Student behavior early warning analysis method and system based on big data

Abstract

The invention discloses a student behavior early warning analysis method and system based on big data, and relates to the technical field of behavior early warning. Data preprocessing is performed on personal data and behavior data, data from different sources are integrated, and a comprehensive student information database is established; the method comprises the following steps: selecting key feature data, including scores, attendance rates and homework completion conditions, closely associated with student behaviors from a student information database, analyzing the key feature data through a regression model, outputting behavior indexes of students, and judging abnormal behavior conditions of the students according to a comparison result of the behavior indexes and a preset analysis threshold value. And corresponding early warning measures are generated according to the abnormal behavior condition and are sent to an administrator. According to the analysis method, after multiple items of data of the students are comprehensively analyzed, the behavior conditions of the students are judged, when the behaviors of the students are abnormal, corresponding early warning measures are generated according to the abnormal conditions, analysis is more comprehensive, manual intervention is avoided, and analysis is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of behavior warning, and specifically relates to a method and system for analyzing students' behavior warning based on big data. Background Art

[0002] Student behavior warning is an important task in educational management, aiming to timely discover potential problems by monitoring students' behaviors and performances, and taking effective intervention measures to improve students' academic achievements, mental health and social adaptation abilities. With the rapid development of information technology and the popularization of tools such as school management systems and student information systems, educational institutions can more conveniently collect, store and analyze students' data. This technical foundation provides strong support for student behavior warning, enabling educators to more accurately understand students' performances and states;

[0003] The existing technologies have the following defects:

[0004] The existing analysis methods usually manually judge whether there is an abnormality in students' behaviors based on a certain behavior of the students. However, analyzing only one behavior easily leads to analysis errors and incomplete analysis, and the manual judgment is highly subjective, further increasing the analysis error rate. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and system for analyzing students' behavior warning based on big data to solve the deficiencies in the background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions: A method for analyzing students' behavior warning based on big data, the analysis method includes the following steps:

[0007] Collect students' personal data, including academic achievements, attendance records, social activities, and collect students' behavior data in various places such as school systems, classrooms, libraries, etc., including login time, participation in discussions, and homework submission situations;

[0008] Perform data preprocessing on personal data and behavior data, including denoising processing and data integration, integrate data from different sources, and establish a comprehensive student information database;

[0009] Select feature data closely related to students' behaviors from the student information database, analyze the feature data through a regression model, and output students' behavior indices;

[0010] Judge the abnormal situation of students' behaviors based on the comparison result between the behavior index and the preset analysis threshold, and generate corresponding warning measures according to the abnormal situation of behaviors and send them to the administrator.

[0011] Preferably, analyze the key feature data through a regression model to output the behavior index of the student, including the following steps:

[0012] Obtain the absenteeism rate, the rate of being late or leaving early, the proportion of online course duration, and the proportion of network usage time of the student;

[0013] Substitute the absenteeism rate, the rate of being late or leaving early, the proportion of online course duration, and the proportion of network usage time into the regression model for calculation to obtain the behavior index of the student. The expression is:

[0014]

[0015] , where XWZ is the behavior index, ql, cdz, tsl, and wzb are the absenteeism rate, the rate of being late or leaving early, the proportion of online course duration, and the proportion of network usage time respectively, and a 1 、a 2 、a 3 、a 4 are the regression coefficients of the absenteeism rate, the rate of being late or leaving early, the proportion of online course duration, and the proportion of network usage time respectively, and a 1 、a 2 、a 3 、a 4 are all greater than 0.

[0016] Preferably, judge the abnormal situation of the student's behavior through the comparison result of the behavior index and the preset analysis threshold, including the following steps:

[0017] After obtaining the behavior index, compare the behavior index with the preset analysis threshold. The analysis threshold includes the first abnormal threshold and the second abnormal threshold. The first abnormal threshold is used to judge whether the student has an abnormality, and the second abnormal threshold is used to judge the severity of the student's abnormality;

[0018] If the behavior index ≤ the first abnormal threshold, judge that the student's behavior is normal;

[0019] If the behavior index > the first abnormal threshold, judge that the student's behavior is abnormal;

[0020] If the behavior index > the first abnormal threshold and the behavior index ≤ the second abnormal threshold, judge that the student's behavior has a minor abnormality;

[0021] If the behavior index > the second abnormal threshold, judge that the student's behavior has a serious abnormality.

[0022] Preferably, integrate the data from different sources to establish a comprehensive student information database, including the following steps:

[0023] For data from different sources, perform field mapping so that the same or similar information is mapped to the same field, determine the association relationships between different tables, establish connections between data through primary keys and foreign keys, keep the data in the database synchronized with the source data based on a regular update mechanism, set different levels of data access permissions according to user roles and requirements, and use encryption and other means to protect sensitive information.

[0024] Preferably, perform data preprocessing on personal data and behavioral data, including the following steps:

[0025] Detect and handle missing values in personal data and behavioral data, find and delete existing duplicate records, use statistical methods or machine learning methods to detect outliers in the data, including personal data and behavioral data, and select a denoising method for denoising according to the nature of the outliers, standardize continuous data through Z-score normalization, map the data to the corresponding range, for data containing categorical information, convert it into numerical data using one-hot encoding or other encoding methods, perform moving average processing on time series data, for data in the field of signal processing, use a filter for smoothing to remove high-frequency noise, segment time series data according to time windows, and aggregate the data according to the required granularity.

[0026] Preferably, collect students' behavioral data, including the following steps:

[0027] Authenticate students through student ID numbers and login name information, extract students' login logs from the school management system or relevant platforms, record students' login times in the database, including login dates and specific times, for the classroom online platform, monitor students' discussion participation, including the number of speeches and participating topics, record students' behavioral data in the discussion in the database, obtain students' homework submission status from the school homework management system, and record students' homework submission status in the database, including submission time and homework type.

[0028] Preferably, collect students' personal data, including the following steps:

[0029] Obtain students' subject score data from the school management system or educational institutions, record students' subject scores in the database by subject and time, if the school has an attendance system, extract students' attendance records from the attendance system, use social media monitoring tools to understand students' interactions on social platforms, record students' participation in school activities, including clubs, societies, and volunteer services, store the collected data in a secure database, and regularly check the accuracy and integrity of the data.

[0030] A student behavior early warning analysis system based on big data, including a data collection module, a preprocessing module, a data integration module, a feature extraction module, a regression analysis module, and an anomaly judgment module;

[0031] Data collection module: Collect students' personal data and collect students' behavior data;

[0032] Preprocessing module: Perform data preprocessing on personal data and behavior data;

[0033] Data integration module: Integrate data from different sources and establish a comprehensive student information database;

[0034] Feature extraction module: Select feature data closely related to students' behavior from the student information database;

[0035] Regression analysis module: Analyze the feature data through a regression model and output the student's behavior index;

[0036] Anomaly judgment module: Judge the abnormal situation of students' behavior based on the comparison result between the behavior index and the preset analysis threshold, and generate corresponding early warning measures according to the abnormal situation of behavior and send them to the administrator.

[0037] In the above technical solution, the technical effects and advantages provided by the present invention:

[0038] The present invention collects students' personal data and behavior data, performs data preprocessing on personal data and behavior data, integrates data from different sources, establishes a comprehensive student information database, selects key feature data closely related to students' behavior from the student information database, including grades, attendance rates, and homework completion situations, analyzes the key feature data through a regression model, outputs the student's behavior index, judges the abnormal situation of students' behavior based on the comparison result between the behavior index and the preset analysis threshold, and generates corresponding early warning measures according to the abnormal situation of behavior and sends them to the administrator. After comprehensively analyzing multiple data of students, this analysis method judges the behavior status of students. When there are abnormalities in the behavior of students, corresponding early warning measures are generated according to the abnormal situation. It is not only more comprehensive in analysis, but also has no manual intervention and is more accurate in analysis. Description of the Drawings

[0039] In order 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. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0040] Figure 1 It is the method flow chart of the present invention. Detailed Embodiments

[0041] 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 accompanying 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.

[0042] Example 1: Please refer to Figure 1 As shown, the method for early warning analysis of student behavior based on big data in this embodiment includes the following steps:

[0043] Collect personal data of students, including academic performance, attendance records, and social activities, collect behavioral data of students in various places such as school systems, classrooms, and libraries, such as login time, participation in discussions, and homework submission, perform data preprocessing on personal data and behavioral data, including denoising processing and data integration, process possible abnormal data or noise to ensure data quality, integrate data from different sources, establish a comprehensive student information database, select key feature data closely related to student behavior from the student information database, including grades, attendance rate, and homework completion, analyze the key feature data through a regression model, output the behavior index of the student, judge the abnormal situation of the student's behavior based on the comparison result between the behavior index and the preset analysis threshold, and generate corresponding early warning measures according to the abnormal situation and send them to the administrator.

[0044] This application collects personal data and behavioral data of students, performs data preprocessing on personal data and behavioral data, integrates data from different sources, establishes a comprehensive student information database, selects key feature data closely related to student behavior from the student information database, including grades, attendance rate, and homework completion, analyzes the key feature data through a regression model, outputs the behavior index of the student, judges the abnormal situation of the student's behavior based on the comparison result between the behavior index and the preset analysis threshold, and generates corresponding early warning measures according to the abnormal situation and sends them to the administrator. After comprehensively analyzing multiple data of students, this analysis method judges the behavior status of students. When the behavior of students is abnormal, corresponding early warning measures are generated according to the abnormal situation. It is not only more comprehensive in analysis but also more accurate without manual intervention.

[0045] Example 2: Collect personal data of students, including academic performance, attendance records, and social activities, including the following steps:

[0046] Clarify the data objective: Determine the data types to be collected: clarify the data types such as academic performance, attendance records, and social activities that need to be collected.

[0047] Set the purpose of data collection: Clearly define the purpose of using the data, such as academic assessment, student behavior early warning, social interaction analysis, etc.

[0048] Obtain authorization and consent: Seek consent from students and parents: Under the requirements of laws and ethics, ensure clear authorization and consent from students and their parents for the collection and use of personal data.

[0049] Establish a data collection system: Select data collection tools: Choose appropriate tools, which may include school management systems, online questionnaires, social media monitoring tools, etc.

[0050] Formulate data collection standards: Ensure the consistency and standardization of data for subsequent analysis.

[0051] Collect academic performance data: Obtain students' transcripts: Obtain students' academic performance data from the school management system or educational institutions.

[0052] Record academic performance: Record students' academic performance by subject and time in the database.

[0053] Collect attendance record data: Use the attendance system: If the school has an attendance system, directly extract students' attendance records from the system.

[0054] Manually record attendance: In the absence of an automated system, teachers or school staff may need to manually record students' attendance.

[0055] Collect social activity data: Monitor social media: If it is necessary to collect social activity data, social media monitoring tools can be used to understand students' interactions on social platforms.

[0056] Record school activities: Record students' participation in school activities, including clubs, societies, volunteer services, etc.

[0057] Data security and privacy protection: Strengthen data security measures: Ensure appropriate measures are taken to protect the security of students' data, including encryption, access control, etc.

[0058] Comply with privacy regulations: Comply with relevant regulations and policies to protect students' privacy rights and ensure the legal use of data.

[0059] Data storage and management: Establish a database: Store the collected data in a secure database to ensure easy management and retrieval.

[0060] Formulate a data retention policy: Formulate a clear data retention policy that stipulates the time and method of data storage.

[0061] Data quality control: Monitor data quality: Regularly check the accuracy and completeness of data to ensure data quality.

[0062] Correct erroneous data: If erroneous or inaccurate data is found, correct it in time to maintain the credibility of the data.

[0063] Monitoring and Updating: Regularly monitor the data collection process: Ensure the effectiveness and efficiency of the data collection process.

[0064] Update data as needed: As the semester progresses, update data such as subject grades, attendance records, and social activities as needed.

[0065] Collecting student behavior data in various places such as school systems, classrooms, libraries, etc., such as login time, discussion participation, and homework submission, includes the following steps:

[0066] System permissions and permissions: Obtain permission: Ensure legality and compliance, and obtain permission from the school system and other relevant systems to access student behavior data.

[0067] Data source identification and integration: Determine the data source: clarify the source of behavioral data that needs to be collected, such as school management system, classroom online platform, library system, etc.

[0068] Integrate data sources: Integrate behavioral data from different systems into a centralized data store or database.

[0069] Data field definition and standardization: Clearly define data fields: Clearly define the behavioral data fields that need to be collected, such as login time, number of discussions participated in, homework submission status, etc.

[0070] Standardize data: Ensure data consistency and standardization for subsequent analysis.

[0071] Data collection tools and technology selection: Choose appropriate tools: Select tools for collecting student behavior data, which may include system logs, API interfaces, sensors, etc.

[0072] Configure data collection technology: Configure the corresponding data collection technology according to different data sources to ensure accurate acquisition of required information.

[0073] Student identity verification: Ensure student identity: Ensure that the collected behavioral data matches the identity of the corresponding student. Identity verification can be performed through information such as student ID and login name.

[0074] Login time data collection: Extract login logs: Extract students’ login logs from the school management system or related platforms.

[0075] Record login time: Record students' login time in the database, including login date, specific time and other information.

[0076] Participation in discussion data collection: Monitoring online platforms: For classroom online platforms, etc., monitor students' discussion participation, including the number of speeches, topics of participation, etc.

[0077] Record discussion data: Record students’ behavioral data during the discussion in a database for subsequent analysis.

[0078] Homework submission status data collection: Obtain homework system data: Obtain students' homework submission status from the school homework management system.

[0079] Record homework data: Record students' homework submission status in the database, including submission time, homework type and other information.

[0080] Data security and privacy protection: Strengthen data security measures: Take measures to protect the security of student behavior data, including data encryption, access control, etc.

[0081] Comply with privacy regulations: Ensure compliance with relevant regulations and policies to protect students' privacy rights.

[0082] Data storage and management: Establish database: Store the collected behavioral data in a secure database and ensure that the data can be easily managed and retrieved.

[0083] Set a data retention policy: Clarify when and how data should be kept and follow appropriate data management policies.

[0084] Monitoring and Updating: Regularly monitor the data collection process: Ensure the effectiveness and efficiency of the data collection process.

[0085] Update data as needed: As the semester progresses, update student behavior data as needed to ensure data is current and accurate.

[0086] Data preprocessing is performed on personal data and behavioral data, including denoising, to handle possible abnormal data or noise and ensure data quality, including the following steps:

[0087] Data cleaning: Missing value processing: Detect and process missing values ​​in personal and behavioral data. You can choose to delete missing values, fill them with the mean or median, etc.

[0088] Duplicate value processing: Find and delete possible duplicate records to ensure the uniqueness of the data.

[0089] Denoising: Outlier detection: Use statistical or machine learning methods to detect outliers in data, including personal and behavioral data.

[0090] Select an appropriate denoising method: Based on the nature of the outliers, select an appropriate method for denoising, such as truncation, substitution, interpolation, etc.

[0091] Data standardization and normalization: Standardization: Standardize continuous data to ensure that it has similar scales and ranges. Common methods include Z-score standardization.

[0092] Normalization: Map the data to a specific range to ensure consistent weights between different features, such as scaling the data to the range [0, 1].

[0093] Process categorical data: One-hot encoding: For data containing categorical information, use one-hot encoding or other encoding methods to convert it into numerical data for easier processing by machine learning algorithms.

[0094] Smoothing: Moving average: Perform a moving average on time series data to smooth out possible fluctuations.

[0095] Filter application: For data in the field of signal processing, filters can be used for smoothing and removing high-frequency noise.

[0096] Data splitting and aggregation: Split time series: Split time series data according to a certain time window for better analysis and modeling.

[0097] Aggregation: Aggregate the data according to the required granularity to reduce the data volume and highlight key information.

[0098] Feature engineering: Create new features: Based on the original data, create new features to better express the information in the data, such as extracting time information, calculating statistical metrics, etc.

[0099] Dimensionality reduction: For high-dimensional data, dimensionality reduction methods such as principal component analysis (PCA) can be considered.

[0100] Data visualization and exploratory analysis: Plot charts: Explore the distribution and characteristics of the data by plotting charts such as histograms, scatter plots, box plots, etc.

[0101] Discover abnormal patterns: During the visualization process, discover potential abnormal patterns or trends.

[0102] Monitoring and iteration: Regularly monitor data: During the modeling and analysis process, regularly monitor the data quality, discover potential problems, and make corresponding adjustments.

[0103] Iterative processing: According to the actual application requirements, iteratively perform data preprocessing to optimize the data quality and model performance.

[0104] Integrate data from different sources to establish a comprehensive student information database, including the following steps:

[0105] Define the integration objectives: Determine the data types to be integrated: Clearly define the types of student information data to be integrated, including academic performance, attendance records, behavior data, etc.

[0106] Ensure data consistency: Define the integration objectives to ensure that the integrated data is reliable in terms of logic and consistency.

[0107] Data field mapping and standardization: Field mapping: For data from different sources, perform field mapping to ensure that the same or similar information is mapped to the same field.

[0108] Standardize the data: Ensure that the integrated data fields have similar formats and units for better analysis and comparison.

[0109] Establish the database architecture: Select a database system: Select an appropriate database system, such as a relational database (e.g., MySQL, PostgreSQL) or a non-relational database (e.g., MongoDB).

[0110] Design the table structure: Design the table structure of the student information database, including primary keys, foreign keys, etc., to meet the requirements of data relationships.

[0111] Data import and cleaning: Data import: Import data from different sources into the student information database to ensure data integrity and accuracy.

[0112] Clean the data: Clean the imported data to handle possible redundant, incorrect, or missing data and maintain the quality of the database.

[0113] Establish data association relationships: Define association relationships: Determine the association relationships between different tables and establish connections between data through primary keys and foreign keys.

[0114] Ensure data consistency: Through association relationships, ensure data consistency between different tables and avoid redundancy and inconsistency.

[0115] Develop a data update strategy: Regular updates: Establish a regular update mechanism to ensure that the data in the database is synchronized with the source data.

[0116] Real-time updates: For cases where real-time data is required, consider using a real-time data synchronization mechanism.

[0117] Data security and access control: Set permissions: Set different levels of data access permissions according to user roles and requirements.

[0118] Encrypt sensitive information: For sensitive information, use encryption and other means to protect it to ensure data security.

[0119] Establish indexing and query optimization: Establish indexing: Create appropriate indexes in the database tables to improve the efficiency of data retrieval.

[0120] Query optimization: Optimize common queries to ensure that the database can still provide fast responses in the case of a large amount of data.

[0121] Backup and recovery strategy: Regular backup: Establish a regular database backup strategy to prevent data loss.

[0122] Disaster recovery plan: Develop a disaster recovery plan to ensure that the database can be quickly restored in case of problems.

[0123] Monitoring and maintenance: Monitor database performance: Regularly monitor database performance, detect potential problems and solve them in a timely manner.

[0124] Maintain the database: Perform database maintenance work, including index rebuilding, statistics update, etc.

[0125] File documentation: Document the database structure: Write documentation for the student information database, detailing the database structure, table relationships, field meanings, etc.

[0126] Update the documentation: As the database structure or data source changes, update the documentation in a timely manner to maintain accuracy.

[0127] Select key feature data closely related to student behavior from the student information database, including grades, attendance, and assignment completion, including the following steps:

[0128] Clarify the problem objective: Define the analysis objective: Clearly define your objective, such as student behavior prediction, academic performance analysis, or other aspects.

[0129] Identify key behaviors: Clearly define the specific behaviors or activities that you think are closely related to student behavior.

[0130] Understand the database structure: View the table structure: Carefully understand the table structure of the student information database, including field names, data types, and relationships.

[0131] Identify key tables: Find the tables that contain information related to student behavior, such as subject grade tables, attendance record tables, assignment submission record tables, etc.

[0132] Data preview and exploration: Execute preliminary queries: Run some simple SQL queries to preview the student information data and understand the data distribution and format.

[0133] Plot charts: Plot charts such as histograms and scatter plots to explore the relationships between features.

[0134] Select key features: Select features related to the target: Select features related to the student behavior target from the database, such as academic performance, attendance rate, homework completion, etc.

[0135] Consider time factors: If there is timestamp information, consider selecting time-related features, such as semester average grades, monthly attendance rates, etc.

[0136] Calculate new features: Derive new features: Calculate new derived features based on existing features, such as semester average grades, number of absences, etc.

[0137] Combine features: Consider combining multiple related features into a more meaningful feature to improve model performance.

[0138] Handle missing values and outliers: Missing value handling: For features with missing values, select appropriate methods for handling, such as deletion, filling with mean, using interpolation, etc.

[0139] Outlier handling: Detect and handle outliers to ensure data accuracy and stability.

[0140] Standardization and normalization: Standardization: For numerical features, perform standardization to ensure they have similar scales.

[0141] Normalization: Map feature values to a certain range to avoid a certain feature having too much influence on the model.

[0142] Data splitting: Divide the training set and the test set: For model training and evaluation, divide the dataset into a training set and a test set to ensure the generalization ability of the model.

[0143] Feature importance analysis: Use the model for analysis: By training the model, utilize the feature importance analysis of the model to understand which features contribute more to student behavior prediction.

[0144] Regular update: Regularly review features: Over time, student behavior and performance may change, so it is necessary to regularly review and update the selected features.

[0145] Analyze the key feature data through a regression model and output the student's behavior index, including the following steps:

[0146] Obtain the student's absenteeism rate, late arrival and early departure rate, proportion of online course duration, and proportion of network usage time;

[0147] Substitute the absenteeism rate, the rate of being late or leaving early, the proportion of online course duration, and the proportion of network usage time into the regression model for calculation to obtain the student's behavior index. The expression is:

[0148]

[0149] , where XWZ is the behavior index, ql, cdz, tsl, and wzb are the absenteeism rate, the rate of being late or leaving early, the proportion of online course duration, and the proportion of network usage time respectively, and a 1 、a 2 、a 3 、a 4 are the regression coefficients of the absenteeism rate, the rate of being late or leaving early, the proportion of online course duration, and the proportion of network usage time respectively, and a 1 、a 2 、a 3 、a 4 are all greater than 0.

[0150] Absenteeism rate and rate of being late or leaving early:

[0151] School attendance system: Schools usually use attendance systems to record students' attendance. These systems can obtain the number of absences, lateness, and early departures of students online.

[0152] Classroom monitoring system: Some schools or classrooms may be equipped with monitoring systems to obtain information on absences, lateness, and early departures by monitoring students' activities in the classroom.

[0153] Proportion of online course duration:

[0154] Obtain the online course duration of students and the total school course duration of the day. Divide the online course duration by the total school course duration of the day to obtain the proportion of online course duration.

[0155] Proportion of network usage time:

[0156] Records of the school's online learning platform: The online learning platforms used by schools usually record students' online activities, including login time, course browsing time, etc. These data can be used to calculate the proportion of network usage time of students.

[0157] Network monitoring tools: The school network system may record students' activities on the school network, including accessing websites, using applications, etc. The network usage time of students can be estimated through these records.

[0158] Judging the abnormal behavior status of students based on the comparison result between the behavior index and the preset analysis threshold, and generating corresponding warning measures according to the abnormal behavior status and sending them to the administrator, including the following steps: After obtaining the behavior index, compare the behavior index with the preset analysis threshold, where the analysis threshold includes a first abnormal threshold and a second abnormal threshold. The first abnormal threshold is used to judge whether a student has an abnormality, and the second abnormal threshold is used to judge the severity of the student's abnormality;

[0159] If the behavior index ≤ the first abnormal threshold, it is judged that the student's behavior is normal;

[0160] If the behavior index > the first abnormal threshold, it is judged that the student's behavior is abnormal;

[0161] If the behavior index > the first abnormal threshold and the behavior index ≤ the second abnormal threshold, it is judged that the student's behavior has a minor abnormality;

[0162] If the behavior index > the second abnormal threshold, it is judged that the student's behavior has a serious abnormality;

[0163] When it is judged that the student's behavior has a minor abnormality, the following warning measures are generated:

[0164] Individual tutoring: Purpose: For minor abnormalities, the problems of students can be understood through individual tutoring or conversations.

[0165] Implementation: The school can arrange for counselors, teachers or psychological counselors to have one-on-one conversations with students to understand their problems and provide support and suggestions.

[0166] Family contact: Purpose: Contact the parents or guardians of the student to understand the family background and possible influencing factors.

[0167] Implementation: The school can communicate with the parents about the student's performance and behavior through family contact methods to obtain a more comprehensive understanding.

[0168] Formulate a personalized plan: Purpose: For the minor abnormalities of students, formulate a personalized improvement plan.

[0169] Implementation: The school can formulate personalized learning plans and behavior management plans together with the students and parents to promote the adaptation and progress of the students.

[0170] Provide support resources: Purpose: Provide additional support resources for students, such as tutoring classes, subject tutoring or mental health services.

[0171] Implementation: The school can provide additional academic and mental health support for students to help them overcome problems.

[0172] When it is judged that the student's behavior has a serious abnormality, the following warning measures are generated:

[0173] Emergency Parent Meeting: Purpose: In response to serious anomalies, immediately hold a parent meeting to discuss the student's issues with parents.

[0174] Implementation: The school can invite relevant teachers, counselors, students, and parents to participate in the meeting to jointly formulate countermeasures.

[0175] Mental Health Professional Intervention: Purpose: For students who may be involved in mental health issues, introduce professional mental health services.

[0176] Implementation: The school can arrange for professional mental health experts to conduct consultations and assessments with students and provide appropriate mental health support.

[0177] Develop an Emergency Action Plan: Purpose: In response to serious anomalies, develop an emergency action plan to clarify the problems and solutions.

[0178] Implementation: The school can cooperate with relevant departments to develop an emergency action plan for students to ensure that problems are resolved in a timely manner.

[0179] Monitoring and Tracking: Purpose: For students with serious anomalies, establish a monitoring and tracking mechanism to ensure the long-term resolution of problems.

[0180] Implementation: The school can set up a special monitoring team to regularly evaluate the progress of students and adjust intervention measures when necessary.

[0181] School Social Worker Intervention: Purpose: In response to family and social problems, introduce school social workers to provide comprehensive support.

[0182] Implementation: The school can cooperate with social workers to provide support and intervention at the family and social levels for students.

[0183] Example 3: The big data-based student behavior early warning analysis system described in this example includes a data collection module, a preprocessing module, a data integration module, a feature extraction module, a regression analysis module, and an anomaly judgment module;

[0184] Data Collection Module: Collect students' personal data, including academic performance, attendance records, and social activities, and collect students' behavior data in various places such as the school system, classrooms, and libraries, such as login time, participation in discussions, and homework submission;

[0185] Preprocessing Module: Perform data preprocessing on personal data and behavior data, including denoising processing and data integration, and process possible abnormal data or noise to ensure data quality;

[0186] Data Integration Module: Integrate data from different sources to establish a comprehensive student information database;

[0187] Feature extraction module: Select key feature data closely related to students' behaviors from the student information database, including grades, attendance rates, and assignment completion status.

[0188] Regression analysis module: Analyze the key feature data through a regression model and output the students' behavior indices.

[0189] Abnormality judgment module: Judge the students' behavior abnormality status based on the comparison result between the behavior index and the preset analysis threshold, and generate corresponding warning measures according to the behavior abnormality status and send them to the administrator.

[0190] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0191] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0192] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to the specific implementation manners. Obviously, many modifications and variations can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art in the relevant technical field can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A student behavior early warning analysis method based on big data, characterized by: The analytical method comprises the following steps: Collect students' personal data, including academic performance, attendance records, social activities, and collect students' behavioral data in various places such as the school system, classrooms, and libraries, including login time, participation in discussions, and homework submission; Perform data preprocessing on personal data and behavioral data, including denoising and data integration, integrating data from different sources to establish a comprehensive student information database; Select characteristic data closely related to student behavior from the student information database, analyze the characteristic data through a regression model, and output the student's behavior index; The abnormal behavior of students is judged by comparing the behavior index with the preset analysis threshold, and corresponding early warning measures are generated based on the abnormal behavior and sent to the administrator.

2. The student behavior early warning analysis method based on big data according to claim 1 is characterized by: The key feature data is analyzed through a regression model to output the student's behavior index, including the following steps: Obtain students’ absenteeism rate, lateness and early departure rate, proportion of online course time, and proportion of Internet usage time; Substitute the absenteeism rate, lateness and early departure rate, online course duration and network usage time into the regression model for calculation to obtain the student behavior index, which is expressed as: ， In the formula, XWZ is the behavior index, ql, cdz, tsl, wzb are the absenteeism rate, lateness and early departure rate, the proportion of online course time and the proportion of network usage time, respectively, a1, a2, a3, a4 are the regression coefficients of absenteeism rate, lateness and early departure rate, the proportion of online course time and the proportion of network usage time, respectively, and a1, a2, a3, a4 are all greater than 0.

3. The student behavior early warning analysis method based on big data according to claim 2 is characterized by: The abnormal behavior of students is judged by comparing the behavior index with the preset analysis threshold, including the following steps: After obtaining the behavior index, the behavior index is compared with a preset analysis threshold, the analysis threshold includes a first abnormal threshold and a second abnormal threshold, the first abnormal threshold is used to determine whether the student has an abnormality, and the second abnormal threshold is used to determine the severity of the student's abnormality; If the behavior index is ≤ the first abnormal threshold, it is judged that the student's behavior is not abnormal; If the behavior index is greater than the first abnormal threshold, the student's behavior is judged to be abnormal; If the behavior index > the first abnormal threshold, and the behavior index ≤ the second abnormal threshold, the student's behavior is judged to be slightly abnormal; If the behavior index is greater than the second abnormal threshold, it is judged that the student's behavior is seriously abnormal.

4. The student behavior early warning analysis method based on big data according to claim 3 is characterized by: Integrate data from different sources to build a comprehensive student information database, including the following steps: For data from different sources, field mapping is performed so that the same or similar information is mapped to the same field, the association relationship between different tables is determined, and the connection between data is established through primary keys and foreign keys. The data in the database is kept synchronized with the source data based on a regular update mechanism. Different levels of data access permissions are set according to user roles and needs, and encryption and other means are used to protect sensitive information.

5. The student behavior early warning analysis method based on big data according to claim 4 is characterized by: Data preprocessing of personal data and behavioral data includes the following steps: Detect and process missing values ​​in personal and behavioral data, find and delete existing duplicate records, use statistical or machine learning methods to detect outliers in data, including personal and behavioral data, select denoising methods based on the nature of the outliers, standardize continuous data through Z-score standardization, map data to the corresponding range, use one-hot encoding or other encoding methods to convert data containing categorical information into numerical data, perform moving average processing on time series data, use filters to smooth data in the field of signal processing, remove high-frequency noise, split time series data into time windows, and aggregate data at the required granularity.

6. The student behavior early warning analysis method based on big data according to claim 5 is characterized by: Collecting student behavior data includes the following steps: Authenticate students through their student ID and login name information, extract student login logs from the school management system or related platforms, and record student login times in the database, including login date and specific time. For online classroom platforms, monitor student discussion participation, including the number of speeches and topics of participation, and record student behavior data in discussions in the database. Obtain student homework submission status from the school homework management system, and record student homework submission status in the database, including submission time and homework type.

7. The student behavior early warning analysis method based on big data according to claim 6 is characterized by: Collecting students’ personal data involves the following steps: Obtain students' subject performance data from the school management system or educational institution, record students' subject performance in the database by subject and time, extract students' attendance records from the attendance system if the school has an attendance system, use social media monitoring tools to understand students' interactions on social platforms, and record students' participation in school activities, including societies, clubs, and volunteer services, store the collected data in a secure database, and regularly check the accuracy and completeness of the data.

8. A student behavior early warning analysis system based on big data, used to implement the analysis method according to any one of claims 1 to 8, characterized in that: It includes data acquisition module, preprocessing module, data integration module, feature extraction module, regression analysis module and abnormality judgment module; Data collection module: collect students’ personal data and behavioral data; Preprocessing module: preprocess personal data and behavioral data; Data integration module: integrate data from different sources to establish a comprehensive student information database; Feature extraction module: selects feature data closely related to student behavior from the student information database; Regression analysis module: Analyze the characteristic data through the regression model and output the student's behavior index; Abnormal judgment module: judge the abnormal behavior of students by comparing the behavior index with the preset analysis threshold, and generate corresponding early warning measures based on the abnormal behavior and send them to the administrator.

Images