset ability dimension sensitive period feature model. The step of calculating the intervention priority score for each child across different ability dimensions based on the ability strength value and the sensitive period strength value specifically includes: substituting the ability strength value into the formula for calculating the urgency of ability shortcomings to obtain the urgency of the child's ability shortcomings across different ability dimensions; mapping the sensitive period strength value to the maturity of the timing window; and calculating the intervention priority score for each child across different ability dimensions based on the urgency of ability shortcomings and the maturity of the timing window. The formula for calculating the urgency of ability shortcomings is as follows: Wherein, Ud represents the urgency of the child's ability shortcomings under ability dimension d, Ed (age) represents the expected ability strength value of the child at the current age under ability dimension d, and Sd represents the ability strength value of the child under ability dimension d.

2. The method according to claim 1, characterized in that, The step of parsing the voice-text input through a semantic analysis module to obtain the voice content structure and determine the ability strength value corresponding to each ability dimension of the child specifically includes: inputting the voice-text input into a pre-trained semantic understanding model and outputting the voice content structure; identifying multiple key feature words based on the voice content structure; classifying and mapping the multiple key feature words based on a preset ability dimension feature word list to obtain a set of key feature words under different ability dimensions; calculating the keyword frequency and word importance score under each ability dimension based on the set of key feature words, and determining the ability strength value corresponding to each ability dimension by weighting the keyword frequency and word importance score.

3. The method according to claim 1, characterized in that, The process of fusing the target ability development ladder strategy with the child's personalized interest characteristics to obtain a content query vector, and then searching a preset intervention content library using the content query vector to obtain an intervention content set, specifically includes: encoding the target ability development ladder strategy into a strategy feature vector; extracting the child's personalized interest characteristics from the child's historical interaction data to construct an interest profile vector; fusing the strategy feature vector and the interest profile vector to obtain a fused feature vector; performing dimensionality reduction on the fused feature vector to obtain the content query vector; calculating the cosine similarity between the content query vector and each content feature vector in the preset intervention content library; and selecting the top N content feature vectors with the highest cosine similarity to obtain the intervention content set.

4. The method according to claim 1, characterized in that, After the steps of calculating the intervention priority score corresponding to the child in different ability dimensions based on the ability strength value and the sensitive period strength value, and determining the ability dimension with the highest intervention priority score as the target ability dimension that needs to be coordinated intervention, the method further includes: obtaining the child's historical intervention data, calculating the historical intervention frequency and historical intervention effect score of the target ability dimension based on the historical intervention data; when the historical intervention frequency is greater than a preset frequency threshold and the historical intervention effect score is lower than a preset first score threshold, updating the ability dimension with the second highest intervention priority score as the target ability dimension that needs to be coordinated intervention.

5. The method according to claim 1, characterized in that, After the steps of fusing the target ability development ladder strategy with the child's personalized interest characteristics to obtain a content query vector, searching in a preset intervention content library using the content query vector to obtain an intervention content set, and pushing the intervention content set, the method further includes: monitoring the child's interaction behavior data with the intervention content set, calculating an interaction effect score based on the interaction behavior data; when the interaction effect score is lower than a preset second score threshold, updating the content query vector, re-searching in the preset intervention content library, and pushing the re-matched intervention content set to the child.

6. A computer device, characterized in that, The computer device includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the computer device to perform the method as described in any one of claims 1 to 5.

7. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are executed on a computer device, the computer device performs the method as described in any one of claims 1 to 5.

8. A computer program product, characterized in that, When the computer program product is run on a computer device, the computer device performs the method as described in any one of claims 1 to 5.

Images