Learner ability development prediction method and system based on big data analysis

Abstract

The embodiment of the invention provides a learner ability development prediction method and system based on big data analysis, and belongs to the technical field of education informatization and big data analysis. The method comprises the following steps: acquiring collection channel information; acquiring learning behaviors and evaluation data according to the collection channel information; acquiring learning state analysis data according to the learning behavior, the evaluation data and the collection channel information; acquiring learning development feature data according to the learning behavior and evaluation data and the learning state analysis data; according to the learning development feature data, training a learner ability development prediction model; and obtaining student ability development prediction data according to the learner ability development prediction model. The learning potential of a learner can be stimulated, and comprehensive improvement of the comprehensive ability of the learner is promoted.

Description

Technical Field

[0001] The present application relates to the field of educational informatization and big data analysis technology, and in particular to a method and system for predicting learner ability development based on big data analysis. Background Art

[0002] In the traditional education model, the assessment and prediction of learners' ability development mainly rely on limited methods such as standardized test scores and teachers' subjective evaluations, which have the following shortcomings:

[0003] (1) Single evaluation method: It is difficult to comprehensively cover learners’ comprehensive ability performance in different aspects, such as innovation ability, teamwork ability and other non-cognitive abilities.

[0004] (2) Lack of personalized analysis: It is impossible to make targeted predictions on ability development based on each learner’s unique learning path, interest preferences, and knowledge weaknesses.

[0005] (3) Insufficient data accuracy and completeness: Due to the limitations of data collection methods and scope, the learning-related data obtained are often incomplete and accompanied by many errors.

[0006] (4) Poor prediction timeliness and dynamism: It is unable to respond promptly to the rapid changes in learners' abilities at different learning stages, and it is difficult to provide real-time and continuous prediction results. Summary of the Invention

[0007] The main purpose of the embodiments of the present application is to provide a learner ability development prediction method and system based on big data analysis.

[0008] The technical solution adopted by the present invention is:

[0009] In one aspect, an embodiment of the present invention provides a method for predicting learner ability development based on big data analysis, the method comprising the following steps:

[0010] Obtain collection channel information;

[0011] Acquire learning behavior and assessment data based on the collection channel information;

[0012] Acquiring learning status analysis data based on the learning behavior and evaluation data and the collection channel information;

[0013] Acquiring learning development characteristic data based on the learning behavior and evaluation data and the learning status analysis data;

[0014] training a learner ability development prediction model based on the learning development characteristic data;

[0015] According to the learner ability development prediction model, student ability development prediction data is obtained.

[0016] Furthermore, the acquisition of collection channel information includes the following steps:

[0017] Obtain education application platform data;

[0018] Obtain test assessment data;

[0019] Obtain data from learners’ wearable devices;

[0020] Obtain community interaction platform data;

[0021] Collection channel information is obtained based on the education application platform data, the test evaluation data, the learner wearable device data and the community interaction platform data.

[0022] Furthermore, the acquisition of learning behavior and evaluation data based on the collection channel information includes the following steps:

[0023] Acquire learning behavior data based on the collection channel information; the learning behavior data includes course clickstream data, learning time data, content preference data, and task completion status data;

[0024] Acquire learner assessment data based on the information collected through the collection channels; the assessment data includes homework score data, semester course exam score data, and online assessment score data;

[0025] According to the learning behavior data and the evaluation data, learning behavior and evaluation data are obtained.

[0026] Furthermore, obtaining learning status analysis data based on the learning behavior and evaluation data and the collection channel information includes the following steps:

[0027] Based on the learning behavior and assessment data and the collection channel information, physiological indicators of learners under different learning tasks are obtained through wearable devices and emotion monitoring technology; the physiological indicators include brain waves, heart rate, eye tracking, and facial expressions;

[0028] Obtain learners' psychological state data through questionnaire surveys; the psychological state data includes learning stress level data, learning motivation data, and emotional stability data;

[0029] Obtaining social interaction data from learners' interactive platforms; the interactive platforms include learning communities, online forums, and group collaboration platforms; the social interaction data includes the number of posts, reply content, performance data on participation in learning group projects, number of active speeches, and contribution of speech content;

[0030] Learning state analysis data is obtained according to the physiological indicators, the psychological state data and the social interaction data.

[0031] Furthermore, obtaining learning development characteristic data based on the learning behavior and evaluation data and the learning status analysis data includes the following steps:

[0032] Performing a data cleaning operation on the learning behavior and evaluation data and the learning status analysis data to obtain first data; the data cleaning operation includes eliminating invalid data, filling missing data, and eliminating duplicate data;

[0033] Based on the first data, for data from different sources and different dimensions, a unified conversion standard is used to integrate and convert the data to obtain second data;

[0034] Extracting characteristic variables for predicting learner ability development based on the second data using statistical analysis and machine learning algorithms; the characteristic variables include correlation characteristics between learning time and learner knowledge proficiency, and correlation characteristics between number of practice sessions and learner knowledge proficiency;

[0035] According to the characteristic variables, the target data dimension is screened out using principal component analysis dimensionality reduction technology;

[0036] According to the target data dimension, learning development characteristic data is obtained.

[0037] Furthermore, the training of a learner capability development prediction model based on the learning development feature data comprises the following steps:

[0038] Obtain learner ability type information and prediction target information; the learner ability type information includes cognitive ability, non-cognitive ability, and professional skills; the prediction target information includes short-term progress trend data and long-term development potential data;

[0039] Selecting a prediction model structure based on the learner ability type information and the prediction target information; the prediction model structure includes support vector machine, random forest, long short-term memory network, and Bayesian network;

[0040] Model training is performed based on the prediction model structure and the learning development feature data to obtain a learner ability development prediction model; the model training includes using a cross-validation method to evaluate model performance and adjust model parameters and hyperparameters.

[0041] Furthermore, obtaining student ability development prediction data according to the learner ability development prediction model includes the following steps:

[0042] Obtain the data to be predicted;

[0043] Obtain user demand data; the user demand data includes report format and analysis type; the report format includes ability development curve chart and radar chart; the analysis type includes class overall ability distribution analysis, growth path planning analysis, and improvement measure suggestion analysis;

[0044] The student ability development prediction data is obtained according to the data to be predicted, the user demand data and the learner ability development prediction model.

[0045] Furthermore, the learner ability development prediction method based on big data analysis further includes the following steps:

[0046] Set up an adaptive model update mechanism;

[0047] Setting the learner ability development prediction model to automatically perform incremental updates according to the adaptive model update mechanism within a preset time to obtain an updated model;

[0048] Collect user feedback information;

[0049] The learner ability development prediction model is optimized based on the updated model and the user feedback information.

[0050] On the other hand, an embodiment of the present invention further provides a learner ability development prediction system based on big data analysis, which is used to implement the learner ability development prediction method based on big data analysis described above. The learner ability development prediction system based on big data analysis includes:

[0051] The first module is used to obtain collection channel information;

[0052] The second module is used to obtain learning behavior and evaluation data based on the collection channel information;

[0053] The third module is used to obtain learning status analysis data based on the learning behavior and evaluation data and the collection channel information;

[0054] A fourth module is configured to obtain learning development characteristic data based on the learning behavior and evaluation data and the learning status analysis data;

[0055] A fifth module is used to train a learner ability development prediction model based on the learning development feature data;

[0056] The sixth module is used to obtain student ability development prediction data based on the learner ability development prediction model.

[0057] Furthermore, the learner ability development prediction system based on big data analysis also includes:

[0058] A data collection module is used to collect raw data on learners' ability development from multiple channels; the raw data includes learning behavior data, academic performance data, physiological and psychological data, and social interaction data;

[0059] A data preprocessing module, used for cleaning, integrating and converting the raw data;

[0060] Model training and updating module, used to build and optimize the learner ability development prediction model;

[0061] Prediction and report generation module, used to generate student ability development prediction data;

[0062] The feedback and optimization module is used to collect user feedback on the student ability development prediction data and optimize the learner ability development prediction model based on the feedback.

[0063] The embodiments of the present application include at least the following beneficial effects: This application provides a method and system for predicting learner ability development based on big data analysis. The steps of the present invention include obtaining collection channel information; obtaining learning behavior and assessment data based on the collection channel information; obtaining learning status analysis data based on the learning behavior and assessment data and the collection channel information; obtaining learning development feature data based on the learning behavior and assessment data and the learning status analysis data; training a learner ability development prediction model based on the learning development feature data; and obtaining student ability development prediction data based on the learner ability development prediction model. This invention can help stimulate learners' learning potential and promote the overall improvement of their comprehensive abilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 is a schematic diagram of a learner ability development prediction method based on big data analysis provided by an embodiment of the present invention;

[0065] Figure 2 Schematic diagram of a learner ability development prediction system based on big data analysis provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are merely examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the appended claims.

[0067] It will be understood that the terms "first", "second", etc. used in this application may be used herein to describe various concepts, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0068] The terms "at least one", "plurality", "each", "any", etc. used in this application include "at least one", "two" or more, "plurality" or "each", "any" or "any one", "each" or "any one" in the context of the present invention, and "at least one" or "at least one" includes one, two or more, "plurality" or "any one" includes two or more, "each" or "each one" in the context of the present invention, and "any" or "any one

[0069] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0070] Before explaining the embodiments of the present application in detail, some of the nouns and terms involved in the embodiments of the present application are first explained. The nouns and terms involved in the embodiments of the present application are subject to the following explanations.

[0071] 1) LSTM (Long Short-Term Memory), long short-term memory network;

[0072] 2) GAN (Generative Adversarial Network), generative adversarial network.

[0073] The embodiments of the present invention are further described below with reference to the accompanying drawings.

[0074] On the one hand, the embodiment of the present invention provides a method for predicting learner ability development based on big data analysis, referring to Figure 1 ,The learner ability development prediction method based on big data ,analysis includes the following steps:

[0075] S100, obtaining collection channel information;

[0076] S200, obtaining learning behavior and evaluation data based on the collection channel information;

[0077] S300, obtaining learning status analysis data based on learning behavior and assessment data and collection channel information;

[0078] S400, obtaining learning development characteristic data based on learning behavior and assessment data and learning status analysis data;

[0079] S500, training a learner ability development prediction model based on learning development characteristic data;

[0080] S600. Obtain student ability development prediction data based on the learner ability development prediction model.

[0081] The step S100 of obtaining the collection channel information disclosed in the embodiment of the present invention includes the following steps:

[0082] S110, obtaining education application platform data;

[0083] S120, obtaining test evaluation data;

[0084] S130, obtaining learner's wearable device data;

[0085] S140, obtaining community interaction platform data;

[0086] S150. Obtain collection channel information based on education application platform data, test assessment data, learner wearable device data, and community interaction platform data.

[0087] S200 disclosed in the embodiment of the present invention obtains learning behavior and evaluation data based on the collection channel information, including the following steps:

[0088] S210. Acquire learning behavior data based on the collection channel information; the learning behavior data includes course clickstream data, learning time data, content preference data, and task completion status data;

[0089] S220. Obtain learner assessment data based on the information collected through the collection channel; the assessment data includes homework score data, semester course exam score data, and online assessment score data;

[0090] S230. Obtain learning behavior and evaluation data based on the learning behavior data and the evaluation data.

[0091] S300 disclosed in the embodiment of the present invention obtains learning status analysis data based on learning behavior and evaluation data and collection channel information, including the following steps:

[0092] S310. Based on learning behavior and assessment data and collection channel information, obtain learners' physiological indicators under different learning tasks through wearable devices and emotion monitoring technology; physiological indicators include brain waves, heart rate, eye tracking, and facial expressions;

[0093] S320, obtaining learners' psychological state data through questionnaire survey; the psychological state data includes learning stress level data, learning motivation data, and emotional stability data;

[0094] S330, obtaining social interaction data of learners' interactive platforms; interactive platforms include learning communities, online forums, and group collaboration platforms; social interaction data includes the number of posts, reply content, performance data on participation in learning group projects, number of active speeches, and contribution of speech content;

[0095] S340. Obtain learning status analysis data based on physiological indicators, psychological status data, and social interaction data.

[0096] S400 disclosed in the embodiment of the present invention obtains learning development characteristic data based on learning behavior and evaluation data and learning status analysis data, including the following steps:

[0097] S410, performing a data cleaning operation on the learning behavior and evaluation data and the learning status analysis data to obtain first data; the data cleaning operation includes eliminating invalid data, filling missing data, and eliminating duplicate data;

[0098] S420: Based on the first data, for data from different sources and in different dimensions, a unified conversion standard is used to integrate and convert the data to obtain second data;

[0099] S430. Extracting characteristic variables for predicting learner ability development based on the second data using statistical analysis and machine learning algorithms; the characteristic variables include correlation characteristics between learning time and learner knowledge proficiency, and correlation characteristics between number of practice sessions and learner knowledge proficiency.

[0100] S440. Based on the characteristic variables, the target data dimension is screened out using principal component analysis dimensionality reduction technology;

[0101] S450. Obtain learning development characteristic data according to the target data dimension.

[0102] As an optional implementation manner, the target data dimensions of the embodiment of the present invention include the data dimensions that are most representative for predicting innovation capabilities.

[0103] S500 disclosed in the embodiment of the present invention trains a learner's ability development prediction model based on learning development feature data, including the following steps:

[0104] S510, acquiring learner ability type information and predicted target information; learner ability type information includes cognitive ability, non-cognitive ability, and professional skills; predicted target information includes short-term progress trend data and long-term development potential data;

[0105] S520, selecting a prediction model structure based on the learner's ability type information and the prediction target information; the prediction model structure includes support vector machine, random forest, long short-term memory network, and Bayesian network;

[0106] S530. Perform model training based on the prediction model structure and learning development characteristic data to obtain a learner ability development prediction model; the model training includes using a cross-validation method to evaluate the performance of the model and adjust model parameters and hyperparameters.

[0107] As an optional implementation method, the embodiment of the present invention first performs standardization and feature engineering (such as sliding window statistics, key indicator extraction) on the data based on the prediction model structure (such as LSTM, random forest, etc.) and learning development feature data (including time series or static features such as historical grades and behavior logs), and then divides the data into training sets and validation sets, uses cross-validation methods to evaluate model performance, and adjusts hyperparameters (such as the number of hidden layers of LSTM and the tree depth of random forest) through grid search or Bayesian optimization. Finally, a learner ability development prediction model is trained to output prediction results such as the probability of short-term improvement trend and the level of long-term development potential based on the input new student feature data.

[0108] S600 disclosed in the embodiment of the present invention obtains student ability development prediction data based on the learner ability development prediction model, including the following steps:

[0109] S610, obtaining data to be predicted;

[0110] S620. Obtain user demand data; user demand data includes report format and analysis type; report format includes ability development curve chart and radar chart; analysis type includes class overall ability distribution analysis, growth path planning analysis, and improvement measure suggestion analysis;

[0111] S630: Obtain student ability development prediction data based on the data to be predicted, user demand data, and the learner ability development prediction model.

[0112] As an optional implementation, the data to be predicted in the embodiment of the present invention includes original academic data (including test scores, homework completion), behavioral data (including online learning time, wrong question repetition rate), physiological data (including EEG concentration index, heart rate variability), etc.

[0113] The embodiment of the present invention processes the data to be predicted (such as structured data such as students' historical grades, learning behavior logs, etc.) and user demand data (including parameters such as report format and analysis type) through a trained learner ability development prediction model (such as an LSTM multi-task model), and finally outputs student ability development prediction data.

[0114] The student ability development prediction data of the embodiment of the present invention includes ability dimension scores, progress trends, development potential levels, learning path recommendations, resource matching (recommendations for learning materials or courses that are adapted to ability characteristics), etc.

[0115] The method for predicting learner ability development based on big data analysis disclosed in an embodiment of the present invention further includes the following steps:

[0116] S700, setting an adaptive model update mechanism;

[0117] S800, setting the learner ability development prediction model to automatically perform incremental updates according to the adaptive model update mechanism within a preset time to obtain an updated model;

[0118] S900, collecting user feedback information;

[0119] S1000. Optimize the learner ability development prediction model based on the updated model and user feedback information.

[0120] On the other hand, an embodiment of the present invention further provides a learner ability development prediction system based on big data analysis, which is used to implement the above-mentioned learner ability development prediction method based on big data analysis. The learner ability development prediction system based on big data analysis includes:

[0121] The first module is used to obtain collection channel information;

[0122] The second module is used to obtain learning behavior and evaluation data based on the collection channel information;

[0123] The third module is used to obtain learning status analysis data based on learning behavior and evaluation data and collection channel information;

[0124] The fourth module is used to analyze data based on learning behavior and assessment data and learning status to obtain learning development feature data;

[0125] The fifth module is used to train the learner ability development prediction model based on learning development characteristic data;

[0126] The sixth module is used to obtain students' ability development prediction data based on the learner ability development prediction model.

[0127] The learner ability development prediction system based on big data analysis disclosed in the embodiment of the present invention is referred to Figure 2 , also includes:

[0128] The data collection module is used to collect raw data on learners' ability development from multiple channels; the raw data includes learning behavior data, academic performance data, physiological and psychological data, and social interaction data;

[0129] Data preprocessing module, used to clean, integrate and transform raw data;

[0130] Model training and updating module, used to build and optimize the learner ability development prediction model;

[0131] Prediction and report generation module, used to generate student ability development prediction data;

[0132] The feedback and optimization module is used to collect users' feedback on students' ability development prediction data and optimize the learner ability development prediction model based on the feedback.

[0133] As an optional implementation method, the embodiment of the present invention proposes a learner ability development prediction method and system based on big data analysis, which aims to establish an accurate prediction model by widely collecting and integrating multi-dimensional learning-related data and applying advanced data analysis and mining technologies, so as to proactively grasp the development trends of learners' abilities and provide strong decision-making support for educators, learners and parents.

[0134] The system architecture of the embodiment of the present invention includes:

[0135] 1. Data acquisition module

[0136] Responsible for collecting various data related to learners' ability development from multiple channels, including but not limited to:

[0137] (1) Learning behavior data: Real-time acquisition of learners’ clickstream data, learning duration, content preferences, task completion status, etc. from various online learning platforms and digital teaching applications. For example, the number of times students pause and replay videos in an online course, and the speed and accuracy of completing homework exercises can be recorded.

[0138] (2) Academic performance data (learner assessment data): Connect to the school's academic management system and various online assessment tools to collect learners' scores in various exams, homework, and quizzes, as well as fluctuation trends in scores. For example, collect students' exam scores for each course each semester, as well as specific scores in unit tests.

[0139] (3) Physiological and psychological data (physiological indicators and psychological state data): physiological indicators such as brain waves, heart rate, eye tracking, and facial expressions of learners are obtained through wearable devices and emotion monitoring technology, as well as psychological state data such as learning stress level, learning motivation, and emotional stability obtained through questionnaires. For example, brain wave monitoring equipment can be used to analyze learners' concentration and fatigue levels under different learning tasks.

[0140] (4) Social interaction data: This data is collected from learning communities, online forums, and group collaboration platforms to collect information about learners’ interactions with others, such as the number of posts, the content of replies, and their performance in group projects. This data can reflect the development of their social skills, such as teamwork and communication. For example, the number of times students actively speak in group discussions and the degree of contribution of their speeches can be counted.

[0141] 2 Data preprocessing module

[0142] Clean, integrate, and transform the massive and diverse raw data collected to ensure the effectiveness and accuracy of subsequent analysis. The main steps include:

[0143] (1) Data cleaning: Exception processing such as eliminating invalid data, filling missing data, and eliminating duplicate data. For example, incorrect login records caused by system failures in online learning platforms are deleted, and the grades of homework that learners occasionally fail to submit are reasonably estimated and supplemented through interpolation methods.

[0144] (2) Data normalization: For data from different sources and different dimensions, a unified conversion standard is used. For example, the score range of academic performance (such as percentage and five-point system) is standardized so that the data can be compared and analyzed on the same basis.

[0145] (3) Feature extraction and selection: Using statistical analysis, machine learning algorithms, and other methods, we extract from the raw data the characteristic variables that have a significant impact on the prediction of learner ability development. For example, we use correlation analysis to identify the correlation characteristics between learning time, number of practice sessions, and learner knowledge proficiency, and use principal component analysis dimensionality reduction technology to screen out the data dimensions that are most representative of predicting innovation ability.

[0146] 3. Model training and update module

[0147] Build and continuously optimize the learner ability development prediction model to achieve accurate prediction of future ability development trends.

[0148] (1) Model construction: Based on the learner's ability type information (such as cognitive ability, non-cognitive ability, professional skills, etc.) and prediction target (such as short-term progress trend data, long-term development potential data, etc.), select the appropriate prediction model structure to build the model. The optional prediction model structure includes but is not limited to support vector machine, random forest, neural network (especially deep learning model such as long short-term memory network LSTM for time series prediction), Bayesian network, etc. For example, for the short-term development prediction of learners' mathematical ability, the LSTM model is used to process the time series characteristics of their learning behavior and performance data over a period of time; for the long-term potential exploration of learners' artistic creativity, the deep learning-based generative adversarial network (GAN) is used to extract implicit artistic style preferences and creative inspiration clues.

[0149] (2) Model training: The model is fully trained using a large amount of pre-processed historical learning data. During the training process, cross-validation and other methods are used to evaluate the performance of the model, and the prediction accuracy is improved by adjusting the model parameters and hyperparameters. For example, when training a learner's reading comprehension ability prediction model based on random forests, multiple cross-validations (such as 10-fold cross-validation) are used to determine the optimal number of decision trees, tree depth and other parameters, so that the model achieves the highest accuracy and the lowest overfitting risk on the validation set.

[0150] (3) Model update: Considering the dynamic changes in learners' abilities and educational environment, an adaptive model update mechanism is set up. When the system collects a certain amount of new data or detects a significant change in the learners' learning patterns, it automatically triggers the retraining and updating of the model to ensure the timeliness and accuracy of the prediction results. For example, when a school introduces a new teaching method or a learner begins to learn a new field of study, the system automatically updates the relevant prediction model incrementally within a certain period (such as weekly or monthly).

[0151] 4. Forecasting and report generation module

[0152] Based on the trained model and the latest data, prediction results about learners' ability development are generated and presented in an intuitive and easy-to-understand form.

[0153] (1) Predictive Calculation: Using the optimized prediction model, quantitatively calculate the learner's ability development trend over a certain period of time in the future. The prediction results can include information such as the possibility of ability improvement or decline, the time required to reach the expected ability level, and the key factors affecting ability development. For example, it is predicted that the probability of a student's math score improving from the current 70 points to around 85 points in the next semester is 80%, and it is pointed out that improving problem-solving speed and strengthening the mastery of geometry knowledge are the keys to achieving this goal.

[0154] (2) Report generation: Generate personalized learner ability development prediction reports according to user needs. The report content can be customized according to specific scenarios and user roles. For example, the report for teachers focuses on the overall ability distribution and strengths and weaknesses analysis of the class; the report for students and parents focuses more on individual ability growth path planning and improvement measures. The report format can be diverse, including text descriptions, tabular data, and visual charts (such as ability development curves, radar charts, etc.). Taking visual charts as an example, by drawing trend curves of learners' various ability indicators over time, the dynamic trajectory of their ability development can be clearly displayed. At the same time, with the help of radar charts, the relative strengths and weaknesses of learners' different abilities can be compared to intuitively present the comprehensive ability level.

[0155] 5. Feedback and optimization module

[0156] Collect user feedback on prediction results and use it to further optimize system performance and prediction results.

[0157] (1) User feedback collection: Provide convenient feedback channels, such as built-in feedback modules and questionnaire links, to encourage users to evaluate and rate the accuracy and practicality of prediction results, and allow them to add specific feedback content, such as analysis of the reasons for the deviation between the prediction results and actual ability development (such as interruption of learning progress due to unexpected factors), suggestions for improving the content and format of the prediction report, etc. For example, a teacher can use the system feedback function to point out that although the ability prediction results of a student are accurate, they hope that the report can include more in-depth analysis of ability development strategies.

[0158] (2) Optimization and iteration: Based on user feedback and other monitoring indicators during system operation (such as prediction error rate, data update timeliness, etc.), regularly evaluate and optimize data collection strategies, preprocessing methods, model algorithms, etc. If it is found that a large number of user feedback prediction reports ignore certain important non-cognitive ability factors, such as the impact of learners' interests and hobbies on their innovation ability, it is necessary to re-examine the data collection module and feature selection process, consider broadening the data collection scope or improving the feature mining algorithm, so as to gradually improve the processing and prediction capabilities of non-cognitive ability-related data in subsequent system iterations.

[0159] The relationship between the various modules and the data flow process in the learner ability development prediction system based on big data analysis in an embodiment of the present invention include: a data acquisition module collects data from different data sources; a data preprocessing module cleans, normalizes, and extracts features from the data; a model training and update module uses the processed data to train the prediction model and updates it according to new data; a prediction and report generation module generates prediction results and personalized reports based on the prediction model; and a feedback and optimization module collects user feedback and optimizes system performance accordingly.

[0160] The learner ability development prediction system based on big data analysis in an embodiment of the present invention provides a user operation interface for users such as learners, teachers, or parents. The user operation interface includes a navigation bar, a prediction report display area, an interactive toolbar, and other parts. The navigation bar contains options such as system settings and personal center; the prediction report display area mainly uses visual charts, combined with tables and text descriptions to present detailed learner ability development prediction information; the interactive toolbar provides function buttons such as data filtering, report export, and feedback submission. Users can click the corresponding buttons to perform operations, such as filtering prediction reports for different time periods, exporting report files in PDF format, or submitting usage feedback.

[0161] The system advantages of the present invention are:

[0162] (1) Comprehensiveness and accuracy: It integrates cross-platform, multi-dimensional learning-related data, which can more comprehensively reflect the learner's ability status compared to the traditional single data source evaluation method; at the same time, with the help of advanced big data analysis and machine learning technology, it greatly improves the accuracy of ability development prediction.

[0163] (2) Personalization and targeting: Customized data collection, analysis, and prediction are conducted based on the unique characteristics of each learner to achieve truly personalized education and precise training.

[0164] (3) Dynamic and real-time: It has the ability to update data in real time and dynamically adjust the prediction model, which can respond to changes in learners' ability development in a timely manner and provide timely and effective basis for educational decision-making.

[0165] (4) Decision support: Provide educators, learners, and parents with comprehensive predictive information on learner ability development, helping them plan learning paths in advance, adjust teaching strategies, and accurately allocate resources to promote the optimal development of learners' abilities.

[0166] As an optional implementation, the embodiment of the present invention takes the prediction of student learning ability as an example, and includes the following steps:

[0167] Step 1: System Installation and Deployment: Install and deploy the system in the school's local area network or on a cloud server. Simultaneously, embed data acquisition interfaces in the school's various online learning platforms, the educational administration system, and teachers' teaching terminals to ensure smooth data transmission to the system server.

[0168] Step 2: Data Collection and Accumulation: Once the system is operational, it automatically collects student learning data from multiple connected data sources. For example, over a two-month period, it collects learning behavior data (such as the number of micro-lessons watched and homework completed) from all first-year high school students on learning platforms for core subjects like Chinese, mathematics, and English. This data also includes test score data from the school's academic affairs system, physiological data monitored by wearable devices (such as average heart rate during class), and social interaction data, such as the number of times students participate in discussions within the learning community.

[0169] Step 3: Data Preprocessing: Preprocess the collected data from the past two months. First, invalid learning behavior records due to student errors were deleted. Next, the exam scores for each course were normalized to a percentage. Then, principal component analysis was used to extract key features that influence learning ability from the learning behavior data, such as daily online learning time, average chapter quiz scores, and the number of repeated learning points. Finally, for social interaction data, text mining techniques were used to conduct sentiment analysis on students' forum posts to extract indicators of participation and collaboration.

[0170] Step 4: Model Training and Initial Prediction: Using preprocessed data and the school's complete study materials for first-year high school students from previous years, a recurrent neural network model based on deep learning was constructed. This model was trained using various student characteristic variables as input and final exam scores and other comprehensive ability evaluation indicators as output targets. After multiple iterations and parameter adjustments, the model achieved prediction accuracy exceeding 90% on the validation set. Subsequently, the model was fed with the latest data from the current first-year high school students to generate preliminary predictions for each student's ability development over the following semester, including information such as expected grade ranking changes and development trends in various subject areas.

[0171] Step 5: Report generation and release: Generate a personalized ability development prediction report for each student based on the prediction results of the model. The report uses a combination of pictures and texts, and uses clear line graphs to intuitively present the expected development trends of students' various ability indicators (such as Chinese reading comprehension ability, mathematical logical thinking ability) in the next semester; uses bar graphs to compare the changes in the relative positions of students' current abilities in the class; and uses radar charts to show the distribution of students' comprehensive abilities. At the same time, in the text description part, targeted learning suggestions are made based on the prediction results, such as encouraging students with weak Chinese abilities to increase their extracurricular reading, and suggesting that students with strong mathematical abilities become leaders in group learning to drive the team's common progress. Finally, these reports are sent to the corresponding teachers, students, and parents through the system platform.

[0172] Step 6: Feedback and system optimization: After receiving the prediction report, teachers, students and parents can log in to the system feedback interface to view and submit feedback. For example, some teachers pointed out that the accuracy of the prediction of experimental operation ability in the prediction report needs to be improved due to the lack of actual experimental data support. After collecting these feedbacks, the technical staff will consult with the school laboratory management personnel to add an experimental equipment usage monitoring interface to the data acquisition module to collect data such as students' operation steps, time consumption and experimental results in physics, chemistry and biology experiments in real time. In the subsequent model update process, the data features related to experimental operation ability will be included in the training set, and the model will be retrained to improve the accuracy of the prediction of practical ability. After a month of optimization and adjustment, the prediction error rate of experimental operation ability was reduced by 30% when it was predicted again, and the credibility and practicality of the prediction report were significantly enhanced.

[0173] The present invention's learner ability development prediction system, based on big data analysis, achieves dynamic and accurate prediction of learner ability development through comprehensive, multi-level data collection, precise data preprocessing, efficient model training and updating, intuitive prediction report generation, and continuous system optimization. This system not only overcomes the limitations of traditional educational assessment and prediction methods but also provides strong technical support for personalized education, precise resource allocation, and teaching strategy optimization. It helps stimulate learners' learning potential and promotes the overall improvement of their comprehensive abilities. It has extremely broad application prospects and profound educational significance.

[0174] On the other hand, an embodiment of the present invention also provides a learner ability development prediction device based on big data analysis, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, it implements the learner ability development prediction method based on big data analysis as described above.

[0175] The processor and the memory can be connected via a bus or other means. The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0176] On the other hand, an embodiment of the present invention further provides a computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions are used to enable a computer to execute the above-mentioned learner ability development prediction method based on big data analysis.

[0177] Those skilled in the art will appreciate that all or some of the steps and systems in the method disclosed above can be implemented as software, firmware, hardware, and appropriate combinations thereof. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, and the computer-readable medium can include computer storage media (or non-transitory media) and communication media (or temporary media). As known to those skilled in the art, the term computer storage media is included in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data) and is volatile and non-volatile, removable, and non-removable. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory, or other memory technology, CD-ROM, digital versatile disks (DVD), or other optical disk storage, magnetic cassettes, magnetic tapes, disk storage, or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, as is well known to those skilled in the art, communication media typically embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0178] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A method for predicting learner ability development based on big data analysis, characterized in that: The learner ability development prediction method based on big data analysis includes the following steps: Obtain collection channel information; Acquire learning behavior and assessment data based on the collection channel information; Acquiring learning status analysis data based on the learning behavior and evaluation data and the collection channel information; Acquiring learning development characteristic data based on the learning behavior and evaluation data and the learning status analysis data; training a learner ability development prediction model based on the learning development characteristic data; According to the learner ability development prediction model, student ability development prediction data is obtained.

2. The method for predicting learner ability development based on big data analysis according to claim 1 is characterized in that: The acquisition of collection channel information includes the following steps: Obtain education application platform data; Obtain test assessment data; Obtain data from learners’ wearable devices; Obtain community interaction platform data; Collection channel information is obtained based on the education application platform data, the test evaluation data, the learner wearable device data and the community interaction platform data.

3. The method for predicting learner ability development based on big data analysis according to claim 1 is characterized in that: The step of obtaining learning behavior and evaluation data based on the collection channel information includes the following steps: Acquire learning behavior data based on the collection channel information; the learning behavior data includes course clickstream data, learning time data, content preference data, and task completion status data; Acquire learner assessment data based on the information collected through the collection channels; the assessment data includes homework score data, semester course exam score data, and online assessment score data; According to the learning behavior data and the evaluation data, learning behavior and evaluation data are obtained.

4. The method for predicting learner ability development based on big data analysis according to claim 1 is characterized in that: The step of obtaining learning status analysis data based on the learning behavior and evaluation data and the collection channel information includes the following steps: Based on the learning behavior and assessment data and the collection channel information, physiological indicators of learners under different learning tasks are obtained through wearable devices and emotion monitoring technology; the physiological indicators include brain waves, heart rate, eye tracking, and facial expressions; Obtain learners' psychological state data through questionnaire surveys; the psychological state data includes learning stress level data, learning motivation data, and emotional stability data; Obtaining social interaction data from learners' interactive platforms; the interactive platforms include learning communities, online forums, and group collaboration platforms; the social interaction data includes the number of posts, reply content, performance data on participation in learning group projects, number of active speeches, and contribution of speech content; Learning state analysis data is obtained according to the physiological indicators, the psychological state data and the social interaction data.

5. The method for predicting learner ability development based on big data analysis according to claim 1 is characterized in that: The step of obtaining learning development characteristic data based on the learning behavior and evaluation data and the learning status analysis data comprises the following steps: Performing a data cleaning operation on the learning behavior and evaluation data and the learning status analysis data to obtain first data; the data cleaning operation includes eliminating invalid data, filling missing data, and eliminating duplicate data; Based on the first data, for data from different sources and different dimensions, a unified conversion standard is used to integrate and convert the data to obtain second data; Extracting characteristic variables for predicting learner ability development based on the second data using statistical analysis and machine learning algorithms; the characteristic variables include correlation characteristics between learning time and learner knowledge proficiency, and correlation characteristics between number of practice sessions and learner knowledge proficiency; According to the characteristic variables, the target data dimension is screened out using principal component analysis dimensionality reduction technology; According to the target data dimension, learning development characteristic data is obtained.

6. The method for predicting learner ability development based on big data analysis according to claim 1 is characterized in that: The step of training a learner's ability development prediction model based on the learning development feature data comprises the following steps: Obtain learner ability type information and prediction target information; the learner ability type information includes cognitive ability, non-cognitive ability, and professional skills; the prediction target information includes short-term progress trend data and long-term development potential data; Selecting a prediction model structure based on the learner ability type information and the prediction target information; the prediction model structure includes support vector machine, random forest, long short-term memory network, and Bayesian network; Model training is performed based on the prediction model structure and the learning development feature data to obtain a learner ability development prediction model; the model training includes using a cross-validation method to evaluate model performance and adjust model parameters and hyperparameters.

7. The method for predicting learner ability development based on big data analysis according to claim 1 is characterized in that: The step of obtaining student ability development prediction data according to the learner ability development prediction model comprises the following steps: Obtain the data to be predicted; Obtain user demand data; the user demand data includes report format and analysis type; the report format includes ability development curve chart and radar chart; the analysis type includes class overall ability distribution analysis, growth path planning analysis, and improvement measure suggestion analysis; The student ability development prediction data is obtained according to the data to be predicted, the user demand data and the learner ability development prediction model.

8. The method for predicting learner ability development based on big data analysis according to claim 1 is characterized in that: The method for predicting learner ability development based on big data analysis further includes the following steps: Set up an adaptive model update mechanism; Setting the learner ability development prediction model to automatically perform incremental updates according to the adaptive model update mechanism within a preset time to obtain an updated model; Collect user feedback information; The learner ability development prediction model is optimized based on the updated model and the user feedback information.

9. A learner ability development prediction system based on big data analysis, used to implement the learner ability development prediction method based on big data analysis as described in any one of claims 1 to 8, characterized in that: The learner ability development prediction system based on big data analysis includes: The first module is used to obtain collection channel information; The second module is used to obtain learning behavior and evaluation data based on the collection channel information; The third module is used to obtain learning status analysis data based on the learning behavior and evaluation data and the collection channel information; A fourth module is configured to obtain learning development characteristic data based on the learning behavior and evaluation data and the learning status analysis data; A fifth module is used to train a learner ability development prediction model based on the learning development feature data; The sixth module is used to obtain student ability development prediction data based on the learner ability development prediction model.

10. The learner ability development prediction system based on big data analysis according to claim 9 is characterized in that: The learner ability development prediction system based on big data analysis also includes: A data collection module is used to collect raw data on learners' ability development from multiple channels; the raw data includes learning behavior data, academic performance data, physiological and psychological data, and social interaction data; A data preprocessing module, used for cleaning, integrating and converting the raw data; Model training and updating module, used to build and optimize the learner ability development prediction model; Prediction and report generation module, used to generate student ability development prediction data; The feedback and optimization module is used to collect user feedback on the student ability development prediction data and optimize the learner ability development prediction model based on the feedback.

Images