Children research accomplishment intelligent evaluation and promotion system integrating AI and big data

Abstract

The invention discloses an AI and big data integrated intelligent evaluation and promotion system for children's research literacy, and relates to the technical field of children's research literacy evaluation. The system comprises a data acquisition module, a data processing and analysis module, an intelligent evaluation module and a personalized suggestion module. The system can automatically collect expression, behavior and learning result data of children in research activities, perform preprocessing and intelligent analysis, and provide personalized learning suggestions and promotion strategies based on evaluation results. The invention aims to improve the accuracy and individuation degree of child research literacy evaluation and promote the comprehensive development of child research ability.

Description

Technical Field

[0001] The present invention relates to the technical field of children's research literacy assessment, and in particular to an intelligent assessment and promotion system for children's research literacy integrating AI and big data. Background Art

[0002] At present, with the continuous advancement of educational informatization, the assessment and cultivation of children's research literacy has become an important issue in the field of education. However, existing technologies still have many shortcomings in the assessment of children's research literacy, as shown below:

[0003] 1) The evaluation method is highly subjective

[0004] Existing technologies mostly rely on traditional tests, teacher observations, and project presentations for evaluation. These methods are often affected by teachers’ subjective judgments, resulting in a lack of objectivity and consistency in the evaluation results. Teachers’ personal experience, bias, or emotional fluctuations may have a significant impact on the evaluation results, making it difficult for the evaluation results to accurately reflect children’s true research literacy;

[0005] 2) Data collection is not comprehensive and timely enough

[0006] Traditional evaluation methods are difficult to comprehensively and real-time collect multi-dimensional data about children during their research process. Key information such as facial expressions, behaviors, interactions, and subtle changes in the learning process are often overlooked or difficult to capture. The lack of such data limits the depth and accuracy of the evaluation, making it impossible for the evaluation results to fully reflect children’s research literacy and potential.

[0007] 3) Inefficient evaluation

[0008] Faced with the assessment needs of large-scale students, traditional manual assessment methods are particularly time-consuming and labor-intensive. Teachers need to spend a lot of time and energy to observe and record students' performance, and then make subjective evaluations based on these data. This assessment method is not only inefficient, but also prone to omissions and errors, and it is difficult to meet the needs of modern education for efficient and accurate assessments;

[0009] 4) Lack of personalized guidance

[0010] Existing technologies can only provide general evaluation results, which are difficult to directly transform into targeted learning suggestions and promotion strategies. As a result, after obtaining the evaluation results, students often do not know how to improve their learning methods and research abilities in a targeted manner. The lack of personalized learning guidance makes it easy for students to lose their way in the research process and it is difficult to achieve the best learning results.

[0011] In summary, existing technologies have problems in the assessment of children's research literacy, such as strong subjectivity, incomplete data collection, low assessment efficiency, and lack of personalization. In order to solve these problems, it is necessary to develop a more scientific, comprehensive, efficient and personalized assessment and promotion system. Summary of the invention

[0012] In view of the defects in the prior art, the present invention provides an intelligent assessment and promotion system for children's research literacy integrating AI and big data, the system comprising:

[0013] Data collection module, used to collect children's facial expression data, behavioral data and learning outcomes during research activities;

[0014] A data processing and analysis module, used to pre-process and intelligently analyze the expression data, behavior data and learning outcomes, and determine expression features, behavior features and learning outcome features according to pre-defined dimensionality reduction rules for different types of data;

[0015] An intelligent assessment module, for inputting the expression features, behavior features and learning outcome features into a pre-built research literacy assessment model to obtain scores for various research capabilities;

[0016] The personalized suggestion module is used to obtain a research capability improvement priority sequence based on the research capability score and in combination with associated capability gains; and to formulate learning suggestions and promotion strategies based on the research capability improvement priority sequence.

[0017] Preferably, performing intelligent analysis on the expression data includes:

[0018] Using an expression recognition model to recognize the expression data, and obtaining the expression category and quantity;

[0019] Determine the proportion of focus time based on the type and quantity of expressions and their duration.

[0020] Preferably, the dimensionality reduction rules of expression data include:

[0021]

[0022] In the formula, A1 represents the first expression feature, V1 represents the heart rate variability during activity, V2 represents the heart rate variability during rest, and V2 represents the proportion of concentration time;

[0023]

[0024] Where A2 represents the second expression feature, N represents the total number of expressions of the i-th expression recognized, ω 1i represents the emotional weight of the i-th expression, ω 2i represents the time weight of the i-th expression, f irepresents the frequency of occurrence of the ith expression in the research activity, and T represents the total number of all expressions identified in the research activity;

[0025] A=α×A1+β×A1

[0026] In the formula, A represents the facial expression feature, and α and β represent the weight coefficients of the first facial expression feature and the second facial expression feature, respectively.

[0027] Preferably, the dimensionality reduction rules of behavior data include:

[0028]

[0029]

[0030] In the formula, B represents the behavior characteristics, T tasks represents the total number of tasks completed in the research activity, T tasks represents the average task completion time, e avg represents the average number of errors, i avg represents the average number of interactions, θ and ω represent the weight coefficients of the number of errors and the number of interactions, respectively.

[0031] Preferably, the dimensionality reduction rules of learning outcomes include:

[0032]

[0033] Where C represents the learning outcome feature, CA represents the number of correctly answered questions, TQ represents the total number of questions, IW represents the number of innovative solutions or innovative works provided, SP represents the number of problems solved, TT represents the total time required to solve the problems, and η, μ, and λ represent the corresponding weight coefficients.

[0034] Preferably, based on the research capability score and in combination with the associated capability gain, a research capability improvement priority sequence is obtained; and based on the research capability improvement priority sequence, learning suggestions and promotion strategies are formulated, including:

[0035] Determine initial capability enhancement priorities based on research capability scores;

[0036] Considering the associated capability gain, constructing an associated capability dependency matrix, and adjusting the gain coefficient of each research capability based on the associated capability dependency matrix;

[0037] According to the initial capability improvement priority and the gain coefficient, the research capability improvement priority is determined, and the research capability improvement priority is sorted in descending order to obtain a research capability improvement priority sequence.

[0038] Preferably, the calculation formula for the initial capability improvement priority is:

[0039]

[0040] Where P v0 Indicates the initial capability improvement priority of the vth capability, M u represents the target score value of the vth research capability, S v represents the current rating value of the vth research capability, ω v Represents the importance weight of the vth research capability.

[0041] Preferably, when there is a dependency between two research capability items, their gain coefficient is:

[0042]

[0043] Where Gain v represents the dependent gain of the v-th research capability, D vu M represents the gain coefficient of the u-th research capability to the v-th research capability, u represents the research ability score of the u-th research ability, S u Represents the importance weight of the u-th research capability.

[0044] The beneficial effects of the present invention are as follows: The present invention provides an intelligent assessment and promotion system for children's research literacy that integrates AI and big data, and the system includes four modules: data collection, data processing and analysis, intelligent assessment, and personalized suggestions. The system can automatically collect children's expressions, behaviors, and learning outcomes in research activities, perform preprocessing and intelligent analysis, and provide personalized learning suggestions and promotion strategies based on the assessment results. The present invention can improve the accuracy and personalization of children's research literacy assessment and promote the comprehensive development of children's research ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0046] Figure 1 A system block diagram of an intelligent assessment and promotion system for children's research literacy that integrates AI and big data provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0048] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0049] like Figure 1 As shown, an embodiment of the present invention provides an intelligent assessment and promotion system for children's research literacy integrating AI and big data, the system comprising:

[0050] The data collection module is used to collect children’s facial expression data, behavioral data and learning outcomes during research activities.

[0051] Use the Internet of Things, sensors (such as smart bracelets to monitor heart rate, eye trackers to track line of sight, etc.) and smart devices (such as cameras, tablets, learning machines, etc.) to collect children's behavioral data, learning outcomes and interaction information during research activities, providing a data basis for subsequent assessments of children's research literacy.

[0052] The data processing and analysis module is used to pre-process and intelligently analyze the expression data, behavior data and learning results, and determine the expression features, behavior features and learning results features according to pre-defined dimensionality reduction rules for different types of data.

[0053] Before data dimensionality reduction, the collected data needs to be preprocessed and preliminarily analyzed.

[0054] Specifically, preprocessing includes but is not limited to data cleaning and data integration, among which data cleaning is used to remove noise data, outliers and incomplete records; data integration is used to integrate data from different devices and different times into a unified format. Specifically, preliminary analysis includes identifying the type, quantity and duration of facial expressions from expression data; extracting task completion status, number of interactions, etc. from behavioral data; and extracting the number of correctly answered questions, innovative solutions, etc. from learning outcomes. The preliminary analysis of these data is a conventional technology and will not be repeated in this embodiment.

[0055] After preprocessing and preliminary analysis of the collected data, the data also needs to be reduced in dimension to improve evaluation efficiency.

[0056] Specifically, the dimensionality reduction rules for expression data include:

[0057]

[0058] In the formula, A1 represents the first expression feature, V1 represents the heart rate variability during activity, V2 represents the heart rate variability during rest, and V2 represents the proportion of concentration time;

[0059]

[0060] Where A2 represents the second expression feature, N represents the total number of expressions of the i-th expression recognized, ω 1i represents the emotional weight of the i-th expression, ω 2i represents the time weight of the i-th expression, f i represents the frequency of occurrence of the i-th expression in the research activity, T represents the total number of all expressions identified in the research activity; among them, the time weight is used to indicate the attenuation degree of children's interest in the research activity with different expressions in a research activity.

[0061] For facial expression features, although they mainly reflect children's immediate emotional state and concentration, the time effect can also be considered to a certain extent. For example, if a child shows higher concentration and positive emotions during a specific time period of a research activity, this may have a positive impact on their overall research experience and learning effect. Therefore, adding the time effect to the evaluation model can provide a more comprehensive understanding of children's emotional changes and concentration in research activities, thereby better evaluating their research literacy. Specifically, a time weight can be introduced in the calculation of facial expression features, which gives different importance according to different time periods of the research activity. For example, it can be considered that facial expression features at the beginning and end of the activity are more important because they may reflect children's initial attitude and final experience of the activity. In this way, the time effect of facial expression features and research activities can be comprehensively considered in the evaluation model.

[0062] A=α×A1+β×A1

[0063] In the formula, A represents the facial expression feature, and α and β represent the weight coefficients of the first facial expression feature and the second facial expression feature, respectively.

[0064] Specifically, the dimensionality reduction rules for behavioral data include:

[0065]

[0066]

[0067] In the formula, B represents the behavior characteristics, T tasks represents the total number of tasks completed in the research activity, T tasks represents the average task completion time, e avg represents the average number of errors, i avg represents the average number of interactions, θ and ω represent the weight coefficients of the number of errors and the number of interactions, respectively.

[0068] Specifically, the dimensionality reduction rules of learning outcomes include:

[0069]

[0070] Where C represents the learning outcome feature, CA represents the number of correctly answered questions, TQ represents the total number of questions, IW represents the number of innovative solutions or innovative works provided, SP represents the number of problems solved, TT represents the total time required to solve the problems, and η, μ, and λ represent the corresponding weight coefficients.

[0071] The dimensionality reduction rules provided by the embodiments of the present invention can comprehensively consider information from multiple dimensions and extract the most representative features, thereby effectively reducing the complexity of the data and improving the efficiency and accuracy of the evaluation. At the same time, by adjusting the weight coefficients, it can flexibly adapt to different needs.

[0072] The intelligent assessment module is used to input the expression characteristics, behavioral characteristics and learning outcome characteristics into a pre-built research literacy assessment model to obtain scores for various research capabilities.

[0073] In the embodiment of the present invention, a suitable machine learning or deep learning algorithm is selected as the basis of the evaluation model according to actual needs. Common algorithms include support vector machine (SVM), random forest, neural network, etc. There is no limitation in the embodiment of the present invention.

[0074] It should be noted that the start time, duration and end time of research activities have a significant impact on research capabilities. The closer the research activities are, the greater the impact on research capabilities. Therefore, after obtaining the research capabilities through the above model evaluation, it is also necessary to set the time decay function to correct the calculated research capability score. Of course, you can also directly add time input during the model training process. At this time, you should use the LSTM network to process time information and provide model accuracy. Of course, you can also use a machine learning model to weight each feature and multiply it by the time decay function, and then train the model to obtain the optimal model parameters.

[0075] In a preferred embodiment, research literacy can also be divided into research awareness, research ability and research evaluation. The evaluation dimensions of research awareness include curiosity, imagination and thirst for knowledge. The evaluation dimensions of research ability include problem discovery, problem solving and summary and reflection. The evaluation dimensions of research character include truth-seeking, self-help and collaboration, and responsibility. The dimension level is defined for each evaluation dimension, and these features are extracted from data such as expression data, behavioral data and learning outcomes. The research elements and dimensions corresponding to these features are combined to obtain the original data set. Based on machine learning algorithms, deep learning algorithms and other methods, the original data set can be used to train a comprehensive research literacy evaluation model.

[0076] According to the predefined dimensionality reduction rules for different types of data, expression features, behavior features and learning outcome features are determined.

[0077] The personalized suggestion module is used to obtain a research capability improvement priority sequence based on the research capability score and in combination with associated capability gains; and to formulate learning suggestions and promotion strategies based on the research capability improvement priority sequence.

[0078] In an embodiment of the present invention, based on the research capability score and in combination with the associated capability gain, a research capability improvement priority sequence is obtained; based on the research capability improvement priority sequence, learning suggestions and promotion strategies are formulated, including: determining the initial capability improvement priority based on the research capability score; considering the associated capability gain, constructing an associated capability dependency matrix, and adjusting the gain coefficient of each research capability based on the associated capability dependency matrix; determining the research capability improvement priority according to the initial capability improvement priority and the gain coefficient, and sorting the research capability improvement priorities in descending order to obtain a research capability improvement priority sequence.

[0079] In the embodiment of the present invention, the calculation formula of the initial capability improvement priority is:

[0080]

[0081] Where P v0 Indicates the initial capability improvement priority of the vth capability, M u represents the target score value of the vth research capability, S v represents the current rating value of the vth research capability, ω v Represents the importance weight of the vth research capability.

[0082] In the embodiment of the present invention, when there is a dependency relationship between two research capability items, their gain coefficient is:

[0083]

[0084] Where Gain v represents the dependent gain of the v-th research capability, D vu M represents the gain coefficient of the u-th research capability to the v-th research capability, u represents the research ability score of the u-th research ability, S u Represents the importance weight of the u-th research capability.

[0085] The specific implementation process of the personalized suggestion module is to receive the students' research ability score data from the intelligent evaluation module; obtain the dependency relationship and importance weight information between research abilities from the system database. Perform the above-mentioned initial ability improvement priority calculation, associated ability gain analysis and gain coefficient adjustment process; determine the research ability improvement priority sequence based on the processing results; and automatically generate personalized learning suggestions and promotion strategies based on the research ability improvement priority sequence. Among them, the learning suggestions and promotion strategies are presented to students in a visual form, such as through online platforms, APP push, etc.

[0086] The personalized suggestion module intelligently obtains the priority sequence of research ability improvement based on the student's research ability score and the associated ability gain, and formulates personalized learning suggestions and promotion strategies accordingly. This module is designed to help students improve their research ability in a targeted manner and achieve personalized learning.

[0087] In a preferred embodiment, the personalized suggestion module can also study the comprehensive evaluation model of literacy based on the evaluation results of the comprehensive evaluation model of literacy.

[0088] In summary, the present invention provides an intelligent assessment and promotion system for children's research literacy that integrates AI and big data. The system includes four modules: data collection, data processing and analysis, intelligent assessment, and personalized suggestions. The system can automatically collect children's expressions, behaviors, and learning outcomes in research activities, perform preprocessing and intelligent analysis, and provide personalized learning suggestions and promotion strategies based on the assessment results. The present invention can improve the accuracy and personalization of children's research literacy assessment and promote the comprehensive development of children's research ability.

[0089] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. An intelligent assessment and promotion system for children's research literacy integrating AI and big data, characterized by: include: Data collection module, used to collect children's facial expression data, behavioral data and learning outcomes during research activities; A data processing and analysis module, used to pre-process and intelligently analyze the expression data, behavior data and learning outcomes, and determine expression features, behavior features and learning outcome features according to pre-defined dimensionality reduction rules for different types of data; An intelligent assessment module, for inputting the expression features, behavior features and learning outcome features into a pre-built research literacy assessment model to obtain scores for various research capabilities; The personalized suggestion module is used to obtain a research capability improvement priority sequence based on the research capability score and in combination with associated capability gains; and to formulate learning suggestions and promotion strategies based on the research capability improvement priority sequence.

2. The system according to claim 1, characterized in that: Intelligent analysis of the expression data includes: Using an expression recognition model to recognize the expression data, and obtaining the expression category and quantity; Determine the proportion of focus time based on the type and quantity of expressions and their duration.

3. The system according to claim 1, characterized in that: The dimensionality reduction rules for expression data include: In the formula, A1 represents the first expression feature, V1 represents the heart rate variability during activity, V2 represents the heart rate variability during rest, and V2 represents the proportion of concentration time; Where A2 represents the second expression feature, N represents the total number of expressions of the i-th expression recognized, ω 1i represents the emotional weight of the i-th expression, ω 2i represents the time weight of the i-th expression, f i represents the frequency of occurrence of the ith expression in the research activity, and T represents the total number of all expressions identified in the research activity; A=α×A1+β×A1 In the formula, A represents the facial expression feature, and α and β represent the weight coefficients of the first facial expression feature and the second facial expression feature, respectively.

4. The system according to claim 1, characterized in that: The dimensionality reduction rules for behavioral data include: In the formula, B represents the behavior characteristics, T tasks represents the total number of tasks completed in the research activity, T tasks represents the average task completion time, e avg represents the average number of errors, i avg represents the average number of interactions, θ and ω represent the weight coefficients of the number of errors and the number of interactions, respectively.

5. The system according to claim 1, characterized in that: The dimensionality reduction rules for learning outcomes include: Where C represents the learning outcome feature, CA represents the number of correctly answered questions, TQ represents the total number of questions, IW represents the number of innovative solutions or innovative works provided, SP represents the number of problems solved, TT represents the total time required to solve the problems, and η, μ, and λ represent the corresponding weight coefficients.

6. The system according to claim 1, characterized in that: Based on the research capability score and in combination with the associated capability gain, a research capability improvement priority sequence is obtained; Develop learning recommendations and promotion strategies based on the research capacity enhancement priority sequence, including: Determine initial capability enhancement priorities based on research capability scores; Considering the associated capability gain, constructing an associated capability dependency matrix, and adjusting the gain coefficient of each research capability based on the associated capability dependency matrix; According to the initial capability improvement priority and the gain coefficient, the research capability improvement priority is determined, and the research capability improvement priority is sorted in descending order to obtain a research capability improvement priority sequence.

7. The system according to claim 1, characterized in that The calculation formula for the initial capability improvement priority is: Where P v0 Indicates the initial capability improvement priority of the vth capability, M u represents the target score value of the vth research capability, S v represents the current rating value of the vth research capability, ω v Represents the importance weight of the vth research capability.

8. The system according to claim 1, characterized in that: When there is a dependency between two research capability items, the gain coefficient is: Where Gain v represents the dependent gain of the v-th research capability, D vu M represents the gain coefficient of the u-th research capability to the v-th research capability, u represents the research ability score of the u-th research ability, S u Represents the importance weight of the u-th research capability.

Images