AI calling operation rule model based on preschool education key indexes

Abstract

The invention discloses a model for calling an operation rule based on a key index AI of preschool education, and relates to the technical field of preschool education. Multi-source data of preschool education is collected by constructing a cloud big data center, and the multi-source data is processed to obtain a data group with an association relationship; a preschool education AI model is constructed and comprises a judgment model, a child development evaluation and diagnosis model, a personalized gamification learning content recommendation model and a family cooperative rearing interaction support model, input of the preschool education AI model is analyzed through the judgment model, and what type of model is adopted for key index output is judged. And meanwhile, a dynamic data pipeline is constructed to connect the cloud big data center and the preschool education AI, and a detection window is arranged to protect data security. Through the cloud big data center, the preschool education AI model and the dynamic data pipeline, more intelligent and accurate preschool education intervention is realized.

Description

Technical Field

[0001] The present invention relates to the field of preschool education technology, and in particular to an AI-based operation rule model for key preschool education indicators. Background Art

[0002] At present, the assessment and intervention of children's development in the field of preschool education mainly rely on manual observation, recording and analysis, which is inefficient and highly subjective. Traditional data collection methods are scattered and incomplete, making it difficult to integrate data on individual children, teaching environments, families and communities, and physical environments, resulting in an insufficient understanding of children's development. In addition, existing data analysis methods are usually relatively simple and difficult to mine the complex relationships and patterns hidden in the data, making it impossible to provide effective decision-making support for personalized education. More importantly, data security and privacy protection face severe challenges during data transmission and use. Traditional data transmission methods are vulnerable to attacks and tampering, lack effective permission control and security protection mechanisms, and pose a risk of data leakage;

[0003] Therefore, it is of great significance to construct an AI-based operation rule model based on key indicators of preschool education. Summary of the Invention

[0004] The purpose of the present invention is to provide an AI-based operation rule model for key indicators of preschool education to address the shortcomings of the background technology.

[0005] To achieve the above objectives, the present invention provides the following technical solutions: Based on the AI call operation rule model of key indicators of preschool education, it includes: building a cloud-based big data center to collect multi-source data on preschool education;

[0006] Build a preschool education AI model and define calling rules;

[0007] Build a dynamic data pipeline through which the cloud big data center and the preschool education AI model interact.

[0008] In a preferred embodiment, the steps of building a cloud-based big data center and collecting multi-source data on preschool education are as follows:

[0009] Collect multi-source data on preschool education and save it in the cloud big data center;

[0010] Multi-source data include individual child data, teaching environment data, family and community data, and physical environment data;

[0011] Feature extraction is performed on multi-source data to obtain mapping indicators, correlation analysis is performed on the mapping indicators to obtain a correlation matrix, and data clusters are constructed based on the correlation matrix.

[0012] In a preferred embodiment, the steps of extracting features from multi-source data to obtain mapping metrics, performing correlation analysis on the mapping metrics to obtain a correlation matrix, and constructing a data cluster based on the correlation matrix are as follows:

[0013] Use the analysis of variance method and the principal component analysis method to extract features from multi-source data to obtain mapping metrics;

[0014] Define a set M of mapping metrics i (r1, r2,..., r n ) and calculate the correlation coefficients between pairwise mapping metrics, and construct a correlation matrix R according to the correlation coefficients:

[0015]

[0016] where R represents the correlation matrix and r nn represents the correlation coefficient between mapping metrics;

[0017] Preset a coefficient threshold, form an index atlas with the mapping metrics whose correlation coefficients exceed the coefficient threshold, and connect the multi-source data pointed to by the index atlas to generate a data cluster;

[0018] The data cluster includes the multi-source data pointed to by the index atlas and the directed edges connecting them, and the directed edges are composed of the values of the correlation coefficients.

[0019] In a preferred embodiment, the steps of constructing a preschool education AI model and defining a call rule are as follows:

[0020] Constructing a preschool education AI model includes a judgment model, a child development assessment and diagnosis model, a personalized gamified learning content recommendation model, and a home-school co-education interaction support model;

[0021] Among them, a preset key index table is set for the child development assessment and diagnosis model, the personalized gamified learning content recommendation model, and the home-school co-education interaction support model;

[0022] Based on the judgment model, process the input of the preschool education AI model to generate key indicators and call requests.

[0023] In a preferred embodiment, the steps of processing the input of the preschool education AI model based on the judgment model to generate key indicators and call requests are as follows:

[0024] The judgment module receives the input of the preschool education AI, and extracts keywords through the semantic segmentation algorithm;

[0025] Use the semantic similarity algorithm to judge the similarity between the keywords and the key indicators in the key indicator table, record the similarity values and sort them to obtain a similarity ranking table;

[0026] Set a preset similarity threshold, select keywords exceeding the similarity threshold as similarity indicators according to the similarity ranking table, and record the key indicators pointed to by the similarity indicators;

[0027] Compare the key indicators based on the key indicators pointed to by the similarity indicators with the key indicator table, send a call application to the model to which the key indicator table belongs, and use the call application and the key indicators as call data.

[0028] In a preferred embodiment, the steps of constructing the dynamic data pipeline connecting the cloud big data center and the preschool education AI model are as follows:

[0029] Set data ports on both the cloud big data center and the preschool education AI model, and connect the data ports of the cloud big data center and the preschool education AI model through the dynamic data pipeline;

[0030] Set a comparison unit in the data port of the cloud big data center. The comparison unit is used to receive the call data sent by the preschool education AI model, including the call application and the key indicators, and perform semantic analysis on the key indicators to retrieve similar mapping indicators to obtain corresponding data keys;

[0031] The data key is used to access the index atlas and call the corresponding data group;

[0032] Transmit the called data group as reply data to the preschool education AI model through the dynamic data pipeline, and set a detection window on the dynamic data pipeline. The detection window is used to confirm the security of the data group.

[0033] In a preferred embodiment, the steps of transmitting the called data group as reply data to the preschool education AI model through the dynamic data pipeline and setting a detection window on the dynamic data pipeline are as follows:

[0034] Store the reply data in the comparison unit, use the comparison unit to encapsulate the reply data and assign an audit watermark layer, and use the audit watermark layer to wrap the encapsulated reply data to obtain an encapsulated reply data;

[0035] Transmit the encapsulated reply data through the dynamic data pipeline, and open multiple detection windows on the dynamic data pipeline;

[0036] When the encapsulated reply data passes through the detection window, a unique ID scratch of the detection window will be engraved on the audit watermark layer, and all detection window ID scratches engraved on the audit watermark layer will be identified;

[0037] If it is identified that there are scratches on the audit watermark layer that do not belong to the detection window of the dynamic data pipeline where they are located, the detection window will destroy the encapsulated reply data, generate a detection log including the scratch information, and send a retransmission application to the cloud big data center.

[0038] In a preferred embodiment, a system for AI call operation rules based on key indicators of preschool education is constructed, including:

[0039] Cloud server module: Construct a cloud big data center to collect multi-source data of preschool education;

[0040] Preschool education AI module: Construct a preschool education AI model and define call rules;

[0041] Data security module: Construct a dynamic data pipeline through which the cloud big data center and the preschool education AI model interact.

[0042] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0043] 1. By constructing a cloud big data center, the present invention can collect and integrate preschool education data from multiple sources. These multi-source data include children's individual data such as health and development assessment, teaching environment data such as teacher-student interaction and teaching resources, family and community data such as family background and community support, and physical environment data such as classroom facilities and outdoor activity venues. Integrating these diverse data breaks the situation of traditional educational data silos and makes it possible to comprehensively understand the development status of young children. In the solution, variance analysis method and principal component analysis method are used to extract features from multi-source data to obtain mapping indicators, and an association matrix and data clusters are constructed through correlation analysis. This deep feature extraction and association analysis can reveal patterns and rules hidden in complex data. For example, analyzing the relationship between family economic status, parents' education level and children's language development. Through this deep analysis, educators can more accurately identify key factors affecting children's development and provide a scientific basis for formulating personalized education programs. In addition, the construction of data clusters also helps to quickly retrieve and call relevant data, improving data utilization efficiency;

[0044] 2. By constructing a preschool education AI model and judging the call rule mechanism of the model, the present invention can significantly improve the accuracy and efficiency of the call of the preschool education AI model. The judging model extracts keywords input to the preschool education AI through semantic segmentation algorithm and matches these keywords with the indicators in the preset key indicator table using semantic similarity algorithm. By setting a similarity threshold, the system can accurately select the key indicators most relevant to the input and send a call application to the corresponding AI model accordingly. This method avoids triggering all AI models indiscriminately, thus greatly reducing unnecessary consumption of computing resources. The process of key indicator extraction and similarity judgment ensures that only the AI model most suitable for processing the current input will be called;

[0045] 3. The present invention constructs a dynamic data pipeline to connect the cloud big data center and the preschool education AI model, realizing secure and reliable data transmission. During the data transmission process, by setting a comparison unit in the data port of the cloud big data center, semantic analysis is performed on key indicators to retrieve similar mapping indicators and obtain corresponding data keys, thereby realizing permission control for data access. Further, to ensure the security of data during transmission, the solution also adopts technical means such as data encapsulation, an audit watermark layer, and detection windows. Specifically, the reply data is stored in the comparison unit, and the comparison unit encapsulates the reply data and assigns an audit watermark layer. Multiple detection windows are opened on the dynamic data pipeline. By engraving the unique ID scratches of the detection windows on the audit watermark layer, it can be identified whether the data has been tampered with or leaked. If the detection window identifies scratches that do not belong to the detection window of the dynamic data pipeline where it is located, the encapsulated reply data is immediately destroyed, and a detection log and a retransmission application are generated. This multi-level security protection mechanism effectively guarantees the security and privacy of data, providing strong support for the popularization and application of preschool education AI. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0047] Figure 1 It is a flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0049] Example 1, please refer to Figure 1 As shown, the AI call operation rule model based on key indicators of preschool education in this embodiment includes:

[0050] S1. Construct a cloud big data center to collect multi-source data of preschool education;

[0051] S2. Construct a preschool education AI model and define call rules;

[0052] S3. Build a dynamic data pipeline through which the cloud big data center and the preschool education AI model interact.

[0053] As described in the above steps S1 - S3, currently, the assessment and intervention of children's development status in the field of preschool education mainly rely on manual observation, recording, and analysis, which are inefficient and highly subjective. Traditional data collection methods are scattered and incomplete, making it difficult to integrate data from multiple aspects such as children's individuals, teaching environments, families and communities, and physical environments, resulting in a lack of in-depth understanding of children's development status. In addition, existing data analysis methods are usually relatively simple, making it difficult to uncover complex relationships and patterns hidden in the data and unable to provide effective decision-making support for personalized education. More importantly, in the process of data transmission and use, data security and privacy protection face severe challenges. Traditional data transmission methods are vulnerable to attacks and tampering, lacking effective permission control and security protection mechanisms, and there is a risk of data leakage;

[0054] By building a cloud big data center, the present invention can collect and integrate preschool education data from multiple sources. These multi-source data include children's individual data such as health and development assessment, teaching environment data such as teacher-student interaction and teaching resources, family and community data such as family background and community support, and physical environment data such as classroom facilities and outdoor activity venues. Integrating these diverse data breaks the situation of traditional education data silos and makes it possible to comprehensively understand children's development status. In the solution, variance analysis and principal component analysis are used to extract features from multi-source data to obtain mapping indicators, and an association matrix and data clusters are constructed through correlation analysis. This deep feature extraction and association analysis can reveal patterns and laws hidden in complex data. For example, analyze the relationship between family economic status, parental education level, and children's language development. Through this in-depth analysis, educators can more accurately identify key factors affecting children's development and provide a scientific basis for formulating personalized education programs. In addition, the construction of data clusters also helps to quickly retrieve and call relevant data, improving data utilization efficiency;

[0055] By building a preschool education AI model and determining the call rule mechanism of the model, the accuracy and efficiency of calling the preschool education AI model can be significantly improved. The judgment model extracts keywords input to the preschool education AI through semantic segmentation algorithms and matches these keywords with the indicators in the preset key index table using semantic similarity algorithms. By setting a similarity threshold, the system can accurately select the key indicators most relevant to the input and send a call application to the corresponding AI model accordingly. This method avoids indiscriminate triggering of all AI models, thus greatly reducing unnecessary consumption of computing resources. The process of extracting key indicators and judging similarity ensures that only the AI model most suitable for processing the current input will be called;

[0056] By constructing a dynamic data pipeline to connect the cloud big data center and the preschool education AI model, secure and reliable data transmission is achieved. During data transmission, by setting a comparison unit in the data port of the cloud big data center, semantic analysis is performed on key indicators to retrieve similar mapping indicators and obtain corresponding data keys, thereby realizing access control of data. Furthermore, to ensure the security of data during transmission, the solution also adopts technical means such as data encapsulation, audit watermark layer, and detection window. Specifically, the reply data is stored in the comparison unit, and the comparison unit is used to encapsulate the reply data and assign an audit watermark layer. Multiple detection windows are opened on the dynamic data pipeline. By engraving the unique ID scratch of the detection window on the audit watermark layer, it can be identified whether the data has been tampered with or leaked. If the detection window identifies a scratch that does not belong to the detection window of the corresponding dynamic data pipeline, the encapsulated reply data is immediately destroyed, and a detection log and a retransmission application are generated. This multi-level security protection mechanism effectively ensures the security and privacy of data, providing strong support for the popularization and application of preschool education AI.

[0057] In one embodiment, step S1 of constructing the cloud big data center and collecting multi-source data of preschool education includes:

[0058] S11. Collect multi-source data of preschool education and save it to the cloud big data center;

[0059] S12. The multi-source data includes children's individual data, teaching environment data, family and community data, and physical environment data;

[0060] S13. Extract features from the multi-source data to obtain mapping indicators, perform correlation analysis on the mapping indicators to obtain a correlation matrix, and construct a data cluster based on the correlation matrix;

[0061] As described in the above steps S11 - S13, the technical solution for constructing a cloud - based big data center to collect multi - source data in preschool education first involves data collection. Data needs to be collected from multiple sources, including children's individual data, teaching environment data, family and community data, and physical environment data, and securely stored in the cloud - based big data center. Among them, children's individual data includes children's age, gender, development assessment results, health records, etc.; teaching environment data includes class size, teacher - student ratio, teacher qualification, teaching methods, curriculum settings, etc.; family and community data includes family income, parental education level, community resources, parent participation, etc.; physical environment data includes classroom area, lighting, noise level, safety facilities, etc. Subsequently, through feature extraction, key features are extracted from these multi - source data and mapped into quantifiable indicators. Then, correlation analysis is carried out to determine the associations between these indicators, construct an association matrix, and based on this, form data clusters to organize related data together for subsequent analysis and application.

[0062] In one embodiment, step S13 of performing feature extraction on multi - source data to obtain mapped indicators, performing correlation analysis on the mapped indicators to obtain an association matrix, and constructing a data cluster based on the association matrix includes:

[0063] S131. Using the analysis of variance method and the principal component analysis method to perform feature extraction on multi - source data to obtain mapped indicators;

[0064] S132. Define a set M of mapped indicators i (r1,r2,...,r n ) and calculate the correlation coefficients between pairwise mapped indicators, and construct an association matrix R according to the correlation coefficients:

[0065]

[0066] S133. Among them, R represents the association matrix, and r nn represents the correlation coefficient between mapped indicators;

[0067] S134. Preset a coefficient threshold, form an index atlas with the mapped indicators whose correlation coefficients exceed the coefficient threshold, and connect the multi - source data pointed to by the index atlas to generate a data cluster;

[0068] S135. The data cluster includes the multi - source data pointed to by the index atlas and the directed edges connecting them, and the directed edges are composed of the values of the correlation coefficients.

[0069] As described in the above steps S131 - S135, the aim is to extract key information from multi - source data and organize it into data clusters that are easy to understand and analyze. This process first uses analysis of variance (ANOVA) and principal component analysis (PCA) for dimensionality reduction and feature selection of multi - source data. ANOVA is used to identify whether there are significant differences between different categories of data and extract discriminative features. PCA, through linear transformation, converts high - dimensional data into a few principal components, retaining the main information of the data, thus obtaining more representative mapping indicators. Subsequently, a set of mapping indicators is defined, and the Pearson correlation coefficient between each pair of indicators in the set is calculated. This coefficient reflects the degree of linear correlation between the indicators, with a value range from - 1 to 1. Based on these correlation coefficients, an association matrix is constructed. To screen out indicators with strong correlations, a coefficient threshold is preset, and the mapping indicators whose absolute value of the correlation coefficient exceeds this threshold are classified into an indicator atlas. The indicator atlas can be understood as a set of key indicators. Finally, according to the indicator atlas, the multi - source data pointed to by the indicators in the atlas are connected by directed edges to form data clusters. Each data cluster contains the multi - source data pointed to by the indicator atlas and the directed edges connecting these data. The weight of the directed edge is composed of the value of the correlation coefficient, indicating the association strength and direction between the data. Such a data cluster structure can clearly display the internal relationships between multi - source data, facilitating subsequent data analysis and mining.

[0070] In one embodiment, step S2 of constructing the preschool education AI model and defining the calling rules includes:

[0071] S21. Constructing the preschool education AI model includes a judgment model, a child development assessment and diagnosis model, a personalized gamified learning content recommendation model, and a home - school co - education interaction support model;

[0072] S22. Among them, a preset key indicator table is set for the child development assessment and diagnosis model, the personalized gamified learning content recommendation model, and the home - school co - education interaction support model;

[0073] S23. Based on the judgment model, the input of the preschool education AI model is processed to generate key indicators and calling applications.

[0074] As described in the above steps S21 - S23, the aim is to build four key AI models to support a comprehensive preschool education experience. This process includes building a judgment model, a child development assessment and diagnosis model, a personalized gamified learning content recommendation model, and a home - school co - education interaction support model. This model is the "command center" of the entire AI system, responsible for receiving inputs, analyzing situations, and deciding which sub - model or sub - models to call. Among them, the child development assessment and diagnosis model is used to comprehensively analyze children's individual data, including development assessment, learning behavior, health status, etc., and combine teaching environment, family, and community data to evaluate children's cognitive, language, social - emotional, and motor development levels, and diagnose potential development problems. The implementation method selects Bayesian networks to establish a dependency relationship model between different fields of child development for comprehensive assessment and diagnosis; the personalized gamified learning content recommendation model can recommend personalized gamified learning content based on children's development assessment results, interest preferences, learning behavior data, combined with the teaching environment and family resources, to stimulate children's learning interest and promote their all - round development. The implementation method selects a reinforcement learning network to continuously optimize the recommendation strategy through interaction with children to improve learning effects. The home - school co - education interaction support model can provide personalized parenting advice and interaction support for parents based on children's development assessment results, learning situations, family environments, and community resources to promote home - school co - education and jointly promote children's healthy growth. The implementation method selects natural language generation to generate personalized parenting advice and activity recommendations according to children's individual situations; at the same time, the three models are used as sub - models, and a key indicator table is constructed for each sub - model. The key indicator table is used to reflect the functions of the sub - models and serve as the judgment criteria for the judgment model. For example, the key indicator table of the child development assessment and diagnosis model can include cognitive development level, language development level, motor development level, etc., the key indicator table of the personalized gamified learning content recommendation model can include interest areas, preferences for gamified elements, and learning style preferences, etc., and the key indicator table of the home - school co - education interaction support model can include parenting styles, quality of parent - child interaction, and degree of family environment support, etc. Finally, the judgment model processes the inputs of the preschool education AI and calls the most suitable sub - model for processing by comparing the key indicator table.

[0075] In one embodiment, step S23 of the judgment model processing the input of the preschool education AI model to generate key indicators and call requests includes:

[0076] S231. The judgment module receives the input of the preschool education AI and extracts keywords through a semantic segmentation algorithm;

[0077] S232. Use a semantic similarity algorithm to judge the similarity between the keywords and the key indicators in the key indicator table, record the similarity values and sort them to obtain a similarity ranking table;

[0078] S233. A preset similarity threshold. Select keywords exceeding the similarity threshold as similarity indicators according to the similarity ranking table, and record the key indicators pointed to by the similarity indicators.

[0079] S234. Compare the key indicators pointed to by the similarity indicators with the key indicator table, send a call application to the model to which the key indicator table belongs, and use the call application and the key indicators as call data.

[0080] As described in the above steps S231 - S234, the judgment module needs to be able to receive inputs from multiple sources, including text data such as teachers' observation records, parents' feedback, children's voice inputs, etc., and structured data including children's age, gender, learning records, etc. The semantic segmentation algorithm is used to identify keywords in the input text. The semantic segmentation algorithm can choose the Transformer model, such as BERT, GPT, etc., which can better understand the semantic information of the text. Then, the semantic similarity algorithm is used to judge the similarity between the keywords and the key indicators in the key indicator table. The semantic similarity algorithm is used to measure the similarity degree between the extracted keywords and the key indicators in the key indicator table. Commonly used semantic similarity algorithms can be selected, including Word2Vec, GloVe, and BERT. The key indicator table contains predefined key indicators and their descriptions. These key indicators reflect the functions and concerns of the preschool education AI model. For each extracted keyword, calculate its semantic similarity with all key indicators in the key indicator table, sort the calculated similarity values to generate a similarity ranking table. The similarity ranking table contains the correspondence between keywords and key indicators, as well as their similarity values. Preset a similarity threshold for screening out keywords that are similar enough to the key indicators. The setting of the similarity threshold needs to be adjusted according to the actual situation. Too high a threshold may result in failure to identify effective similarity indicators, and too low a threshold may lead to the introduction of noise. Further, a dynamic threshold can be considered, such as adjusting the threshold according to factors such as the frequency of keywords and the weights of key indicators. According to the similarity ranking table, select keywords with similarity values exceeding the threshold as similarity indicators and record the key indicators pointed to by each similarity indicator. According to the key indicators pointed to by the similarity indicators, send a call application to the model to which the key indicator table belongs. Further, if a similarity indicator points to multiple key indicators and these key indicators belong to different models, a certain strategy needs to be used to select which model to call. The priority can be set for different models, and the model with a higher priority is called first. It is also possible to vote according to the number of similarity indicators pointing to different models and select the model with the most votes. It is also possible to call multiple models simultaneously and fuse their outputs. Use the call application and the key indicators as call data and pass them to the corresponding sub - models.

[0081] In one embodiment, step S3 of constructing a dynamic data pipeline to connect the cloud big data center and the preschool education AI model includes:

[0082] S31. Set data ports on both the cloud big data center and the preschool education AI model, and connect the data ports of the cloud big data center and the preschool education AI model through a dynamic data pipeline;

[0083] S32. Set a comparison unit in the data port of the cloud big data center. The comparison unit is used to receive the call data sent by the preschool education AI model, including a call application and key metrics, and perform semantic analysis on the key metrics to retrieve similar mapped metrics and obtain corresponding data keys;

[0084] S33. The data key is used to access the metric atlas and call the corresponding data group;

[0085] S34. Transmit the called data group as reply data to the preschool education AI model through the dynamic data pipeline, and set a detection window on the dynamic data pipeline. The detection window is used to confirm the security of the data group;

[0086] As described in the above steps S31 - S34, to construct a dynamic data pipeline to connect the cloud big data center and the preschool education AI model, first set data ports on the cloud big data center and the AI model respectively, and connect them through a dynamic data pipeline. Set a comparison unit in the data port of the big data center. This unit receives the call data from the AI model, including a call application and key metrics, and performs semantic analysis on the key metrics to retrieve similar mapped metrics, thereby obtaining corresponding data keys. The data key is a credential for accessing a specific data group. The data key contains access permission information to ensure that only authorized users can access the data. Associate the data key with the mapped metrics to ensure that only models related to the key metrics can access the corresponding data. Use the data key to access the metric atlas, verify the validity of the key, and call the corresponding data group. Finally, transmit the called data group as reply data to the preschool education AI model via the dynamic data pipeline. At the same time, set a detection window on the data pipeline to ensure the security of the data group during transmission. This process realizes the dynamic and secure access of the AI model to the data in the big data center.

[0087] In one embodiment, step S34 of transmitting the called data group as reply data to the preschool education AI model through the dynamic data pipeline and setting a detection window on the dynamic data pipeline includes:

[0088] S341. Store the reply data in the comparison unit, use the comparison unit to encapsulate the reply data and assign an audit watermark layer, and use the audit watermark layer to wrap the encapsulated reply data to obtain an encapsulated reply data;

[0089] S342. Transmit the encapsulated reply data through a dynamic data pipeline, and open multiple detection windows on the dynamic data pipeline;

[0090] S343. When the encapsulated reply data passes through a detection window, a unique ID scratch of the detection window will be engraved on the audit watermark layer, and all the detection window ID scratches engraved on the audit watermark layer will be identified;

[0091] S344. If scratches that do not belong to the detection windows of the dynamic data pipeline are identified on the audit watermark layer, the detection window will destroy the encapsulated reply data, generate a detection log including the scratch information and send a retransmission request to the cloud big data center;

[0092] As described in the above steps S341 - S344, in order to ensure the secure transmission of data, first, the data group called from the cloud big data center is stored as reply data in the comparison unit. The comparison unit encapsulates these reply data and assigns an audit watermark layer, thus forming the encapsulated reply data. The principle of the audit watermark layer is to embed a specific digital signal into the reply data using a specific watermark algorithm. The receiving party can verify the integrity and source of the data by detecting this signal. Subsequently, the encapsulated reply data is transmitted through a dynamic data pipeline. Multiple detection windows are pre - opened on this pipeline and a unique ID is set for each detection window. When the encapsulated reply data passes through any detection window, this window will engrave its unique ID scratch on the audit watermark layer and simultaneously identify all the existing detection window ID scratches on the audit watermark layer. The detection window uses the watermark algorithm to detect the audit watermark layer, encrypts its own unique ID using an encryption algorithm and then embeds it into the audit watermark layer as an ID scratch using the watermark algorithm. If the detection window identifies that there are ID scratches on the audit watermark layer that do not belong to the detection windows of this dynamic data pipeline, it will immediately determine that the data may have been tampered with or leaked. The detection window will immediately destroy the encapsulated reply data and generate a detailed detection log, including all the identified scratch information. At the same time, this detection window will also send a retransmission request to the cloud big data center to ensure that the AI model can receive complete and secure data, thereby ensuring the reliability and security of the entire system. Through this multiple - detection mechanism, data can be effectively prevented from being illegally tampered with or stolen during transmission, and the overall security protection ability of the system can be improved.

[0093] In one embodiment, the system for constructing an AI call operation rule model based on key indicators of preschool education includes:

[0094] Cloud server module: Construct a cloud big data center to collect multi - source data of preschool education;

[0095] Preschool education AI module: Construct a preschool education AI model and define call rules;

[0096] Data security module: Build a dynamic data pipeline through which the cloud big data center and the preschool education AI model interact.

[0097] The above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. An AI call operation rule model based on key indicators of preschool education, characterized in that: Construct a cloud big data center to collect multi-source data of preschool education; Construct a preschool education AI model and define call rules; Construct a dynamic data pipeline through which the cloud big data center and the preschool education AI model interact.

2. The AI call operation rule model based on the key indicators of preschool education according to claim 1, wherein: The steps of constructing the cloud big data center to collect multi-source data of preschool education are as follows: Collect multi-source data of preschool education and save it to the cloud big data center; The multi-source data includes children's individual data, teaching environment data, family and community data, and physical environment data; Extract features from the multi-source data to obtain mapping indicators, perform correlation analysis on the mapping indicators to obtain a correlation matrix, and construct a data cluster based on the correlation matrix.

3. The operation rule model for AI call based on key indicators of preschool education according to claim 2, wherein: The steps of extracting features from the multi-source data to obtain mapping indicators, performing correlation analysis on the mapping indicators to obtain a correlation matrix, and constructing a data cluster based on the correlation matrix are as follows: Use the analysis of variance method and the principal component analysis method to extract features from the multi-source data to obtain mapping indicators; Define the mapping index set M i (r1, r2,..., r n ), calculate the correlation coefficient between pairwise mapping indices, and construct the correlation matrix R based on the correlation coefficient: Among them, R represents the correlation matrix, and r nn represents the correlation coefficient between mapping indicators; Preset a coefficient threshold, form an index atlas with mapping indicators whose correlation coefficients exceed the coefficient threshold, and connect the multi-source data pointed to by the index atlas to generate a data cluster; The data cluster includes the multi-source data pointed to by the index atlas and the directed edges connecting them, and the directed edges are composed of the values of the correlation coefficients.

4. The AI call operation rule model based on the key indicators of preschool education according to claim 1, wherein: The steps of constructing the preschool education AI model and defining call rules are as follows: The constructed preschool education AI model includes a judgment model, a children's development assessment and diagnosis model, a personalized gamified learning content recommendation model, and a home-school co-education interaction support model; Among them, a preset key indicator table is set for the children's development assessment and diagnosis model, the personalized gamified learning content recommendation model, and the home-school co-education interaction support model; Based on the judgment model, process the input of the preschool education AI model to generate key indicators and call requests.

5. The operation rule model based on AI call of key indicators of preschool education according to claim 4, wherein: The steps of processing the input of the preschool education AI model based on the judgment model to generate key indicators and call requests are as follows: The judgment module receives the input of the preschool education AI, and extracts keywords through a semantic segmentation algorithm; Use the semantic similarity algorithm to judge the similarity between the keywords and the key indicators in the key indicator table, record the values of the similarity and sort them to obtain a similarity ranking table; Preset a similarity threshold, select keywords exceeding the similarity threshold as similar indicators according to the similarity ranking table, and record the key indicators pointed to by the similar indicators; Compare the key indicators pointed to by the similar indicators with the key indicator table, send a call request to the model to which the key indicator table belongs, and use the call request and the key indicators as call data.

6. The AI call operation rule model based on the key indicators of preschool education according to claim 1, wherein: The steps of constructing the dynamic data pipeline to connect the cloud big data center and the preschool education AI model are as follows: Set data ports on both the cloud big data center and the preschool education AI model, and connect the data ports of the cloud big data center and the preschool education AI model through a dynamic data pipeline; Set a comparison unit in the data port of the cloud big data center. The comparison unit is used to receive the call data sent by the preschool education AI model, and perform semantic analysis on the key indicators to retrieve similar mapping indicators to obtain corresponding data keys; The data key is used to access the metric atlas and call the corresponding data group; The called data group is transmitted as reply data to the preschool education AI model through the dynamic data pipeline, and a detection window is set on the dynamic data pipeline, and the detection window is used to confirm the security of the data group.

7. The operation rule model for AI call based on key indicators of preschool education according to claim 6, wherein: The step of transmitting the called data group as reply data to the preschool education AI model through the dynamic data pipeline and setting a detection window on the dynamic data pipeline is as follows: Deposit the reply data into the comparison unit, use the comparison unit to encapsulate the reply data and assign an audit watermark layer, and use the audit watermark layer to wrap the encapsulated reply data to obtain the encapsulated reply data; Transmit the encapsulated reply data through the dynamic data pipeline, and open multiple detection windows on the dynamic data pipeline; When the encapsulated reply data passes through the detection window, a unique ID scratch of the detection window will be engraved on the audit watermark layer, and all detection window ID scratches engraved on the audit watermark layer will be identified; If scratches that do not belong to the detection window of the dynamic data pipeline are identified on the audit watermark layer, the detection window will destroy the encapsulated reply data, generate a detection log including the scratch information and send a retransmission request to the cloud big data center.

8. Based on the AI call operation rule model for key indicators of preschool education, characterized in that: Cloud server module: Build a cloud big data center to collect multi-source data of preschool education; Preschool education AI module: Build a preschool education AI model and define call rules; Data security module: Build a dynamic data pipeline, and the cloud big data center and the preschool education AI model interact through the dynamic data pipeline.

Images