Collaborative education platform based on modern information technology

Abstract

The invention discloses a collaborative education platform based on a modern information technology. The collaborative education platform comprises a data layer for collecting student-related education data; the model layer is used for constructing an appreciation evaluation model and carrying out hierarchical progress evaluation and cooperative group identification, the appreciation evaluation model calculates a teacher appreciation contribution index based on student baseline scores and progress rates, the hierarchical progress evaluation is grouped according to student bases, the progress rates in the groups are calculated respectively, and the cooperative group identification is carried out; the collaborative group recognition module is used for detecting team formation behaviors of students through an interactive teaching module, automatically creating temporary collaborative portraits and calculating semantic similarity of resources and group targets based on BERT codes; the application layer is used for carrying out teaching value calculation, collaborative resource recommendation and collaborative achievement management by adopting a micro-service architecture and an API (Application Program Interface) gateway; the interactive service layer is used for providing a teacher-side teaching value visual instrument panel and a cooperative guiding capability index monitoring interface; and the decision support layer is used for performing assessment value calculation and decision optimization based on a big data analysis platform and a machine learning model technology.

Description

Technical Field

[0001] This invention relates to the field of educational technology, and in particular to a collaborative education platform based on modern information technology. Background Technology

[0002] In the context of the accelerating global advancement of educational informatization, traditional education models face numerous challenges, including geographical limitations, uneven distribution of teachers, high barriers to sharing educational resources, limited channels for teacher-student interaction, and difficulty in accurately meeting personalized learning needs. These challenges necessitate breakthroughs through modern information technology. In response, collaborative education platforms based on modern information technology have emerged. Supported by cutting-edge technologies such as cloud computing, big data, artificial intelligence, and the Internet of Things, these platforms aim to build an open, shared, intelligent, efficient, and interactive digital education ecosystem. They integrate high-quality educational resources to achieve real-time sharing and dynamic allocation across regions, levels, and schools, and rely on intelligent algorithms to accurately analyze student learning behavior data to customize personalized learning paths. Alongside adaptive teaching programs, a multi-dimensional interactive platform is built to promote real-time communication and in-depth collaboration among teachers and students, students among students, families and schools, and educational institutions. This breaks down the time and space limitations and resource barriers of traditional education models, and promotes a paradigm shift in education from one-way instruction to two-way interaction, from standardized production to personalized cultivation, and from closed and isolated to open and collaborative. Its profound significance lies not only in improving the efficiency and quality of teaching and learning, promoting educational equity and balanced resource allocation, and stimulating learners' initiative and creativity, but also in promoting systemic changes in educational concepts, models, and systems by building a new educational ecosystem. This lays a solid foundation for cultivating innovative talents who can meet the needs of future society, and ultimately helps achieve the national education modernization strategy and optimize and improve the global education governance system.

[0003] In existing technologies, the teaching impact value relies on the frequency of above-average scores and the progress value for calculation, but it does not control for differences in students' basic abilities. The same progress value has different teaching significance for students with high and low foundations, leading to assessment distortion. Therefore, this paper proposes a collaborative education platform based on modern information technology. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a collaborative education platform based on modern information technology.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A collaborative education platform based on modern information technology includes: Data layer: Collects relevant educational data of students, and relies on the data governance engine to complete data cleaning, deduplication, tagging and construction of educational data asset catalog; Model layer: Integrating machine learning frameworks and project response theory engines, a value-added evaluation model is constructed to conduct tiered progress assessment and collaborative group identification. The value-added evaluation model calculates the teacher's value-added contribution index based on students' baseline scores and progress rates. The tiered progress assessment is conducted by grouping students according to their basic scores and calculating the progress rate within each group. The collaborative group identification detects students' team-building behavior through the interactive teaching module, automatically creates temporary collaborative profiles, and calculates the semantic similarity between resources and group goals based on BERT encoding. Application Layer: A microservice architecture and API gateway are used for teaching value calculation, collaborative resource recommendation, and collaborative outcome management. The teaching value calculation quantifies the teaching effectiveness of teachers through multiple dimensions. The collaborative resource recommendation improves resource matching efficiency through three-dimensional adaptation of individuals, groups, and roles. The collaborative outcome management collects group project results, uses multi-dimensional evaluation to quantify quality, defines a collaborative outcome contribution index and links it to the instructor, and combines collaborative guidance ability indicators and collaborative effectiveness coefficients for evaluation. Interactive Service Layer: Through the front-end React framework, a responsive interface is implemented, and real-time communication with the back-end RESTful API and WebSocket is achieved. It provides teachers with a visual dashboard of teaching values ​​and a collaborative guidance capability indicator monitoring interface, students with functions for creating collaborative groups, receiving resource recommendations, and tracking individual progress rates, and management with a visual configuration and dynamic adjustment of collaborative efficiency coefficients. Decision support layer: Based on big data analysis platform and machine learning model technology, assessment values ​​are calculated and decision optimization is performed. The assessment value calculation comprehensively evaluates the effectiveness of teachers' collaborative teaching through multi-dimensional dynamic weighting. The decision optimization provides teaching strategy suggestions based on the assessment value calculation results.

[0006] The above technical solution further includes: Furthermore, the specific steps for the value-added evaluation model to calculate the teacher's value-added contribution index based on students' baseline performance and progress rate are as follows: Data Input: Obtain student baseline scores from the data layer. The score of the i-th exam Average of exams Standard deviation ; Difficulty calibration: A Z-Score normalization layer is implemented using PyTorch to dynamically calibrate the score for each exam, represented as... : Progress rate calculation: The progress rate is calculated for each student's calibrated grades and is expressed as follows: And store it in a distributed database; VAI Calculation: Construct an HLM model using TensorFlow, input covariates, and output the teacher value-added contribution index. , represented as ,in, These are the weighting coefficients. This is random error.

[0007] Furthermore, the specific steps for calculating the progress rate within each group based on students' initial abilities in the tiered progress assessment are as follows: Data grouping: Calculate the quantile thresholds for baseline scores and group students accordingly; Intra-group progress rate calculation: The average progress rate of students within each group is obtained by averaging the progress rates of students within that group. ; Weighted summation: Calculate the weights based on the percentage of people in each group. And calculate the total weighted progress rate. .

[0008] Furthermore, the specific steps for collaborative group identification—detecting student team-building behavior through the interactive teaching module, automatically creating temporary collaborative profiles, and calculating the semantic similarity between resources and group targets based on BERT encoding—are as follows: Collaborative behavior detection: Record students' team-building behavior through the interaction service layer and generate collaborative behavior logs; Collaborative profile building: The BERT model is used to encode the target text of the group to obtain semantic vectors; Member roles are identified by analyzing member behavioral characteristics using GNNs. Resource recommendation suitability calculation: Calculate the semantic similarity Sim(G,R) between the resource and the group's objective, where G is the group's objective text and R is the resource description; The recommendation weights are adjusted based on the member's role.

[0009] Furthermore, the teaching value calculation quantifies teacher teaching effectiveness through multiple dimensions, and is expressed as follows: ,in, The above-average score index is the difficulty-calibrated version. Difficulty calibration: A Z-Score normalization layer is implemented using PyTorch to dynamically calibrate the score for each exam, represented as... : Progress rate calculation: The progress rate is calculated for each student's calibrated grades and is expressed as follows: And store it in a distributed database; VAI Calculation: Construct an HLM model using TensorFlow, input covariates, and output the teacher value-added contribution index. , represented as ,in, These are the weighting coefficients. This is random error.

[0010] Furthermore, the specific steps for improving resource matching efficiency through individual-group-role three-dimensional adaptation in the collaborative resource recommendation are as follows: Dynamic weight adjustment mechanism: The weight coefficient λ is dynamically adjusted based on the collaboration intensity. The adjustment formula for the weight coefficient λ is as follows: ,in, To adjust the coefficients, the CII (Collaboration Intensity Index) is calculated based on group interaction data, including interaction frequency, participation balance, and task completion quality. It combines interaction frequency, participation balance, and task completion quality, and forms the collaboration intensity index through standardization and dynamic weighting. Overall Recommendation Score Calculation: Combining individual behavioral scores Group correlation score The final recommendation value is calculated based on the character suitability score. , represented as When the character and resource type are a perfect match, the character fit is 1.2; when the character and resource type are partially matched, the character fit is 1.0; and when the character and resource type do not match, the character fit is 0.8.

[0011] Furthermore, the collaborative outcome management involves collecting group project outcomes, quantifying quality through multivariate evaluation, defining a collaborative outcome contribution index and linking it to the supervising teacher, and combining collaborative guidance ability indicators and collaborative effectiveness coefficients for evaluation. The specific steps are as follows: Group project deliverables collection: Collect group project deliverables and standardize them; Multi-dimensional assessment to quantify quality: A multi-dimensional assessment method is employed to quantify quality, including teacher ratings, peer review, and system plagiarism detection. Teacher ratings are calculated using a weighted formula based on the innovativeness, technical difficulty, and completion of the achievement. Peer review is calculated using an anonymous blockchain voting mechanism based on practicality and scalability, using an average score. System plagiarism detection calculates originality using a text similarity algorithm. The multi-dimensional assessment score is obtained by combining teacher ratings, peer review, and system plagiarism detection, all weighted according to their respective proportions. ; Collaboration Outcome Contribution Index (CI) Definition and Correlation with Instructors: Define the Collaboration Outcome Contribution Index (CI) and correlate it with instructors, utilizing a comprehensive multivariate evaluation score and member contribution. Calculate the group CI value, where, As a weight for the quality of results, Member contributions are assessed through collaboration behavior logs, such as the number of code commits and discussion participation. To assign role weights, the group CI values ​​are weighted and averaged according to the number of groups each instructor is responsible for, resulting in the instructor CI value, expressed as: ,in, The number of groups guided by the teacher. Let g be the number of students in group g. Let g be the CI value of the g-th group; Quantification of Collaborative Guidance Capability Indicators: The collaborative guidance capability indicators are quantified, including the frequency of discussion organization through weekly discussion statistics, the number of conflict mediations through keyword detection in chat logs, and the interaction balance through the standard deviation of member speaking frequency. The guidance capability score is then calculated using a weighted formula to construct a collaborative effectiveness coefficient and form a closed-loop evaluation chain. The collaborative effectiveness coefficient F is obtained by weighting the comprehensive outcome quality, guidance capability, and student satisfaction, and then fed back to the collaborative process through the decision support layer to optimize strategies.

[0012] Furthermore, assessment values The effectiveness of collaborative teaching among teachers is evaluated through a multi-dimensional, dynamically weighted, and comprehensive assessment, represented as follows: ,in, Interaction value, For dynamic weights.

[0013] The present invention has the following beneficial effects: In this invention, the relative progress rate is calculated by collecting students' baseline scores upon enrollment, which effectively controls the impact of differences in students' basic abilities on the assessment. The teaching significance of a 10% improvement for students with low basic abilities is quantified, avoiding the unfairness in assessment caused by the difficulty of improvement for students with high basic abilities. The teacher's value-added contribution index, the weighted sum of the stratified progress rate, and the above-average score index after difficulty calibration together constitute the optimized teaching value, forming a multi-dimensional assessment system. Attached Figure Description

[0014] Figure 1 This is a system block diagram of a collaborative education platform based on modern information technology proposed in this invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Please see Figure 1 As shown, this invention is a collaborative education platform based on modern information technology, comprising: Data Layer: Collects relevant educational data such as student admission scores, collaborative behavior, and exam dynamics. Through the combination of distributed databases (such as Hadoop HDFS) and cloud storage (Alibaba Cloud OSS), it supports the unified storage of structured data (student baseline scores, exam records) and unstructured data (collaborative behavior logs, discussion group texts). It also relies on the data governance engine to complete data cleaning, deduplication, tagging, and the construction of an educational data asset catalog. Model Layer: Integrating machine learning frameworks (TensorFlow / PyTorch) and the Item Response Theory (IRT) engine, a value-added evaluation model is constructed to perform tiered progress assessment and collaborative group identification. The value-added evaluation model calculates the teacher's value-added contribution index based on students' baseline scores and progress rates to eliminate the impact of fluctuations in exam difficulty. The tiered progress assessment divides students into high / medium / low groups according to their foundation, and calculates the progress rate within each group to avoid the assessment bias of "high-foundation students having difficulty making progress". The collaborative group identification detects students' team-building behavior through the interactive teaching module, automatically creates temporary collaborative profiles containing group goals, member roles, and resource preferences, and calculates the semantic similarity between resources and group goals based on BERT encoding. It also performs collaborative filtering by combining the historical resource usage effects of similar groups, providing a basis for role-adaptive recommendations, thereby improving group collaboration efficiency and resource matching. Application Layer: A microservice architecture (Spring Cloud) and an API gateway (Kong) are used for teaching value calculation, collaborative resource recommendation, and collaborative outcome management. The teaching value calculation quantifies teacher teaching effectiveness through multiple dimensions. The collaborative resource recommendation improves resource matching efficiency through three-dimensional adaptation of individuals, groups, and roles. The collaborative outcome management collects group project results, quantifies quality through multi-dimensional evaluation, defines a collaborative outcome contribution index and links it to the instructor, and combines collaborative guidance ability indicators and collaborative effectiveness coefficients for evaluation, forming a closed-loop evaluation chain of "collaboration process - outcome quality - teacher contribution". Interactive Service Layer: Through the front-end React framework, a responsive interface is implemented, and real-time communication with the back-end RESTful API and WebSocket is achieved. It provides teachers with a visual dashboard of teaching values ​​and a collaborative guidance capability indicator monitoring interface, students with functions for creating collaborative groups, receiving resource recommendations, and tracking individual progress rates, and management with a visual configuration and dynamic adjustment of collaborative efficiency coefficients. Decision support layer: Based on big data analysis platform and machine learning model technology, assessment values ​​are calculated and decision optimization is performed. The assessment value calculation comprehensively evaluates the effectiveness of teachers' collaborative teaching through multi-dimensional dynamic weighting. The decision optimization provides teaching strategy suggestions based on the assessment value calculation results.

[0017] In one embodiment, the specific steps of the value-added evaluation model in calculating the teacher's value-added contribution index based on students' baseline performance and progress rate are as follows: Data Input: Obtain student baseline scores from the data layer. The score of the i-th exam Average of exams Standard deviation ; Difficulty calibration: A Z-Score normalization layer is implemented using PyTorch to dynamically calibrate the score for each exam, represented as... : Progress rate calculation: The progress rate is calculated for each student's calibrated grades and is expressed as follows: And store it in a distributed database; VAI Calculation: An HLM model is built using TensorFlow, with input covariates such as progress rate and school fixed effects, and the output is the teacher value-added contribution index. , represented as ,in, These are the weighting coefficients. As for random error, if student k's progress rate Δk is higher than the model's predicted value, A positive result may indicate that the student has positive factors that have not been captured by the model (such as independent learning outside of class); conversely, a negative result may indicate temporary negative interference (such as test anxiety). School fixed effects are statistical techniques in econometrics used to control for unobservable heterogeneity at the school level. In the context of educational assessment (Value-Added Assessment Model, VAM), it captures the impact of school-specific, difficult-to-measure factors (such as campus culture, resource allocation, and student group characteristics) on teaching outcomes, thereby more accurately separating the individual teacher effect (i.e., the teacher's own contribution to teaching).

[0018] In one embodiment, the specific steps for calculating the progress rate within each group based on student baseline in the tiered progress assessment are as follows: Data grouping: Use PyTorch's torch.quantile function to calculate the quantile threshold of baseline scores and group students accordingly; Intra-group progress rate calculation: The average progress rate of students within each group is obtained by averaging the progress rates of students within that group. ; Weighted summation: Calculate the weights based on the percentage of people in each group. And calculate the total weighted progress rate. .

[0019] In one embodiment, the specific steps of the collaborative group identification, which detects student team-building behavior through an interactive teaching module, automatically creates temporary collaborative profiles, and calculates the semantic similarity between resources and group targets based on BERT encoding, are as follows: Collaborative behavior detection: Record students' team-building behaviors (such as collaborative document editing and joining discussion groups) through the interaction service layer and generate collaborative behavior logs; Collaborative profile building: The BERT model is used to encode the target text of the group to obtain semantic vectors; By analyzing member behavioral characteristics (such as the number of times they speak and the frequency of editing) using GNN, member roles (group leader / researcher / recorder) can be identified. Resource recommendation suitability calculation: Calculate the semantic similarity Sim(G,R) between the resource and the group's objective, where G is the group's objective text and R is the resource description; The recommendation weight is adjusted according to the member's role. For example, the team leader recommends "overall guidelines" and the researcher recommends "in-depth literature".

[0020] In one embodiment, the teaching value calculation quantifies teacher teaching effectiveness through multiple dimensions, and is expressed as follows: ,in, The above-average score index is the difficulty-calibrated version. Difficulty calibration: A Z-Score normalization layer is implemented using PyTorch to dynamically calibrate the score for each exam, represented as... : Progress rate calculation: The progress rate is calculated for each student's calibrated grades and is expressed as follows: And store it in a distributed database; VAI Calculation: An HLM model is built using TensorFlow, with input covariates such as progress rate and school fixed effects, and the output is the teacher value-added contribution index. , represented as ,in, These are the weighting coefficients. This is random error.

[0021] In one embodiment, the specific steps of the collaborative resource recommendation to improve resource matching efficiency through individual-group-role three-dimensional adaptation are as follows: Dynamic weight adjustment mechanism: The weight coefficient λ is dynamically adjusted based on the intensity of collaboration. For example, when the collaboration intensity is high (e.g., daily discussions exceed 2 hours), the group relevance weight λ2 is increased, and the individual behavior weight λ1 is decreased; conversely, when the collaboration intensity is low, the weight coefficient λ is adjusted using the following formula: ,in, To adjust the coefficients, the CII (Collaboration Intensity Index) is calculated based on group interaction data, including interaction frequency, participation balance, and task completion quality. It combines interaction frequency, participation balance, and task completion quality, and forms the collaboration intensity index through standardization and dynamic weighting. Overall Recommendation Score Calculation: Combining individual behavioral scores Group correlation score The final recommendation value is calculated based on the character suitability score. , represented as When the character and resource type are a perfect match, the character fit is 1.2; when the character and resource type are partially matched, the character fit is 1.0; and when the character and resource type do not match, the character fit is 0.8.

[0022] In one embodiment, the collaborative outcome management involves collecting group project outcomes, quantifying quality using multivariate assessment methods, defining a collaborative outcome contribution index and linking it to the supervising teacher, and combining collaborative guidance ability indicators and collaborative effectiveness coefficients for evaluation. The specific steps are as follows: Group project deliverables collection: Collect group project deliverables (such as reports, code repositories, experimental data, demonstration videos, and other physical deliverables, as well as process data such as discussion records and modification logs) and standardize them (including format standardization, such as converting PDF reports to structured text, content extraction, such as extracting key functions from code repositories, and metadata annotation, such as submission time and member contribution annotation). Multi-dimensional assessment to quantify quality: A multi-dimensional assessment method is adopted to quantify quality, including teacher ratings, peer review, and system plagiarism detection. Teacher ratings are calculated using a weighted formula based on the innovativeness, technical difficulty, and completion of the achievement. Peer review is calculated using an anonymous blockchain voting mechanism based on practicality and scalability, using an average score. System plagiarism detection calculates originality using text similarity algorithms such as the Jaccard index. The multi-dimensional assessment score is obtained by combining teacher ratings, peer review, and system plagiarism detection according to their respective weights. ; Collaboration Outcome Contribution Index (CI) Definition and Correlation with Instructors: Define the Collaboration Outcome Contribution Index (CI) and correlate it with instructors, utilizing a comprehensive multivariate evaluation score and member contribution. Calculate the group CI value, where, As a weight for the quality of results, Member contributions are assessed through collaboration behavior logs, such as the number of code commits and discussion participation. Assuming role weights such as group leader 1.2 / researcher 1.0 / recorder 0.8, the group CI value is obtained by weighting the group CI value according to the number of groups supervised by the supervising teacher, and is expressed as follows: ,in, The number of groups guided by the teacher. Let g be the number of students in group g. Let g be the CI value of the g-th group; Quantification of Collaborative Guidance Capability Indicators: The collaborative guidance capability indicators are quantified, including the frequency of discussion organization through weekly discussion statistics, the number of conflict mediations through keyword detection of "controversy" and "mediation" in chat records, and the interaction balance calculated through the standard deviation of the number of member speeches. The guidance capability score is then calculated using a weighted formula to construct a collaborative effectiveness coefficient and form a closed-loop evaluation chain. The collaborative effectiveness coefficient F is obtained by weighting the comprehensive outcome quality (teacher CI value), guidance capability (G value), and student satisfaction (through standardized processing of questionnaire scores). The results are then fed back to the collaborative process through the decision support layer optimization strategy (such as adjusting the weight of discussion frequency).

[0023] In one embodiment, the assessment value The effectiveness of collaborative teaching among teachers is evaluated through a multi-dimensional, dynamically weighted, and comprehensive assessment, represented as follows: ,in, Interaction value, For dynamic weighting, the entropy weighting method is used to automatically assign weights based on the dispersion of data in each dimension, expressed as follows: ,in, Let i be the information entropy of the i-th dimension. This represents the percentage of the indicator value.

[0024] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A collaborative education platform based on modern information technology, characterized in that, include: Data layer: Collects relevant educational data of students, and relies on the data governance engine to complete data cleaning, deduplication, tagging and construction of educational data asset catalog; Model layer: Integrating machine learning frameworks and project response theory engines, a value-added evaluation model is constructed to conduct tiered progress assessment and collaborative group identification. The value-added evaluation model calculates the teacher's value-added contribution index based on students' baseline scores and progress rates. The tiered progress assessment is conducted by grouping students according to their basic scores and calculating the progress rate within each group. The collaborative group identification detects students' team-building behavior through the interactive teaching module, automatically creates temporary collaborative profiles, and calculates the semantic similarity between resources and group goals based on BERT encoding. Application Layer: A microservice architecture and API gateway are used for teaching value calculation, collaborative resource recommendation, and collaborative outcome management. The teaching value calculation quantifies the teaching effectiveness of teachers through multiple dimensions. The collaborative resource recommendation improves resource matching efficiency through three-dimensional adaptation of individuals, groups, and roles. The collaborative outcome management collects group project results, uses multi-dimensional evaluation to quantify quality, defines a collaborative outcome contribution index and links it to the instructor, and combines collaborative guidance ability indicators and collaborative effectiveness coefficients for evaluation. Interactive Service Layer: Through the front-end React framework, a responsive interface is implemented, and real-time communication with the back-end RESTful API and WebSocket is achieved. It provides teachers with a visual dashboard of teaching values ​​and a collaborative guidance capability indicator monitoring interface, students with functions for creating collaborative groups, receiving resource recommendations, and tracking individual progress rates, and management with a visual configuration and dynamic adjustment of collaborative efficiency coefficients. Decision support layer: Based on big data analysis platform and machine learning model technology, assessment values ​​are calculated and decision optimization is performed. The assessment value calculation comprehensively evaluates the effectiveness of teachers' collaborative teaching through multi-dimensional dynamic weighting. The decision optimization provides teaching strategy suggestions based on the assessment value calculation results.

2. The collaborative education platform based on modern information technology according to claim 1, characterized in that, The specific steps for calculating the teacher's value-added contribution index based on students' baseline performance and progress rate in the value-added evaluation model are as follows: Data Input: Obtain student baseline scores from the data layer. The score of the i-th exam Average of exams Standard deviation ; Difficulty calibration: A Z-Score normalization layer is implemented using PyTorch to dynamically calibrate the score for each exam, represented as... : Progress rate calculation: The progress rate is calculated for each student's calibrated grades and is expressed as follows: And store it in a distributed database; VAI Calculation: Construct an HLM model using TensorFlow, input covariates, and output the teacher value-added contribution index. , represented as ,in, These are the weighting coefficients. This is random error.

3. A collaborative education platform based on modern information technology according to claim 2, characterized in that, The specific steps for calculating the progress rate within each group based on students' initial levels in the tiered progress assessment are as follows: Data grouping: Calculate the quantile thresholds for baseline scores and group students accordingly; Intra-group progress rate calculation: The average progress rate of students within each group is obtained by averaging the progress rates of students within that group. ; Weighted summation: Calculate the weights based on the percentage of people in each group. And calculate the total weighted progress rate. .

4. A collaborative education platform based on modern information technology according to claim 3, characterized in that, The specific steps for identifying collaborative groups, which involves detecting student team-building behavior through the interactive teaching module, automatically creating temporary collaborative profiles, and calculating the semantic similarity between resources and group targets based on BERT encoding, are as follows: Collaborative behavior detection: Record students' team-building behavior through the interaction service layer and generate collaborative behavior logs; Collaborative profile building: The BERT model is used to encode the target text of the group to obtain semantic vectors; Member roles are identified by analyzing member behavioral characteristics using GNNs. Resource recommendation suitability calculation: Calculate the semantic similarity Sim(G,R) between the resource and the group's objective, where G is the group's objective text and R is the resource description; The recommendation weights are adjusted based on the member's role.

5. A collaborative education platform based on modern information technology according to claim 4, characterized in that, The teaching value calculation quantifies teacher teaching effectiveness through multiple dimensions, and is expressed as follows: ,in, The above-average score index is the difficulty-calibrated version. Difficulty calibration: A Z-Score normalization layer is implemented using PyTorch to dynamically calibrate the score for each exam, represented as... ; Progress rate calculation: The progress rate is calculated for each student's calibrated grades and is expressed as follows: And store it in a distributed database; VAI Calculation: Construct an HLM model using TensorFlow, input covariates, and output the teacher value-added contribution index. , represented as ,in, These are the weighting coefficients. This is random error.

6. A collaborative education platform based on modern information technology according to claim 5, characterized in that, The specific steps for improving resource matching efficiency through individual-group-role three-dimensional adaptation in the collaborative resource recommendation are as follows: Dynamic weight adjustment mechanism: The weight coefficient λ is dynamically adjusted based on the collaboration intensity. The adjustment formula for the weight coefficient λ is as follows: ,in, To adjust the coefficients, the CII (Collaboration Intensity Index) is calculated based on group interaction data, including interaction frequency, participation balance, and task completion quality. It combines interaction frequency, participation balance, and task completion quality, and forms the collaboration intensity index through standardization and dynamic weighting. Overall Recommendation Score Calculation: Combining individual behavioral scores Group correlation score The final recommendation value is calculated based on the character suitability score. , represented as When the character and resource type are a perfect match, the character fit is 1.2; when the character and resource type are partially matched, the character fit is 1.0; and when the character and resource type do not match, the character fit is 0.

8.

7. A collaborative education platform based on modern information technology according to claim 6, characterized in that, The collaborative outcome management involves collecting group project outcomes, quantifying quality through multivariate evaluation, defining a collaborative outcome contribution index and linking it to the supervising teacher, and combining collaborative guidance ability indicators and collaborative effectiveness coefficients for evaluation. The specific steps are as follows: Group project deliverables collection: Collect group project deliverables and standardize them; Multi-dimensional assessment to quantify quality: A multi-dimensional assessment method is employed to quantify quality, including teacher ratings, peer review, and system plagiarism detection. Teacher ratings are calculated using a weighted formula based on the innovativeness, technical difficulty, and completion of the achievement. Peer review is calculated using an anonymous blockchain voting mechanism based on practicality and scalability, using an average score. System plagiarism detection calculates originality using a text similarity algorithm. The multi-dimensional assessment score is obtained by combining teacher ratings, peer review, and system plagiarism detection, all weighted according to their respective proportions. ; Collaborative Outcome Contribution Index (CI) Definition and Correlation with Instructors: Define the Collaborative Outcome Contribution Index (CI) and correlate it with instructors, utilizing a comprehensive multivariate evaluation score and member contribution. Calculate the group CI value, where, As a weight for the quality of results, Member contributions are assessed through collaboration behavior logs, such as the number of code commits and discussion participation. To assign role weights, the group CI values ​​are weighted and averaged according to the number of groups each instructor is responsible for, resulting in the instructor CI value, expressed as: ,in, The number of groups guided by the teacher. Let g be the number of students in group g. Let g be the CI value of the g-th group; Quantification of Collaborative Guidance Capability Indicators: The collaborative guidance capability indicators are quantified, including the frequency of discussion organization through weekly discussion statistics, the number of conflict mediations through keyword detection in chat logs, and the interaction balance through the standard deviation of member speaking frequency. The guidance capability score is then calculated using a weighted formula to construct a collaborative effectiveness coefficient and form a closed-loop evaluation chain. The collaborative effectiveness coefficient F is obtained by weighting the comprehensive outcome quality, guidance capability, and student satisfaction, and then fed back to the collaborative process through the decision support layer to optimize strategies.

8. A collaborative education platform based on modern information technology according to claim 7, characterized in that, Assessment Value The effectiveness of collaborative teaching among teachers is evaluated through a multi-dimensional, dynamically weighted, and comprehensive assessment, represented as follows: ,in, Interaction value, For dynamic weights.

Images